Release Notes
Overview
This section describes the eZuce uniteme 16.02 software release, including:
- Software Enhancements
- Application, Hardware, and Documentation Requirements
- Installation & Upgrade Instructions
- One attachment (PDF file) containing Fixes for internally-reported issues and for customer-reported issues
Software Enhancements
eZuce is pleased to announce the Beta Release of uniteme and reachme 16.02.
This is an exciting release specially for reachme with 53 of the fixes or enhancements targeted at the product. Overall the release contains 83 new features, improvements, and fixes.
Highlights
uniteme New Features:
Easy portal customization for uniteme customers.
uniteme Improvements:
Make New User Welcome eMail customizable
Change some system defaults related to call transfers
UniteWeb - Sort Voicemail by date
UniteWeb - Upload user avatar
Polycom SoundPoint IP 4.0.8 and Later Firmware Support
Yealink plugin updates (contribution)
reachme New Features:
Enhanced Reporting
Create custom reports and they appear in Supervisor Portal
New Jasper Server version
Short CDR, Traffic Detail, Report
Configurable Auto Logout of Reach Agent if in Release mode for Specified Time
reachme Improvements:
Outbound Calling Improvements
Outbound Calls appear in CDR
Outbound calling controlled through Agent and Agent Group Permission
- Outbound call widget integrated in session manager
Important Notes
None.
Software Release History
We're currently running on a bi-monthly release cycle.
- April release for 2014 is 14.04
- December release for 2014 is 14.10 (delayed for 14.04.3)
- April release for 2015 is 15.04
- May release for 2015 is 15.05
- June release for 2015 is 15.06
- August release for 2015 is 15.08
- October release for 2015 is 15.10
- December release for 2015 is 15.12
- February release for 2016 is 16.02
Release Level History
- 14.04 - April 30, 2014
- 14.04.1 - June 01, 2014
- 14.04.2 - July 11, 2014
- 14.04.3 - October 24, 2014
- 14.10 - February 5, 2015
- 15.04 - April 29, 2015
- 15.05 - May 27, 2015
- 15.06 - June 30, 2015
- 15.08 - August 31, 2015
- 15.10.1 - December 9, 2015
- 15.12 - January 6, 2015
- 16.02 - March 14, 2016
System Requirements
For a reasonably performing system we recommend the following configuration.
Minimum hardware requirements
- Pentium 4 or Xeon processor @ 2.0 GHz Core 64bit or higher
- Minimum 4 GB of RAM with sufficient swap space
- 80 GB disk (75 users depending on usage patterns)
Note: uniteme supports an unlimited number of voice-mail boxes,the total number of hours of recorded messages is determined by the size of the hard-disk. As a rule, for every minute of recorded messages you will need 1 MB of disk space (About 3 hours per 10 GB of disk space).
Operating System
CentOS/RHEL 6 x86_64 with latest updates is required.An i386 version is also provided as is (without any support).
Devices
Phones
- Polycom VVX Devices with firmware 5.1.3 (split) are recommended for new installations
- Polycom SoundPoint IP Devices should run firmware 4.0.9 (split)
Gateways
- AudioCodes Gateways are recommended for PSTN connectivity
SBCs
- Sangoma, AudioCodes, Acme Packet and Ingate SBC's are recommended for SIP Trunking and Remote Worker connectivity (commonly referred to as sipXbridge and MediaRelay services respectively).
- NOTE: The eZuce uniteme - "Use built-in SIP Trunk SBC" found in Gateway Details for use with Trunking or Remote Worker solutions should be used only for lab purposes. The openUC "Built-In SIP Trunk SBC" (sipXbridge) will not be supported in any production or live environment.
Documentation
Technical Reference Manuals, User Guides, Reach Reference Manuals, and other technical and user information can be found under the following link: Documentation Page
Installation and Upgrade Notes
Installation note
After uniteme 16.02 is downloaded and installed,the clusterId read tag is unique (same as locationId). Follow these steps to propagate the new read tags to the MongoDB replica set:
- In the uniteme menu, click System>Database.
- Click the Add query metadata button.
- To verify that the MongoDB replica contains the unique read tags, run in the command line:
//mongo rs.config();//
Special MongoDB note
Please be aware of this Mongodb requirementshttp://docs.mongodb.org/manual/reference/ulimit/“Note: Both the “hard” and the “soft” ulimit affect MongoDB’s performance. The “hard” ulimit refers to the maximum number of processes that a user can have active at any time. This is the ceiling: no non-root process can increase the “hard” ulimit. In contrast, the “soft” ulimit is the limit that is actually enforced for a session or process, but any process can increase it up to “hard” ulimit maximum.Every deployment may have unique requirements and settings; however, the following thresholds and settings are particularly important for mongod and mongos deployments:
ulimit –a -f (file size): unlimited -t (cpu time): unlimited -v (virtual memory): unlimited -n (open files): 64000 -m (memory size): unlimited -u (processes/threads): 32000
Always remember to restart your mongod and mongos instances after changing the ulimit settings to make sure that the settings change takes effect.If you limit virtual or resident memory size on a system running MongoDB the operating system will refuse to honor additional allocation requests. After every install/upgrade please check that "cat /proc/$pid_of_mongo/limits" have the recommended value of 655350. To make this value permanent you need to create this file /etc/security/limits.d/99-mongodb-nproc.conf and add the following lines:
mongodb soft nproc 64000
mongodb hard nproc 64000
mongodb soft nofile 64000
mongodb hard nofile 64000
Special Patch Note
If you have a patch installed to your system a new patch may be required. Please contact sa@ezuce.com if think you may have a patch applied as that may be replaced during the update.
Installing from ISO image
Download uniteme ISO
Download the ISO image corresponding to your hardware and write the image to a DVD.
- The ISO files are available here: https://download.ezuce.com/openuc/ISO/
- You will need a valid ezuce.com user ID to login and download.
- We recommend the 64 bit installation in most cases. This ISO file name ends in x86_64.iso
Install uniteme
- Boot from the DVD created with the uniteme ISO image.
- Press Enter at the boot screen below to begin the uniteme installation.
- Select Manual Configuration under Enable IPv4 support and select OK.
- Set a static IPv4 address with the corresponding networking information and click OK.
- In certain situations an warning of the use of indicated storage devices will be displayed.
- Select the language to be used during the installation.
- Select the keyboard layout to be used.
- Select the timezone to be used.
- Set a root password.
- Login to the system as root with the password you provided earlier and continue on to the Configure of uniteme.
Installing from Repository
uniteme can be installed using the below procedure
- Download and install CentOS 6.x minimal ISO
- Run the following command:
yum update && reboot
- Run the following commands to retrieve and run the eZuce uniteme installer:
curl https://download.ezuce.com/openuc-setup > /usr/bin/openuc-setup chmod +x /usr/bin/openuc-setup openuc-setup
This utility will guide you through the process of installing uniteme from the eZuce software repository.
Upgrade from previous versions
When upgrading uniteme from openUC 4.6 Update 11 or 14.4.3 to 15.06 follow these steps to ensure the SEC service is correctly running:
- 1. Upgrade from 4.6 Update 11 or 14.4.3 to 15.06.
- 2. After the upgrade is complete, perform the usual restart.
- 3. Once possible, connect via CLI and monitor processes using top. Notice that the SEC process is using a lot of CPU memory.
- 4. Perform another restart OR restart only the Sipxlogwatcher service.
Support Tips and Contact Information
Please see the Getting Support section for support tips and support contact information.
Enhancements, Fixes and Known Issues
JIRA name | RN Content | Enhancement/Fix/Known Issue | Key words | |
SIPX-397 | Polycom SPIP 4.0.8 & Later firmware | Added an enhancement for some new parameters to firmware 4.0.8 and later firmware for the SoundPoint IP phones. One in particular causes the phone to close TCP socket connections which can break remote phone operation. Add the following new parameter and ensure phone is configured with the value set as 1. tcpIpApp.keepalive.tcp.sip.persistentConnection.enable 0 or 1 (0 = default) If 0, the TCP Socket connection is closed after 1 minute. When the phone sends a new SIP message, a new connection is opened. If 1, the TCP Socket connection remains open indefinitely. Large installations may want to be able to set the value to 0 to keep TCP socket connections lower. | Enhancement | Polycom |
SIPX-400 | 3 level plugin overwrite support | Added an enhancement so that GUI plugins would work to 3 levels making customization of portal easier. Ensure a depth of 3 for plugin overwriting level 1 - sipxplugin2.beans.xml level 2 - sipxplugin.beans.xml level 3 - sipxplugin0.beans.xml level 2 overwrites level 1 level 3 overwrites level 2 level 1 beans will be loaded level 2 beans will be loaded and will overwrite any beans with same name from level 1 level 3 beans will be loaded and will overwrite any beans with same name from level 2 | Enhancement | Config |
SIPX-424 | Yealink plugin V8X | Enhanced the Yealink plugin with new phones, add firmware version 8 model files, and correct number of lines on T4 phones. | Enhancement | Yealink |
SIPX-92 | Change Polycom Phone QOS Default for Call Control so DSCP = 24 | Change Polycom defaults for QoS for call control so DSCP = 24. In SipXConfig Polycom Phones QOS settings page The default TOS settings equate to a DSCP of 44. Call Control (current) Prec Dly Thpt Rlb 5 Chk UnCk UnCk 101 1 0 0 seventh bit (Cost) 0 As DSCP = 44 Call Control (proposed) Prec Dly Thpt Rlb 3 UnCk UnCk UnCk Seventh bit (Cost) Uncheck 011 0 0 0 seventh bit (Cost) 0 As DSCP = 24 RTP equates to a DSCP of 46 and if this Jira is implemented, that the Call Control settings equate to 24. A call control setting of 24 is consistent with recommendations by eZuce and Cisco. | Enhancement | Polycom |
UC-2127 | Add "Traffic Detail" Reach Report | Added an enhancement for a new Traffic Detail report to Reach. | Enhancement | Reach |
UC-2699 | Change Date filter to Date/Time filter on Reach Reports | Added an enhancement to allow the Reach supervisor user to select a date/time range for start/stop rather than just a date start/stop for the following reports: Agent Group Productivity Agent Productivity CDR Queue Group Traffic Overview Queue Traffic Overview | Enhancement | Reach |
UC-2885 | Outbound call status message could be improved | Added an enhancement to improve the status messages provided in the Reach GUI. Steps to reproduce: An agent answers a voicemail call and decides to call back the client. If the client rejects the call then an error message is displayed: "Cannot connect to outbound call destination: USER_BUSY" If the client does not answer the call another error message is displayed: "Cannot connect to outbound call destination: RECOVERY_ON_TIMER_EXPIRE" These dashboard messages could be improved to be more "user friendly". | Enhancement | Reach |
UC-3529 | Reach Freeswitch Outbound Fixes | Umbrella issue for enhancements and fixes to the outbound functionality in Reach. Fixes to consider for freeswitch_outbound: - consistent agents states - the state of an agent should be reflected under the supervisor agent manager widget - CDRs for outbound calls - recordings for outbound calls - caller id for outbound calls - outbound voicemails, conferences and callbacks - transfers to external numbers - permission profiles for allow outbound - integration of the outbound call widget into the session manager | Enhancement | Reach |
UC-3530 | Integrate outbound call widget with the session manager in Reach dashboard | Added an enhancement to integrate the outbound call widget directly in the Session Manager widget. Currently, if a Reach agent wants to make an outbound call, he can use the outbound call widget that he can add on demand into his dashboard. We should integrate this widget with the session manager widget, so that the agent can make outbound calls from the session manager. He will be allowed to transfer or create a conference with another agent in the same way as with inbound calls. | Enhancement | Reach |
UC-3581 | Auto Logout of Reach Agent | Added an enhancement that will automatically log an agent out of the system if they have been sitting in the release mode for a specified period of time. Add a feature bit on the Agent Group admin page for Reach that will allow the admin to set whether or not they want agent's to be auto logged out and if so, what the timer will be set to. The new fields on the admin UI should be displayed as follows: Auto Logout (checkbox) Release Duration (text box) These fields should be at the bottom of the Agent Group admin page. By default, the checkbox should be unchecked, indicating that an agent should never be automatically logged off. When "Auto Logout" is unchecked, the Release Duration text box should be disabled. When "Auto Logout" is checked, "Release Duration" should be enabled and the admin should be able to place a numeric value into this field. There should be "help text" associated with the "Release Duration" field that says ... "the number of minutes that an agent can sit in the release mode before being automatically logged out by the system". If "Auto Logout" is on, Reach should set a timer when an agent goes into the release state. This timer should be set to the number of minutes configured in the "Release Duration" field. Whether the agent places themselves into the release state or the system places them into the release state due to them not answering call(s) that they were offered will not matter. In both cases the system should set the timer. If the timer expires, the agent should be automatically logged off. Reach should provide a notification on the agent's portal that indicates that they have been logged off due to inactivity (like the notification that is provided when an agent is automatically placed into the release mode). | Enhancement | Reach |
UC-3722 | Reach Use Advance Login allows clicking "use advanced" without a URI entered | Added an enhancement to the ReachME login page to have the Use Advanced Login button greyed out and not selectable until the user puts a URI in the advance login area. As of 15.08, once a reach user logs in they are sent to a screen that allows them to use advanced login options. This assumes that the user that logged in has the correct permission to allow them access to advanced login options. They are presented with 2 buttons as follows: Use Advanced Login (in grey) Continue (in blue) Currently, they can click the Use Advanced Login button even if they do not put anything in the Advanced Options section (e.g. a sip URI). If they do that, they are logged in but they will be offered calls on their normal extension (registered line). This is a bit counter intuitive. This request is to have the Use Advanced Login button greyed out and not selectable until the user puts a URI in the advance login area. Once they enter a URI there, the Use Advanced Login button should be enabled and turned to blue. In addition, by default, we should have the Advanced Options sections opened up so that the Endpoint field is displayed by default. | Enhancement | Reach |
UC-3723 | Short version of the CDR report | An enhancement that adds a new short version of the CDR report and will include the following data. For all "initial" calls into Reach (segment 1 calls), show CDR details with the following data fields: Queue Line Agent Login Agent Name Disposition Caller ANI Caller ID Offered To Agent Date/Time Accepted by Agent Date/Time Time in Queue Call End Date/Time Call Duration Input parameters for execution of the report will minimally include start date and end date. | Enhancement | Reach |
UC-3727 | Cannot control outbound call actions through permission profiles | Added an enhancement to permission profiles currently to control what can be done through the outbound call widget . I am mainly referring to being allowed to transfer calls or make conferences via the outbound widget. The existing permissions like Transfer/Conference to Agent/Number apply only to inbound calls. Currently you can only make the outbound widget available or unavailable and that's it. | Enhancement | Reach |
UC-3728 | Add Reach Permission checkbox for "Allow Outbound" | Added as part of the rewrite of the outbound functionality enhancements. Specifically UC-3530 that will integrate outbound into the session manager. With the outbound widget being removed by UC-3530, permissions work will need to be done as follows: Remove the Outbound Widget from the "accessible widgets" list. Add a "Allow Outbound" check box. If this new checkbox is selected, the user will be able to use the outbound functionality integrated in the session manager. If not checked, the outbound functionality will not be usable. Not displayed at all would be preferable. Greying out the functions would be acceptable. | Enhancement | Reach |
UC-3729 | Add columns to Queue Traffic Overview Reach report | Fixed an issue with the outbound call widget that was created when an Agent rejected an incoming ReachME call which caused the call to go to Voicemail. Steps to reproduce: 1. Agent sets his status to available Result : agent is now idle. 2. Agent adds an Outbound call widget on his dashboard Result : successfully added.Status is set to Idle. 3. Agent selects a customer from the Outbound call widget ->Select a customer drop down. Result : agent is alerted by incoming call from reach.The outbound call widget displays status - pre-call 4. Agent Rejects the incoming call Issue : the outbound call widget and the session manager widget still show "pre-call". The Outbound call widget still waits for the agent to enter a number. Workaround : You can bypass this if you refresh the dashboard page. But if you keep waiting for aprox 2 minutes after the reject - reach log will display an error : "[error] <0.21401.1> gen_fsm <0.21401.1> in state awaiting_destination terminated with reason: no such process or port in call to gen_fsm:sync_send_all_state_event(<0.21249.1>, dump_state)" Followed by : "[error] <0.21401.1> CRASH REPORT Process <0.21401.1> with 0 neighbours exited with reason: no such process or port in call to gen_fsm:sync_send_all_state_event(<0.21249.1>, dump_state) in gen_fsm:terminate/7 line 600" (sipxreach_server2.log) If you don't refresh the page and choose to go ahead and enter a number and call it, then that number is called but as soon as it answers it somehow goes to the agent voicemail.. | Enhancement | Reach |
UC-3730 | Add columns to Queue Group Traffic Overview Reach report | An enhancement that added 3 columns to the Queue Group Traffic Overview report as follows: Sent To Voicemail Transfer Out SLA The first 2 new columns will be counts of calls that were sent out of the queue based on a recipe setup. Specifically, the "Sent To Voicemail" column will be a count of calls that went to queue voicemail. The "Transfer Out" column will be a count of calls sent out of the queue to some other destination based on a transfer to sip uri recipe step. In addition, these calls that make up these 2 counts will no longer be counted as abandons. Thus, the abandon count needs to be updated to ensure that calls sent to extension or voicemail are no longer included. For the 3rd new column, add a column called SLA (service level agreement) For this SLA column, allow the user to input a value to define what their SLA is. An SLA is typically something like X % of calls answered within Y seconds. So, we will want the user to input the number of seconds (Y value). From that we will then fill this new column with data that is the % of calls answered within # of seconds defined by Y (the user input). Example: User puts in 60 seconds for the input control (representing 60 seconds as the target answer time) To fill in the SLA column we will need to get the following: NumCalls = total count of calls NumCallsUnderSLA = total count of calls answered within 60 seconds (Time in Q from CDR for that Call) Then a simple division NumCallsUnderSLA/NumCalls will give us the % to display. | Enhancement | Reach |
UC-3766 | Change new user notification email to Uniteme | Modified the new user announcement to not display the user password and change references from openUC to UniteME: Currently when a new user is created and gets notified about his new account, the email looks like : OpenUC User is created openuc@cristi.ezuce.ro Welcome to Uniteme - Unified Communication System Please find below your login information User extension: 200 User Instant Message alias: 200 User password: 12345678 User Portal URL: https://unitemeiso.cristi.ezuce.ro XMPP Server: 200@cristi.ezuce.ro SIP Server: sip:200@cristi.ezuce.ro Documented new user email template in Wiki. | Enhancement | Config |
UC-3786 | JasperReports deployment improvement/update | Added an enhancement to update JasperReports deployment process allowing us to release new reports. | Enhancement | Reach |
UC-3798 | Embed JasperServer report viewer into Reach dashboard | Added an enhancement to embed JasperServer (JRS) report viewer into Reach dashboard's REPORTS tab using an iFrame and JRS HTTP api. Main advantage is avoiding duplicate work on input controls and easier, automatic way of reports integration into Reach interface. Critical part is customization of report viewer's GUI to fit Reach look and feel. | Enhancement | Reach |
UC-3835 | Upgrade JasperServer to 6.2 | Enhancement to update Jasper Server to Version 6.2 in support of new reporting capabilities. In ReachME 15.12 and earlier Jasper Server Version 5.2 offers only old and deprecated REST v1 API. For the new jasper reports deployment mechanism we should use the v2 of the API. useful link: http://community.jaspersoft.com/documentation/tibco-jasperreports-server-upgrade-guide/v610/introduction | Enhancement | Reach |
UC-3857 | Remove locking 'findAndModify' on agent_report and rstats | Added an enhancement to address a Reach performance issue. When two 'findAndModify' operations are executed by Reach they would put a lock on the database, which can badly influence the performance. From mongodb.log: 2015-12-22T10:20:25.567-0500 [conn44] warning: ClientCursor::staticYield can't unlock b/c of recursive lock ns: top: { opid: 45800690, active: true, secs_running: 0, microsecs_running: 2824, op: "query", ns: "reach_reports.agent_report", query: { findAndModify: "agent_report", query: { info: null, _id.agent: "1006", _id.vsn: 1 }, update: { $set: { info: { first_name: "Erlinda", last_name: "Edmunds", profile: "MainGroup" } } } }, client: "10.254.4.251:42399", desc: "conn44", threadId: "0x7f91899b1700", connectionId: 44, locks: { ^: "w", ^reach_reports: "W" }, waitingForLock: false, numYields: 0, lockStats: { timeLockedMicros: {}, timeAcquiringMicros: { r: 0, w: 8 } } } We can mitigate this by using update instead of findAndModify. | Enhancement | Reach |
UC-3859 | Defaults Related to Transfers should be changed | System enhancement to change the following 3 values to the opposite of what they default to in order to avoid transfer issues. We should change the default values of these options. Parameters and current defaults: System-->Media Services (Allow Blind Transfer) default: unchecked System-->Media Services (Simplify Call After Transfer) default: checked System--Voicemail (Transfer by Bridging the call) default: false Parameters and proposed new defaults: System-->Media Services (Allow Blind Transfer) default: checked System-->Media Services (Simplify Call After Transfer) default: unchecked System--Voicemail (Transfer by Bridging the call) default: true | Enhancement | Config |
UC-3876 | Advanced login is not working for outbound calls | Fixed an issue that didn't allow advanced login to work with outbound calls. A Reach agent is allowed to specify a sip endpoint at the login step. All calls will be routed at that endpoint and not at the agent's extension (which is the default). This option was not taken into consideration for outbound calls. When making an outbound call, the phone registered with the agent's extension will now ring, regardless of the specified sip endpoint. | Enhancement | Reach |
UC-3877 | Add filter for media direction to CDR reports | CDR reports should have filter for media direction. Possible values should be "all", "inbound", "outbound". Main CDR report should have two new columns: Media Direction and Media Type. | Enhancement | Reach |
UC-3897 | CDR Reports must be able to be reported by time zone | Allow CDR Reports and Downloads to have time reported by a particular time zone. In Diagnostics -> Call Detail -> Historic and Diagnostics -> Call Detail -> Reports be able to select (and remember) a Time Zone to display call records in. Download of CSV must also be in selected time zone. | Enhancement | Config |
UC-3907 | BLF and MWI URI fields should be empty on phone profiles when not used | Added an enhancement so that BLF and MWI URI fields are empty for phone profiles that don't use BLFs or need MWI. Set "Polycom phone"->Line->Messaging->Subscribe field to blank if the User assigned to this phone has User->Permissions->Voice Mail disabled. ex: "Hoteling User has MWI, normal User on Phone has no MWI Hoteling User will login on the phone and has a waiting Message Hoteling User will logout and the MWI is still there" And if the user does not have any Speed dials with presence or group speed dials with presence defined, then on his phone set "Polycom phone"->BLF->Subscription URL field to blank. For example: "User 200 is registered on phone A with the BLF 201 and 202 The User 201 has no BLF and no permission to use BLFs If User 201 use hoteling to log in on phone A, he will get the BLF of User 200 (BLF 201, 202), because the attendant uri don't change and he tries to get the BLF of user 200." | Enhancement | Polycom |
UC-3917 | CDR data consistency for transfers and conferences with external numbers | Enhancement to display in Reach CDR a row for each external call segment. Currently, we do not track external call segments, as stated in UC-3116. Nonetheless, we have in the raw events collection information about an external transfer or conference being done. We should display in Reach CDR a row for each of these segments and mark as unknown those details that we do not track (e.g. accepted by endpoint timestamp). | Enhancement | Reach |
UW-168 | Allow upload of avatar from the user portal | Added an enhancement to allow users to be able to upload avatar from user portal. | Enhancement | Unite |
UW-319 | Don't Display Conference Bridge Section in Unite | Added an enhancement to not display Conference bridge section if user has no conference bridges defined. Currently Unite will show a conference bridge section even if there are no conferences defined for the user. In that section it will show what appears to be an error message stating: Feature Not Available Please Contact your Administrator This causes confusion for the users since and they do indeed "contact their administrator" even though there is nothing wrong but rather the feature is just not turned on for them. We should remove that section all together in the Unite display if there are no conferences for the user. | Enhancement | Unite |
UW-321 | In UniteWeb and Unite Lite voicemail is not sorted by date | Added an enhancement to sort voicemail messages by date instead of by name. | Enhancement | Unite |
SIPX-158 | Main cdr fixes | Fix an issue with CDR to properly handle forwarded calls. When a call goes out of sipx through a gateway or a SIP Trunk, we rely on the contact that other part sent back to us to build cdr, and this is often useless to get information about the called party. This is even worse when you have some call forwarding active, in this case you miss the actual entity reached from the call. Often can be useful to know about which gateway/trunk is engaged when a call exit from sipx. | Fix | CDR |
SIPX-279 | zen 4787 : user registration page error on user with multiple bria phones assigned | Fixed an issue with registration page for an user assigned to more than one Bria phone. Workaround is to search registrations in global page for that particular user using the web browser's find function. | Fix | Config |
SIPX-335 | In mongo, users point to the old location name after a location name change | Fixed an issue caused by changing a location name for a user, the old location name would remain. Steps to reproduce: 1. Create a location 2. Assign user group to this location and verify in mongodb User points to the correct location 3. Change location name to a new one and verify in mongo Issue: Users still point to the old location name. Workaround: Click Apply on the user group | Fix | Locations |
SIPX-37 | Consultative transfer to park orbit does not work | Fixed an issue with users performing a consultative transfer to park orbits. Steps : 1.Define a park orbit extension from Features->Park Orbit : extension 888 2.2012 calls 2048 3.2012 parks 2048 to the park orbit extension via a consultative transfer (not blind transfer) Expected : 2012 to be able to park the call Actual : 2012 hears a busy tone when trying to park the call Note: Consultative transfer to park orbit may disrupt BLF, recommend blind transfer only. | Fix | Park |
SIPX-398 | Remove IMAP settings from Unified Messaging settings in User Groups | IMAP is no longer supported. Remove IMAP server host, IMAP server port and TLS from User and User Group settings. | Fix | Config |
SIPX-434 | Yealink 7x/8x misleading tooltip | For Yealink Firmware Template 7x/8x Preference > Backlight Time (seconds) Currently the config "Always" means backlight always off Configuration of 0 seconds will mean always on Updated tooltip to make it more reasonable | Fix | Yealink |
SIPX-59 | Scheduled backup by specified day does not work | Fixed an issue caused when scheduling backup for one particular day and time. Steps to recreate: 1.Go to Backup and setup a scheduled backup : let's say Monday 17:00 PM (if current day is Monday) 2.Wait for specified time to arrive Reported issue : backup is not triggered. Additional info : scheduled backup setup as "Everyday" 17:PM will work. Backup Now also works. There is nothing related reported in the backup/sipxconfig.log | Fix | Backup |
SIPX-61 | Call Forwarding CDR problem | Fixed an issue where if an extension (e.g. 200) is forwarded to mobile phone, and another extension or incoming number dials 200, and talks to the mobile number, mobile phone number of the callee cannot be displayed in the CDR. | Fix | CDR |
UC-1382 | Consultative transfer must point to the exact gateway that accepted the initial INVITE | Fixed an issue caused by multiple gateways on a single dial plan, consultative transfer needed to point to the exact gateway that accepted the initial INVITE. When there are multiple gateways configured to handle the call, there is no guaranty that the initial INVITE and the transferred INVITE with replaces will follow the same forked path. The proxy should be able to tag the REPLACEs header with the absolute URI of the gateway that accepted the initial INVITE. The registrar fallback plugin should be able to parse this tag and return it as a solitary contact. How to replicate: 1. Create 2 gateways in the same branch. Let's call them gw01 and gw02 respectively. 2. Point a dial plan to both gw01 and gw02 as tandem failover 3. Bring down gw01 4. Call a user and put it on hold. 5. Using line 2, call the dial plan you have created (gw01 is down so gw02 should accept the call). 6. Bring up gw01 7. Transfer the call. Since gw01 is now online, the transferred call will go to gw01. This will fail because the call being replaced is in gw02. This is just a manual scenario. In a real deployment, all sorts of timeouts can happen making forks failover which increases the chance that an attended transfer hits a different gw. | Fix | SIPCore |
UC-1768 | Call is not recorded when agent calls back | Fixed an issue that caused agent call-backs to not record the call. Setup: Configure a client for Reach, enable call recording. Add a line and a queue for this client and an agent that can receive calls. Steps to reproduce: A client calls the Reach extension and is waiting in queue. While waiting, the client presses star and leaves a voicemail. Put the agent available. The voicemail is now routed to the agent. After listening to the message, the agent calls back the client. The client answers the call. Reported problem: In supervisor's Call Recording widget the audio file corresponding to this call is not available. Expected result: The call between the agent and the client should be recorded. 2015-06-03 14:09:05.450 [error] <0.31681.0> gen_fsm <0.31681.0> in state outbound_ringing terminated with reason: bad record client in freeswitch_outbound:outbound_ringing/2 line 201 2015-06-03 14:09:05.450 [error] <0.31681.0> CRASH REPORT Process <0.31681.0> with 0 neighbours exited with reason: bad record client in freeswitch_outbound:outbound_ringing/2 line 201 in gen_fsm:terminate/7 line 622 | Fix | Reach |
UC-1905 | Call Recording should be included in backup archive | Fixed an issue that caused Reach Recordings to not be included in backups. Configure Reach, enable call recording for a client and perform some calls. Take a backup of the system. The backup archive should include the recorded calls. When performing a restore, in supervisor's dashboard, the call recording widget will display "Audio file not available" for all the calls that were not restored. | Fix | Reach |
UC-2902 | One way talk path between supervisors on the same barge call | Fixed an issue that caused a one-way talk path between multiple Supervisors on the same barge call. Steps to reproduce: Agent is in a call with a client. Supervisor1 barges the call. The supervisor can hear both the agent and the client, both the agent and the client can hear the supervisor. Supervisor2 barges the same call. Reported problem: Supervisor1, the agent and the client can hear Supervisor2, but Supervisor2 can hear only the agent and the client. So, Supervisor1 is not heard by Supervisor2. | Fix | Reach |
UC-2910 | Truncate Reach client recordings in MongoDB and disk | Fixed an issue similar to UC-2324, where the reach.recordings collection in mongo should also be truncated based on the "Days to Retain Recordings" value in the client configuration. This mongo collection holds information about the calls which is displayed on Call Recording gadget of the Reach dashboard. After the Reach recordings change done in UC-2051 (recordings are kept now in MongoFS), the periodic cleanup should be done for: - recordings metadata - gridfs files - disk files The algorithm is: ``` for each client with call recordings enabled do: delete disk directories older than the retain date (today - (days_to_keep)) delete old recordings from gridfs (db: reach; collections: fs.files, fs.chunks) delete old recordings metadata (db: reach; collection: recordings) ``` Log what was deleted in /var/log/sipxpbx/reach-recordings-cleanup.log | Fix | Reach |
UC-3153 | Do not allow presence subscriptions to self | Fixed an issue caused when users subscribed to their own presence. Changed config webui to not allow users to subscribe to themselves. If the user has "Use Group Speed Dials" checked, this should also be filtered to exclude the user him/herself from the group list. Slightly related to UC-134. | Fix | Config |
UC-3164 | Cannot delete agent from UI, if default group for reach agents is deleted | Fixed an issue whereas the Administrator could not delete an agent from GUI if default group for reach agents is deleted. Config: openUC (14.10.20140828070812 2014-08-28EDT07:08:23 localhost.localdomain) Users already existent in openUC. Steps to reproduce: 1. Make an user a Reach agent. AdminPortal --> Features --> OpenUC Reach --> Agents --> Add New Agent --> select Agent Group : default --> select user --> OK. 2. Check Users --> User Groups. If this is the first agent you will see that a new user group is created: OpenUC-reach-agents and the user which now is an agent is added to that group, else the user is just added to this group. 3. From whatever reason the User Group: OpenUC-reach-agents is deleted. 4. Try to delete the previously added agent: OpenUC Reach --> Agents --> delete selected agent. Reported Problem: The agent cannot be deleted. An error page is presented. Comment: Issue reproduces in 14.04 too. Tested on openUC (14.04.20140430131046 2014-04-30EDT13:04:44 localhost.localdomain) | Fix | Reach |
UC-3192 | Data about Outbound Calls missing from CDR Reports | Fixed an issue with Reach CDR Report and outbound calls, no data about outbound calls could be retrieved. This information should be also accessible in CDR reports. Direction field is in this case outbound. | Fix | Reach |
UC-3204 | Reach Avatar displays wrong in Agent Manager widget | Fixed an issue caused by custom client avatars in the Agent manager. When a client avatar is defined, the agent manager displays this avatar over the top of the default handshake avatar in the agent manager. Steps to recreate: 1. Define an avatar on a client in Reach 2. Place a call into a queue such that the client with the avatar is assigned on the call 3. Answer the call with an agent 4. View the agent manager to see the answered call connected to an agent | Fix | Reach |
UC-3226 | For distribution lists, increase setting_value table from its "value" field currently limited to 1000 characters | Fixed an issue caused by creating very large distribution lists whereas the field that contained this only allowed for 1000 characters. | Fix | Config |
UC-3328 | Conference of Reach VM does NOT work when Number is selected from Transfer/Conference | Fixed an issue with conferencing a Reach Voicemail did not work properly when a number was selected from the Transfer / Conference drop-down list. Steps to reproduce: 1. Client calls to Reach line and leaves a voicemail. 2. When an agent becomes available the voicemail message is offered as a VM Callback. 3. When agent answers, the voicemail is played. 4. While listening the message the agent selects Transfer/Conference - Number. Reported problem: When Conference button is pressed nothing is triggered and the following message appears in dashboard: "Agent not available for conference" | Fix | Reach |
UC-3381 | If the client's name is longer than 18 characters the name is no longer displayed in supervisor's dashboard - Call Recordings | Fixed a display issue where if in Reach the client's name was longer than 18 characters the name is no longer displayed in supervisor's dashboard - Call Recordings. Three dots are displayed instead (...). Steps to reproduce: 1. Set in OpenUC Reach a client name longer than 18 characters. 2. Have some completed calls to Reach in order to have some recordings. 3. Check in supervisor's dashboard - Call Recordings widget that the client's name is not displayed. See picture attached. | Fix | Reach |
UC-3567 | When transferring an outbound call, "Outbound Call" is displayed instead of customer outbound caller id | Fixed a display issue when transferring an outbound call, "Outbound Call" was displayed instead of the outbound CallerID. Steps to reproduce: 1.Setup a client with Outbound CallerID - "ezuce support" 2.As an agent,login to dashboard and add the Outbound call widget 3.Use the widget to place a call to this agent (select client,answer call from Reach,dial the clients number and call) 4.Client answers call 5.Agent transfers outbound call to some other number Reported issue : "Outbound Call" is displayed at the last endpoint instead of the defined client Outbound CallerID | Fix | Reach |
UC-3572 | Only one of the agent names is displayed in the Agent Productivity report if the agent names exceed 14 characters in total | Fixed an issue where only one of the Agent names was displayed in the Agent Productivity Report if an Agent name exceeded 14 characters. Steps to reproduce: 1.Setup an user with first & last names that in total exceed 14 characters . 2.Configure Reach and add the above user as an Agent. 3.Call a queue and have this agent answer some calls so that the reach reports can be populated. 4.As a supervisor generate an Agent Productivity report and verify the Agent Name column Issue: the agent displays only one of his names - either the first name or the last name. Workaround : change the first/last name of the user to have under 15 characters (space between names is considered a character) | Fix | Reach |
UC-3591 | Problems moving data from events to call-segment_facts collection | Fixed an issue with call data collection. A customer complained about the lack of reports. During troubleshooting we discovered that this data is on events collection from reach_report but is not moved in call_segment_facts where it was needed. | Fix | Reach |
UC-3628 | REST API for call action doesn't escape + char | Fixed an issue with REST call action API such that it didn't escape the '+' character properly. | Fix | API |
UC-3631 | Phonebook Contacts added via "Additional Contacts" option, not visible for hoteling users | Fixed an issue that caused hoteling users to not be able to see phonebook contacts that were added via the additional contacts option. | Fix | Hoteling |
UC-3641 | Outbound call widget when agent rejects incoming ReachME call | Fixed an issue with the outbound call widget that was created when an Agent rejected an incoming ReachME call which caused the call to go to Voicemail. Steps to reproduce: 1.Agent sets his status to available Result : agent is now idle. 2.Agent adds an Outbound call widget on his dashboard Result : successfully added.Status is set to Idle. 3.Agents selects a customer from the Outbound call widget ->Select a customer drop down. Result : agent is alerted by incoming call from reach.The outbound call widget displays status - pre-call 4.Agent Rejects the incoming call Issue : the outbound call widget and the session manager widget still show "pre-call". The Outbound call widget still waits for the agent to enter a number. Workaround : You can bypass this if you refresh the dashboard page. But if you keep waiting for aprox 2 minutes after the reject - reach log will display an error : "[error] <0.21401.1> gen_fsm <0.21401.1> in state awaiting_destination terminated with reason: no such process or port in call to gen_fsm:sync_send_all_state_event(<0.21249.1>, dump_state)" Followed by : "[error] <0.21401.1> CRASH REPORT Process <0.21401.1> with 0 neighbours exited with reason: no such process or port in call to gen_fsm:sync_send_all_state_event(<0.21249.1>, dump_state) in gen_fsm:terminate/7 line 600" (sipxreach_server2.log) If you don't refresh the page and choose to go ahead and enter a number and call it,then that number is called but as soon as it answers it somehow goes to the agent voicemail.. | Fix | Reach |
UC-3649 | In conferences with outbound calls, agent and conference participant will always see caller id as "Reach" | Fixed an issue with outbound calls where in the Agent widget the Agents would see the outbound CallerID as "Reach" instead of the actual called number. Steps to reproduce: 1.Agent calls customer using the Outbound call widget : selects the customer from the drop-down. Result : after selecting the customer,the agent receives a pre-call ,his phone displays "From:Reach Reach". 2.He answers the incoming call and then dials the customer number in the outbound widget,to which the customer answers. Result : the agent keeps seeing the callerID "From:Reach" 3.The agent uses the outbound call widget to convert the current call into a conference by inviting another extension Result : both the agent and the other conference participant will see the callerID as : "From:Reach" Expected : at step 3 to see the Agent CallerID or the Outbound CallerID (if that one was setup) Not sure if on steps 1 & 2 the caller ID is the correct one. | Fix | Reach |
UC-3660 | Cannot save Reach report in PDF format from Chrome | Fixed an issue where a user could not download the PDF version of a report. After placing some calls, the user generated a Reach report from the supervisor dashboard. I was able to view the report in all available formats including PDF. Issue: i cannot download the PDF version from Chrome browser. I can download the other versions. I can download the PDF version from Firefox/IE. | Fix | Reach |
UC-3696 | Transfer on outbound calls via a voice mail call back is not working. | Fixed an issue where when an agent transferred a call that was initiated from a VoiceMail callback the callback was dropped. Steps: 1.Caller calls Line 303 where Agent 1 is Released. 2.Caller presses * to leave a VM in this queue and then presses * again to finish. 3.Agent is alerted by the newly left VM when returning to idle state. 4.Agent successfully listens the VM and then from the Dashboard clicks the Callback icon. 5.Caller is called back. 6.While on the line with caller, Agent 1 performs a call Transfer to Agent 2 from the Transfer/Conference Widget. 7.Agent 2 answers the call. Reported Problem : Agent 2 should have the called back party on the line. Actual Result : The call back is dropped, Agent 2 is hearing VM left by caller to Agent 1. After listening the routed VM, 2nd agent is capable of doing another callback to reach the caller from the Callback widget. | Fix | Reach |
UC-3812 | Call Park BLF and Line Appearance - Wrong CallerID | Fixed an issue with the CallerID displayed when a call was retrieved from the park orbit. The CallerID would show what was dialed to unpark the call. 1. Call from 2159663354(PSTN) -> 2676380305(alias for x201) 2. Answer call. 3. Blind Transfer from x201->5007(Call Park) 4. MoH to caller and Call Park BLF light and Line Appearance popup on x203 are as expected. 5. x202 fetches CP call via *45007 successfully. 6. BUT x202 does NOT see CallerID of the remote party (2159663354) but rather *45007. 7. ALSO, the CP BLF on x203 is STILL LIT, even though call has been fetched out of park | Fix | Park |
UC-3817 | Outbound call widget when called party rejects outbound call | Fixed an issue where if the called party rejects an outbound call initiated by a Reach agent, then the call is not cleared correctly. The first leg of the call would still remain active, the agent must close it manually from his phone. Nonetheless, the state of the agent is correctly updated. When the same agent tries to make another outbound call, then he will receive the following error: RCV: {"jsonrpc":"2.0","error":{"code":2,"message":"existing_call"},"id":722} | Fix | Reach |
UC-3820 | Reach Call shows Talk Time but No Offered To Agent Time | Fixed an issue where a customer reported that Reach CDR shows a call that had talk time (CPT) with no Offered To Agent date/time stamp. This should not be possible. | Fix | Reach |
UC-3831 | Conferences to external numbers for outbound calls | Fixed some issues with Outbound calling for Agents. Currently, an agent has the possibility to convert an outbound call into a conference and add new parties. This feature had some issues, as follows: - When the third party rejects the conference invite call or does not answer, the dashboard will still show the agent as being on conference with that party - The agent cannot add more than a party because the transfer / conference drop down box will not be shown after the first conference invite. All those issues were found after integrating outbound call widget into the session manager widget. | Fix | Reach |
UC-3843 | Idle time not correctly computed in Reach | Fixed an issue where the Idle Time was not properly computed in the Agent Portal. We have few reports from couple offices when they press idle button on acd portal we display time under the idle/release button showing how long they have been idle. Few users are reporting that even tho when they stay idle for exact 1 hour without changing status or taking call somehow the time shows little less than 1 hour under idle." The problem appears to be timer.js module, that computes the time based on a setInterval method called each second. It increments the number of seconds by one each second, instead of using date/time differences between the current time and the started time. | Fix | Reach |
UC-3845 | Reach Call Stuck in Ringing State | Fixed an issue with agents that did not have a phone assigned to them in the system. The call would show as ringing but never be offered to another agent. This may happen in other cases but it was only really noticed when an agent did not have a phone assigned to them at all in the system. Clearly this means that there would be no registration for a phone for that user as well then. The scenario is as follows: Agent logs into the Reach portal with valid credentials but has no phone. Agent goes idle Call is offered to the agent but the agent cannot answer the call. What should happen is that the agent is set to the release mode and the call is pulled back and offered to another agent. What actually happened is that the call continued to show as being ringing both in the agent reach portal and in the queue manager widget and the call was never offered to another agent. | Fix | Reach |
UC-3850 | Call Recording Widget Icons Are Not Centered | Fixed a graphics issue in the Agent portal that caused icons to not be displayed in the middle of buttons. | Fix | Reach |
UC-3852 | Advanced Login Does Not Allow regular phone numbers | Fixed an issue caused by a data validation check that that ensured that an agent using the advanced login option enters a sip uri in the end point field caused regular phone numbers to not be accepted. This results in a user wanting to use a PSTN number (e.g. 2318232979) must enter something like 2318232979 followed by @domain. So, this workaround was required: 2318232979@domain The used to be able to simply put in the PSTN number like this: 2318232979 This is much better from a usability perspective since users dial external numbers with just a simple 10 digit or perhaps 1+10 or 91+10. We want to remove this check that was added and perhaps replace it with something that checks for only crazy special characters that would not be allowable. | Fix | Reach |
UC-3854 | License count display problem when superadmin is phantom user. | Fixed an issue when superadmin is marked as a phantom user, the "Number of users requiring licenses" field displayed in License page was less by 1. This was because the license count had subtracted the superadmin account from the "Number of users requiring licenses" twice, once because it is superadmin and a second time because it is a phantom user. This problem does not impact the actual validation and license warning mechanism, it is only a problem in the License page display fields. | Fix | Config |
UC-3860 | Reach Callback when the agent is in a conference is not working | Fixed an issue for calling back a caller when the agent was already in a conference (this would be used if the caller got disconnected). Currently, when an agent receives a voicemail, he is allowed to conference in other parties. If he is in a conference and then tries to make a callback to the original caller, the conference would be ended and the caller would just be in a normal call with the agent. Expected behaviour: Add the caller to the conference. | Fix | Reach |
UC-3863 | Audio not cut through for Reach call transferred and then answered during Media Annc | Fixed an issue where calls transferred from one queue to another and then answered during a media announcement in the second queue were not properly connected. To recreate: call enters Sales queue call is offered to agent 2000 and is answered agent 2000 transfers the call to queue General agent 2000 is offered the call again out of that queue but did not answer (this step is not needed to recreate) caller begins hearing media announcement from recipe step on General queue at 20 seconds in the queue agent 2001 becomes available during the media announcement playback to the caller agent 2001 answers the call Note that general queue has a recipe step to play a hold media file every 20 seconds. result: Caller is left hearing in queue like treatment (MOH and media announcements) Agent is left with dead air Call seems to be connected and both legs show up in cdr and in fs_cli | Fix | Reach |
UC-3864 | /var/log/messages " There was a mount error, trying to mount one of the filesystems on this host." | Fixed an issue with the new SIP packet capture tcpdump based service when running on multiple servers. Log entry noted on fresh install Uniteme 15.12 /var/log/messages seems to be related to new tcpdump service .. : [root@uc1 ~]# grep -c 'mount' /var/log/messages 169 [root@uc1 ~]# grep 'mount' /var/log/messages | tail -n 5 Jan 11 11:42:59 uc1 cf3[1656]: There was a mount error, trying to mount one of the filesystems on this host. Jan 11 11:42:59 uc1 cf3[1656]: There was a mount error, trying to mount one of the filesystems on this host. Jan 11 11:42:59 uc1 cf3[1656]: There was a mount error, trying to mount one of the filesystems on this host. Jan 11 11:42:59 uc1 cf3[1656]: There was a mount error, trying to mount one of the filesystems on this host. Jan 11 11:43:00 uc1 cf3[1656]: There was a mount error, trying to mount one of the filesystems on this host. | Fix | Config |
UC-3871 | VM stream can only be played once, if voicemail slider reaches the end | Fixed an issue with the voicemail slider where if it reached the end of the message the stream could not be listened to again. Reported problem: Agent receives a voicemail call, answers and hears the message. The voicemail stream can only be heard until the voicemail slider reaches the end. After that point, even if the slider is moved at the start position, pressing the play button will not resume the stream. | Fix | Reach |
UC-3872 | Issues with websocket connection keep-alive in Reach | Fixed an issue with the ReachME Dashboard and websocket connections being closed randomly. Reach dashboard opens a websocket connection on 8937 / 8936 ports when agents are logged in. Over that connection, the agents are being served realtime statistics, call queues updates and more. There is a keep-alive mechanism implemented to ensure the websocket connection is always open. Nonetheless, this mechanism is not correctly working in the latest Reach version. If a Reach agent remains logged in for a longer period of time, the websocket connection is closed (and not reopened), which makes the dashboard freezing. That agent will not receive calls any more, even if he will see himself as "available". In the browser console, the error displayed is: "Cannot read property 'status' of undefined". This issue could be caused by the embedding of Jasper report viewer in Reach dashboard. | Fix | Reach |
UC-3873 | Missing collections from Reach recordings backup | Fixed an issue with backup of Reach data, the mongo collections fs.files and fs.chunks were not added to the archive. | Fix | Reach |
UC-3874 | Trim client name in agent manager widget | Fixed an issue that caused a display problem in Agent Manager widget if the client name was too long. | Fix | Reach |
UC-3875 | Reach is not trying to reconnect to jasperserver in case the service is down and then returned to service | Fixed an issue such that if the Jasper Server failed or had not started yet, the Reach dashboard would not try to reconnect. Possible scenarios for this issue: - reach process started, but for some reason jasperserver has not started yet; if an agent would login in reach dashboard, he won't see reports, but no indication of an error would be displayed - if jasper server stops while agents are connected in the dashboard, then they won't see reports, but again, no indication of an error would be displayed. We should display the error as a notification and while jasperserver is starting, a loader instead of an empty reports drop down list. | Fix | Reach |
UC-3882 | SSS should consider REFER messages | Fixed an issue with SSS and the way it handles REFER messages. For example, when call retrieved from park orbit by pressing speed dial button with presence, phone is instructed to call *78(park), in sample below *781100. So the INVITE that goes out from phone has To header *781100. When phone tries to park call second time to 1100 it will try to park the call with the To header *781100. This is done in 2 steps: INVITE to *781100 then REFER to 1100 SSS should treat properly this case (the REFER message), by looking at Refer-To header and marking that extension as busy | Fix | SSS |
UC-3883 | Update CDR jrxml to account for outbound calls CDRs | Reach CDRs for outbound calls were implemented in UC-3192. This created the need to make some changes to the underlying CDR jrxml file, in order to properly display CDRs for outbound calls: - rename several columns: Caller ANI - Endpoint ANI Caller ID - Originating Caller ID Offered to Agent - Offered to Endpoint Accepted by Agent - Accepted by Endpoint Time in Queue - Time in Queue/Precall - for columns like 'Queue', 'Line' etc. specify that those fields are not applicable by NA (instead of empty entry) - update formulas timestamps: Time In Queue/Precall = Accepted by Endpoint - Date/Time OR Time in Queue/Precall = CallEnd - Date/Time Call Duration = Time in Precall OR Call Duration = Call End - Date/Time CPT = Call End - Accepted by Endpoint OR CPT = 0 | Fix | Reach |
UC-3892 | A monitored call park extension will show Busy status when it is actually not | Fixed an issue where a park orbit may appear to be busy but there are no calls in the orbit. Steps to reproduce: Park call, un-park, park it again but this time don't un-park the call - make the user waiting in the park extension to drop the call. Issue: the one who monitors the call park extension via presence speed dial , will see the status of the call park extension as still being busy,so it basically remains stuck - though the call park extension is now empty since the one waiting in it,dropped the call. You have to restart sss or the monitoring phone, or do another park-unpark to clear the stuck presence | Fix | SSS |
UC-3896 | Webui rejects longer TLD names as invalid | Fixed an issue with the tapestry email validator where it doesn't allow for top level domain names with more than 4 characters. Fixed by changing validator regular expression. | Fix | Config |
UC-3905 | Polycom Hoteling Server Type | Fixed an issue with the hoteling server type setting. Phone Configuration > Hoteling Server Type will be configured as prov.serverType Parameter must be according to Polycom Admin Guide device.prov.serverType Tested with 16.02, Polycom VVX 410, FW 5.3.1 | Fix | Hoteling |
UC-3910 | Queue page not showing inherited skills correctly | Fixed an issue with the Admin GUI where the Queue page was not showing inherited skills correctly. | Fix | Reach |
UC-3914 | Conference leg with external number fails if the third party is invited by an agent different then the first agent | Fixed an issue that caused a conference leg with external number fails if the third party is invited by an agent different then the first agent. Steps to reproduce: 1. A call is placed into a Reach line 2. The call is routed to an agent, the agent takes the call 3. The agent decides to make a conference with another agent 4. The other agent is invited to the conference and answers the conf call 5. The other agent decides to invite a third party (external number) to the conference. Issue: The external leg fails with function_clause. From Reach log: 11:10:31.445 [error] <0.1229.0>Undefined - Undefined:Undefined:Undefined - gen_fsm <0.1229.0> in state oncall terminated with reason: {{function_clause,[{agent,get_endpoint,[freeswitch_conference,"1000"],[{file,"/home/nlazar/devel/sipxecs/reach-app/apps/reach_core/src/agent.erl"},{line,287}]},{freeswitch_conference,handle_conference_to_outband,7,[{file,"/home/nlazar/devel/sipxecs/reach-app/apps/reach_freeswitch/src/freeswitch_conference.erl"},{line,191}]},{gen_media,oncall,3,[{file,"/home/nlazar/devel/sipxecs/reach-app/apps/reach_core/src/gen_media.erl"},{line,1925}]},{gen_fsm,handle_msg,7,[{file,"gen_fsm.erl"},{line,505}]},...]},...} This issue is not reproducible if the first agent (the one that creates the interaction) invites the third party to the conference. | Fix | Reach |
UC-3938 | Searching for external numbers in the call history often fails | Fixed an issue when searching for external incoming calls in the CDR doesn't return reliable results. The calls are listed in the CDRs, but it is sometimes not possible to search for specific numbers - the search either returns the correct results, or no results at all. The search results are consistent; when one number is entered into the search, it always appears while another number is never found even though it is also listed in the call history. Because the CDRs are much larger for some customers, they rely more heavily on the search function. We've been able to recreate this issue on many software versions; 14.10, 15.06, 15.08, 15.10. | Fix | CDR |
UW-206 | Settings: change pin or password - no validation message | Fixed an issue where if a password is input with less than 8 chars the input changes border color, but there is no validation message. If a user is LDAP managed they should not be able to change their password and there should be a message as such. When changing PIN there is no validation message either. | Fix | Unite |
UC-3682 | Reach warm transfer not working correctly | There is a known issue when answering a Reach call from a Bria Softphone (Stretto for CCS and Bria 3 3.5.5) when selecting transfer/conference the hold button on the portal placed the Bria on hold, this is an issue because it's not obvious to the agent. The call does not release from hold unless manually done from the Bria. If only happens on each initial call to the agent, for example if the call is transferred back the agent, the agent can perform a transfer/conference without thee Bria going on hold. Tested on 15.05 and 14.04, 14.04 has no issue." Steps to reproduce .. : We did further testing. Behavior for Polycom is the same as Bria. Please note Bria was on my PC over VPN and Polycom was on my home network registered through the Ingate SBC. On Reach3 15.05 (see packet capture from Bria "reach3_Bria_hold.pcapng" 1. Agent "AGENT1" answer call 2. Agent places caller on hold - Bria goes on hold 3. Agent "AGENT1" conferences agent "lca" - Agent "AGENT2" answers ***-Bria for agent "AGENT1" does not release from hold*** - Agents cannot converse, (If agent "AGENT1" clicks hold button in UI, customer and agent "AGENT2" can converse). 4. Agent "AGENT1" manually takes Bria off hold 5. All parties converse. On Reach1 14.04 (see packet capture from Bria "reach1_Bria_hold.pcapng" 1. Agent "AGENT1" answer call 2. Agent places caller on hold - Bria goes on hold 3. Agent "AGENT1" conferences agent "AGENT2" - Agent "AGENT2" answers -Bria for agent "AGENT1" releases from hold - Agents can converse, 4. Agent "AGENT1" click hold in UI taking customer off hold 5. All parties converse. Additional information : " I took a closer look at each Notify hold and talk request and the "To:" Address in 14.04 is the sip uri of the agent in both hold and talk. In 15.05 the hold has the correct sip uri while talk changes the "To:" Address to the callers sip uri Multiple PCAPs are attached within the Zendesk ticket, both Bria and Polycom. Workaround is to downgrade Freeswitch to 1.4.6. Please contact TAC for assistance in doing this | Known Issue | Reach |
SIPX-430 | Unable to remove Homer/sipxhomer | On V15.12, with the introduction of Network Packet Capture, Homer cannot be removed due to: dependency on package sipxecs-- root@s01 ~]# yum remove homer sipxhomer sipxhomer-config sipxhomer-proxyplugin Loaded plugins: product-id, rhnplugin, security, subscription-manager This system is receiving updates from RHN Classic or RHN Satellite. Setting up Remove Process Resolving Dependencies --> Running transaction check ---> Package homer.x86_64 0:3.2.5-2.el6 will be erased ---> Package sipxhomer.x86_64 0:15.12-4124.49451 will be erased --> Processing Dependency: sipxhomer >= 15.12 for package: sipxecs-15.12-7798.ee34c.x86_64 ---> Package sipxhomer-config.x86_64 0:15.12-4124.49451 will be erased ---> Package sipxhomer-proxyplugin.x86_64 0:15.12-4124.49451 will be erased --> Running transaction check ---> Package sipxecs.x86_64 0:15.12-7798.ee34c will be erased --> Finished Dependency Resolution Dependencies Resolved ======================================================================================================= Package Arch Version Repository Size ======================================================================================================= Removing: homer x86_64 3.2.5-2.el6 @openuc 1.6 M sipxhomer x86_64 15.12-4124.49451 @openuc 449 k sipxhomer-config x86_64 15.12-4124.49451 @openuc 18 k sipxhomer-proxyplugin x86_64 15.12-4124.49451 @openuc 389 k Removing for dependencies: sipxecs x86_64 15.12-7798.ee34c @openuc 0.0 Transaction Summary ======================================================================================================= Remove 5 Package(s) Installed size: 2.4 M If you proceed with the removal of *homer, then attempt to reinstall sipxecs, *homer is pulled back in. | Config |
Enhancements, Fixes and Known Issues - By Keyword
JIRA name | RN Content | Enhancement/Fix/Known Issue | Key words | |
UC-3628 | REST API for call action doesn't escape + char | Fixed an issue with REST call action API such that it didn't escape the '+' character properly. | Fix | API |
SIPX-59 | Scheduled backup by specified day does not work | Fixed an issue caused when scheduling backup for one particular day and time. Steps to recreate: 1.Go to Backup and setup a scheduled backup : let's say Monday 17:00 PM (if current day is Monday) 2.Wait for specified time to arrive Reported issue : backup is not triggered. Additional info : scheduled backup setup as "Everyday" 17:PM will work. Backup Now also works. There is nothing related reported in the backup/sipxconfig.log | Fix | Backup |
SIPX-158 | Main cdr fixes | Fix an issue with CDR to properly handle forwarded calls. When a call goes out of sipx through a gateway or a SIP Trunk, we rely on the contact that other part sent back to us to build cdr, and this is often useless to get information about the called party. This is even worse when you have some call forwarding active, in this case you miss the actual entity reached from the call. Often can be useful to know about which gateway/trunk is engaged when a call exit from sipx. | Fix | CDR |
SIPX-61 | Call Forwarding CDR problem | Fixed an issue where if an extension (e.g. 200) is forwarded to mobile phone, and another extension or incoming number dials 200, and talks to the mobile number, mobile phone number of the callee cannot be displayed in the CDR. | Fix | CDR |
UC-3938 | Searching for external numbers in the call history often fails | Fixed an issue when searching for external incoming calls in the CDR doesn't return reliable results. The calls are listed in the CDRs, but it is sometimes not possible to search for specific numbers - the search either returns the correct results, or no results at all. The search results are consistent; when one number is entered into the search, it always appears while another number is never found even though it is also listed in the call history. Because the CDRs are much larger for some customers, they rely more heavily on the search function. We've been able to recreate this issue on many software versions; 14.10, 15.06, 15.08, 15.10. | Fix | CDR |
SIPX-400 | 3 level plugin overwrite support | Added an enhancement so that GUI plugins would work to 3 levels making customization of portal easier. Ensure a depth of 3 for plugin overwriting level 1 - sipxplugin2.beans.xml level 2 - sipxplugin.beans.xml level 3 - sipxplugin0.beans.xml level 2 overwrites level 1 level 3 overwrites level 2 level 1 beans will be loaded level 2 beans will be loaded and will overwrite any beans with same name from level 1 level 3 beans will be loaded and will overwrite any beans with same name from level 2 | Enhancement | Config |
UC-3766 | Change new user notification email to Uniteme | Modified the new user announcement to not display the user password and change references from openUC to UniteME: Currently when a new user is created and gets notified about his new account, the email looks like : OpenUC User is created openuc@cristi.ezuce.ro Welcome to Uniteme - Unified Communication System Please find below your login information User extension: 200 User Instant Message alias: 200 User password: 12345678 User Portal URL: https://unitemeiso.cristi.ezuce.ro XMPP Server: 200@cristi.ezuce.ro SIP Server: sip:200@cristi.ezuce.ro Documented new user email template in Wiki. | Enhancement | Config |
UC-3859 | Defaults Related to Transfers should be changed | System enhancement to change the following 3 values to the opposite of what they default to in order to avoid transfer issues. We should change the default values of these options. Parameters and current defaults: System-->Media Services (Allow Blind Transfer) default: unchecked System-->Media Services (Simplify Call After Transfer) default: checked System--Voicemail (Transfer by Bridging the call) default: false Parameters and proposed new defaults: System-->Media Services (Allow Blind Transfer) default: checked System-->Media Services (Simplify Call After Transfer) default: unchecked System--Voicemail (Transfer by Bridging the call) default: true | Enhancement | Config |
UC-3897 | CDR Reports must be able to be reported by time zone | Allow CDR Reports and Downloads to have time reported by a particular time zone. In Diagnostics -> Call Detail -> Historic and Diagnostics -> Call Detail -> Reports be able to select (and remember) a Time Zone to display call records in. Download of CSV must also be in selected time zone. | Enhancement | Config |
SIPX-279 | zen 4787 : user registration page error on user with multiple bria phones assigned | Fixed an issue with registration page for an user assigned to more than one Bria phone. Workaround is to search registrations in global page for that particular user using the web browser's find function. | Fix | Config |
SIPX-398 | Remove IMAP settings from Unified Messaging settings in User Groups | IMAP is no longer supported. Remove IMAP server host, IMAP server port and TLS from User and User Group settings. | Fix | Config |
UC-3153 | Do not allow presence subscriptions to self | Fixed an issue caused when users subscribed to their own presence. Changed config webui to not allow users to subscribe to themselves. If the user has "Use Group Speed Dials" checked, this should also be filtered to exclude the user him/herself from the group list. Slightly related to UC-134. | Fix | Config |
UC-3226 | For distribution lists, increase setting_value table from its "value" field currently limited to 1000 characters | Fixed an issue caused by creating very large distribution lists whereas the field that contained this only allowed for 1000 characters. | Fix | Config |
UC-3854 | License count display problem when superadmin is phantom user. | Fixed an issue when superadmin is marked as a phantom user, the "Number of users requiring licenses" field displayed in License page was less by 1. This was because the license count had subtracted the superadmin account from the "Number of users requiring licenses" twice, once because it is superadmin and a second time because it is a phantom user. This problem does not impact the actual validation and license warning mechanism, it is only a problem in the License page display fields. | Fix | Config |
UC-3864 | /var/log/messages " There was a mount error, trying to mount one of the filesystems on this host." | Fixed an issue with the new SIP packet capture tcpdump based service when running on multiple servers. Log entry noted on fresh install Uniteme 15.12 /var/log/messages seems to be related to new tcpdump service .. : [root@uc1 ~]# grep -c 'mount' /var/log/messages 169 [root@uc1 ~]# grep 'mount' /var/log/messages | tail -n 5 Jan 11 11:42:59 uc1 cf3[1656]: There was a mount error, trying to mount one of the filesystems on this host. Jan 11 11:42:59 uc1 cf3[1656]: There was a mount error, trying to mount one of the filesystems on this host. Jan 11 11:42:59 uc1 cf3[1656]: There was a mount error, trying to mount one of the filesystems on this host. Jan 11 11:42:59 uc1 cf3[1656]: There was a mount error, trying to mount one of the filesystems on this host. Jan 11 11:43:00 uc1 cf3[1656]: There was a mount error, trying to mount one of the filesystems on this host. | Fix | Config |
UC-3896 | Webui rejects longer TLD names as invalid | Fixed an issue with the tapestry email validator where it doesn't allow for top level domain names with more than 4 characters. Fixed by changing validator regular expression. | Fix | Config |
SIPX-430 | Unable to remove Homer/sipxhomer | On V15.12, with the introduction of Network Packet Capture, Homer cannot be removed due to: dependency on package sipxecs-- root@s01 ~]# yum remove homer sipxhomer sipxhomer-config sipxhomer-proxyplugin Loaded plugins: product-id, rhnplugin, security, subscription-manager This system is receiving updates from RHN Classic or RHN Satellite. Setting up Remove Process Resolving Dependencies --> Running transaction check ---> Package homer.x86_64 0:3.2.5-2.el6 will be erased ---> Package sipxhomer.x86_64 0:15.12-4124.49451 will be erased --> Processing Dependency: sipxhomer >= 15.12 for package: sipxecs-15.12-7798.ee34c.x86_64 ---> Package sipxhomer-config.x86_64 0:15.12-4124.49451 will be erased ---> Package sipxhomer-proxyplugin.x86_64 0:15.12-4124.49451 will be erased --> Running transaction check ---> Package sipxecs.x86_64 0:15.12-7798.ee34c will be erased --> Finished Dependency Resolution Dependencies Resolved ======================================================================================================= Package Arch Version Repository Size ======================================================================================================= Removing: homer x86_64 3.2.5-2.el6 @openuc 1.6 M sipxhomer x86_64 15.12-4124.49451 @openuc 449 k sipxhomer-config x86_64 15.12-4124.49451 @openuc 18 k sipxhomer-proxyplugin x86_64 15.12-4124.49451 @openuc 389 k Removing for dependencies: sipxecs x86_64 15.12-7798.ee34c @openuc 0.0 Transaction Summary ======================================================================================================= Remove 5 Package(s) Installed size: 2.4 M If you proceed with the removal of *homer, then attempt to reinstall sipxecs, *homer is pulled back in. | Config | |
UC-3631 | Phonebook Contacts added via "Additional Contacts" option, not visible for hoteling users | Fixed an issue that caused hoteling users to not be able to see phonebook contacts that were added via the additional contacts option. | Fix | Hoteling |
UC-3905 | Polycom Hoteling Server Type | Fixed an issue with the hoteling server type setting. Phone Configuration > Hoteling Server Type will be configured as prov.serverType Parameter must be according to Polycom Admin Guide device.prov.serverType Tested with 16.02, Polycom VVX 410, FW 5.3.1 | Fix | Hoteling |
SIPX-335 | In mongo, users point to the old location name after a location name change | Fixed an issue caused by changing a location name for a user, the old location name would remain. Steps to reproduce: 1. Create a location 2. Assign user group to this location and verify in mongodb User points to the correct location 3. Change location name to a new one and verify in mongo Issue: Users still point to the old location name. Workaround: Click Apply on the user group | Fix | Locations |
SIPX-37 | Consultative transfer to park orbit does not work | Fixed an issue with users performing a consultative transfer to park orbits. Steps : 1.Define a park orbit extension from Features->Park Orbit : extension 888 2.2012 calls 2048 3.2012 parks 2048 to the park orbit extension via a consultative transfer (not blind transfer) Expected : 2012 to be able to park the call Actual : 2012 hears a busy tone when trying to park the call Note: Consultative transfer to park orbit may disrupt BLF, recommend blind transfer only. | Fix | Park |
UC-3812 | Call Park BLF and Line Appearance - Wrong CallerID | Fixed an issue with the CallerID displayed when a call was retrieved from the park orbit. The CallerID would show what was dialed to unpark the call. 1. Call from 2159663354(PSTN) -> 2676380305(alias for x201) 2. Answer call. 3. Blind Transfer from x201->5007(Call Park) 4. MoH to caller and Call Park BLF light and Line Appearance popup on x203 are as expected. 5. x202 fetches CP call via *45007 successfully. 6. BUT x202 does NOT see CallerID of the remote party (2159663354) but rather *45007. 7. ALSO, the CP BLF on x203 is STILL LIT, even though call has been fetched out of park | Fix | Park |
SIPX-397 | Polycom SPIP 4.0.8 & Later firmware | Added an enhancement for some new parameters to firmware 4.0.8 and later firmware for the SoundPoint IP phones. One in particular causes the phone to close TCP socket connections which can break remote phone operation. Add the following new parameter and ensure phone is configured with the value set as 1. tcpIpApp.keepalive.tcp.sip.persistentConnection.enable 0 or 1 (0 = default) If 0, the TCP Socket connection is closed after 1 minute. When the phone sends a new SIP message, a new connection is opened. If 1, the TCP Socket connection remains open indefinitely. Large installations may want to be able to set the value to 0 to keep TCP socket connections lower. | Enhancement | Polycom |
SIPX-92 | Change Polycom Phone QOS Default for Call Control so DSCP = 24 | Change Polycom defaults for QoS for call control so DSCP = 24. In SipXConfig Polycom Phones QOS settings page The default TOS settings equate to a DSCP of 44. Call Control (current) Prec Dly Thpt Rlb 5 Chk UnCk UnCk 101 1 0 0 seventh bit (Cost) 0 As DSCP = 44 Call Control (proposed) Prec Dly Thpt Rlb 3 UnCk UnCk UnCk Seventh bit (Cost) Uncheck 011 0 0 0 seventh bit (Cost) 0 As DSCP = 24 RTP equates to a DSCP of 46 and if this Jira is implemented, that the Call Control settings equate to 24. A call control setting of 24 is consistent with recommendations by eZuce and Cisco. | Enhancement | Polycom |
UC-3907 | BLF and MWI URI fields should be empty on phone profiles when not used | Added an enhancement so that BLF and MWI URI fields are empty for phone profiles that don't use BLFs or need MWI. Set "Polycom phone"->Line->Messaging->Subscribe field to blank if the User assigned to this phone has User->Permissions->Voice Mail disabled. ex: "Hoteling User has MWI, normal User on Phone has no MWI Hoteling User will login on the phone and has a waiting Message Hoteling User will logout and the MWI is still there" And if the user does not have any Speed dials with presence or group speed dials with presence defined, then on his phone set "Polycom phone"->BLF->Subscription URL field to blank. For example: "User 200 is registered on phone A with the BLF 201 and 202 The User 201 has no BLF and no permission to use BLFs If User 201 use hoteling to log in on phone A, he will get the BLF of User 200 (BLF 201, 202), because the attendant uri don't change and he tries to get the BLF of user 200." | Enhancement | Polycom |
UC-2127 | Add "Traffic Detail" Reach Report | Added an enhancement for a new Traffic Detail report to Reach. | Enhancement | Reach |
UC-2699 | Change Date filter to Date/Time filter on Reach Reports | Added an enhancement to allow the Reach supervisor user to select a date/time range for start/stop rather than just a date start/stop for the following reports: Agent Group Productivity Agent Productivity CDR Queue Group Traffic Overview Queue Traffic Overview | Enhancement | Reach |
UC-2885 | Outbound call status message could be improved | Added an enhancement to improve the status messages provided in the Reach GUI. Steps to reproduce: An agent answers a voicemail call and decides to call back the client. If the client rejects the call then an error message is displayed: "Cannot connect to outbound call destination: USER_BUSY" If the client does not answer the call another error message is displayed: "Cannot connect to outbound call destination: RECOVERY_ON_TIMER_EXPIRE" These dashboard messages could be improved to be more "user friendly". | Enhancement | Reach |
UC-3529 | Reach Freeswitch Outbound Fixes | Umbrella issue for enhancements and fixes to the outbound functionality in Reach. Fixes to consider for freeswitch_outbound: - consistent agents states - the state of an agent should be reflected under the supervisor agent manager widget - CDRs for outbound calls - recordings for outbound calls - caller id for outbound calls - outbound voicemails, conferences and callbacks - transfers to external numbers - permission profiles for allow outbound - integration of the outbound call widget into the session manager | Enhancement | Reach |
UC-3530 | Integrate outbound call widget with the session manager in Reach dashboard | Added an enhancement to integrate the outbound call widget directly in the Session Manager widget. Currently, if a Reach agent wants to make an outbound call, he can use the outbound call widget that he can add on demand into his dashboard. We should integrate this widget with the session manager widget, so that the agent can make outbound calls from the session manager. He will be allowed to transfer or create a conference with another agent in the same way as with inbound calls. | Enhancement | Reach |
UC-3581 | Auto Logout of Reach Agent | Added an enhancement that will automatically log an agent out of the system if they have been sitting in the release mode for a specified period of time. Add a feature bit on the Agent Group admin page for Reach that will allow the admin to set whether or not they want agent's to be auto logged out and if so, what the timer will be set to. The new fields on the admin UI should be displayed as follows: Auto Logout (checkbox) Release Duration (text box) These fields should be at the bottom of the Agent Group admin page. By default, the checkbox should be unchecked, indicating that an agent should never be automatically logged off. When "Auto Logout" is unchecked, the Release Duration text box should be disabled. When "Auto Logout" is checked, "Release Duration" should be enabled and the admin should be able to place a numeric value into this field. There should be "help text" associated with the "Release Duration" field that says ... "the number of minutes that an agent can sit in the release mode before being automatically logged out by the system". If "Auto Logout" is on, Reach should set a timer when an agent goes into the release state. This timer should be set to the number of minutes configured in the "Release Duration" field. Whether the agent places themselves into the release state or the system places them into the release state due to them not answering call(s) that they were offered will not matter. In both cases the system should set the timer. If the timer expires, the agent should be automatically logged off. Reach should provide a notification on the agent's portal that indicates that they have been logged off due to inactivity (like the notification that is provided when an agent is automatically placed into the release mode). | Enhancement | Reach |
UC-3722 | Reach Use Advance Login allows clicking "use advanced" without a URI entered | Added an enhancement to the ReachME login page to have the Use Advanced Login button greyed out and not selectable until the user puts a URI in the advance login area. As of 15.08, once a reach user logs in they are sent to a screen that allows them to use advanced login options. This assumes that the user that logged in has the correct permission to allow them access to advanced login options. They are presented with 2 buttons as follows: Use Advanced Login (in grey) Continue (in blue) Currently, they can click the Use Advanced Login button even if they do not put anything in the Advanced Options section (e.g. a sip URI). If they do that, they are logged in but they will be offered calls on their normal extension (registered line). This is a bit counter intuitive. This request is to have the Use Advanced Login button greyed out and not selectable until the user puts a URI in the advance login area. Once they enter a URI there, the Use Advanced Login button should be enabled and turned to blue. In addition, by default, we should have the Advanced Options sections opened up so that the Endpoint field is displayed by default. | Enhancement | Reach |
UC-3723 | Short version of the CDR report | An enhancement that adds a new short version of the CDR report and will include the following data. For all "initial" calls into Reach (segment 1 calls), show CDR details with the following data fields: Queue Line Agent Login Agent Name Disposition Caller ANI Caller ID Offered To Agent Date/Time Accepted by Agent Date/Time Time in Queue Call End Date/Time Call Duration Input parameters for execution of the report will minimally include start date and end date. | Enhancement | Reach |
UC-3727 | Cannot control outbound call actions through permission profiles | Added an enhancement to permission profiles currently to control what can be done through the outbound call widget . I am mainly referring to being allowed to transfer calls or make conferences via the outbound widget. The existing permissions like Transfer/Conference to Agent/Number apply only to inbound calls. Currently you can only make the outbound widget available or unavailable and that's it. | Enhancement | Reach |
UC-3728 | Add Reach Permission checkbox for "Allow Outbound" | Added as part of the rewrite of the outbound functionality enhancements. Specifically UC-3530 that will integrate outbound into the session manager. With the outbound widget being removed by UC-3530, permissions work will need to be done as follows: Remove the Outbound Widget from the "accessible widgets" list. Add a "Allow Outbound" check box. If this new checkbox is selected, the user will be able to use the outbound functionality integrated in the session manager. If not checked, the outbound functionality will not be usable. Not displayed at all would be preferable. Greying out the functions would be acceptable. | Enhancement | Reach |
UC-3729 | Add columns to Queue Traffic Overview Reach report | Fixed an issue with the outbound call widget that was created when an Agent rejected an incoming ReachME call which caused the call to go to Voicemail. Steps to reproduce: 1. Agent sets his status to available Result : agent is now idle. 2. Agent adds an Outbound call widget on his dashboard Result : successfully added.Status is set to Idle. 3. Agent selects a customer from the Outbound call widget ->Select a customer drop down. Result : agent is alerted by incoming call from reach.The outbound call widget displays status - pre-call 4. Agent Rejects the incoming call Issue : the outbound call widget and the session manager widget still show "pre-call". The Outbound call widget still waits for the agent to enter a number. Workaround : You can bypass this if you refresh the dashboard page. But if you keep waiting for aprox 2 minutes after the reject - reach log will display an error : "[error] <0.21401.1> gen_fsm <0.21401.1> in state awaiting_destination terminated with reason: no such process or port in call to gen_fsm:sync_send_all_state_event(<0.21249.1>, dump_state)" Followed by : "[error] <0.21401.1> CRASH REPORT Process <0.21401.1> with 0 neighbours exited with reason: no such process or port in call to gen_fsm:sync_send_all_state_event(<0.21249.1>, dump_state) in gen_fsm:terminate/7 line 600" (sipxreach_server2.log) If you don't refresh the page and choose to go ahead and enter a number and call it, then that number is called but as soon as it answers it somehow goes to the agent voicemail.. | Enhancement | Reach |
UC-3730 | Add columns to Queue Group Traffic Overview Reach report | An enhancement that added 3 columns to the Queue Group Traffic Overview report as follows: Sent To Voicemail Transfer Out SLA The first 2 new columns will be counts of calls that were sent out of the queue based on a recipe setup. Specifically, the "Sent To Voicemail" column will be a count of calls that went to queue voicemail. The "Transfer Out" column will be a count of calls sent out of the queue to some other destination based on a transfer to sip uri recipe step. In addition, these calls that make up these 2 counts will no longer be counted as abandons. Thus, the abandon count needs to be updated to ensure that calls sent to extension or voicemail are no longer included. For the 3rd new column, add a column called SLA (service level agreement) For this SLA column, allow the user to input a value to define what their SLA is. An SLA is typically something like X % of calls answered within Y seconds. So, we will want the user to input the number of seconds (Y value). From that we will then fill this new column with data that is the % of calls answered within # of seconds defined by Y (the user input). Example: User puts in 60 seconds for the input control (representing 60 seconds as the target answer time) To fill in the SLA column we will need to get the following: NumCalls = total count of calls NumCallsUnderSLA = total count of calls answered within 60 seconds (Time in Q from CDR for that Call) Then a simple division NumCallsUnderSLA/NumCalls will give us the % to display. | Enhancement | Reach |
UC-3786 | JasperReports deployment improvement/update | Added an enhancement to update JasperReports deployment process allowing us to release new reports. | Enhancement | Reach |
UC-3798 | Embed JasperServer report viewer into Reach dashboard | Added an enhancement to embed JasperServer (JRS) report viewer into Reach dashboard's REPORTS tab using an iFrame and JRS HTTP api. Main advantage is avoiding duplicate work on input controls and easier, automatic way of reports integration into Reach interface. Critical part is customization of report viewer's GUI to fit Reach look and feel. | Enhancement | Reach |
UC-3835 | Upgrade JasperServer to 6.2 | Enhancement to update Jasper Server to Version 6.2 in support of new reporting capabilities. In ReachME 15.12 and earlier Jasper Server Version 5.2 offers only old and deprecated REST v1 API. For the new jasper reports deployment mechanism we should use the v2 of the API. useful link: http://community.jaspersoft.com/documentation/tibco-jasperreports-server-upgrade-guide/v610/introduction | Enhancement | Reach |
UC-3857 | Remove locking 'findAndModify' on agent_report and rstats | Added an enhancement to address a Reach performance issue. When two 'findAndModify' operations are executed by Reach they would put a lock on the database, which can badly influence the performance. From mongodb.log: 2015-12-22T10:20:25.567-0500 [conn44] warning: ClientCursor::staticYield can't unlock b/c of recursive lock ns: top: { opid: 45800690, active: true, secs_running: 0, microsecs_running: 2824, op: "query", ns: "reach_reports.agent_report", query: { findAndModify: "agent_report", query: { info: null, _id.agent: "1006", _id.vsn: 1 }, update: { $set: { info: { first_name: "Erlinda", last_name: "Edmunds", profile: "MainGroup" } } } }, client: "10.254.4.251:42399", desc: "conn44", threadId: "0x7f91899b1700", connectionId: 44, locks: { ^: "w", ^reach_reports: "W" }, waitingForLock: false, numYields: 0, lockStats: { timeLockedMicros: {}, timeAcquiringMicros: { r: 0, w: 8 } } } We can mitigate this by using update instead of findAndModify. | Enhancement | Reach |
UC-3876 | Advanced login is not working for outbound calls | Fixed an issue that didn't allow advanced login to work with outbound calls. A Reach agent is allowed to specify a sip endpoint at the login step. All calls will be routed at that endpoint and not at the agent's extension (which is the default). This option was not taken into consideration for outbound calls. When making an outbound call, the phone registered with the agent's extension will now ring, regardless of the specified sip endpoint. | Enhancement | Reach |
UC-3877 | Add filter for media direction to CDR reports | CDR reports should have filter for media direction. Possible values should be "all", "inbound", "outbound". Main CDR report should have two new columns: Media Direction and Media Type. | Enhancement | Reach |
UC-3917 | CDR data consistency for transfers and conferences with external numbers | Enhancement to display in Reach CDR a row for each external call segment. Currently, we do not track external call segments, as stated in UC-3116. Nonetheless, we have in the raw events collection information about an external transfer or conference being done. We should display in Reach CDR a row for each of these segments and mark as unknown those details that we do not track (e.g. accepted by endpoint timestamp). | Enhancement | Reach |
UC-1768 | Call is not recorded when agent calls back | Fixed an issue that caused agent call-backs to not record the call. Setup: Configure a client for Reach, enable call recording. Add a line and a queue for this client and an agent that can receive calls. Steps to reproduce: A client calls the Reach extension and is waiting in queue. While waiting, the client presses star and leaves a voicemail. Put the agent available. The voicemail is now routed to the agent. After listening to the message, the agent calls back the client. The client answers the call. Reported problem: In supervisor's Call Recording widget the audio file corresponding to this call is not available. Expected result: The call between the agent and the client should be recorded. 2015-06-03 14:09:05.450 [error] <0.31681.0> gen_fsm <0.31681.0> in state outbound_ringing terminated with reason: bad record client in freeswitch_outbound:outbound_ringing/2 line 201 2015-06-03 14:09:05.450 [error] <0.31681.0> CRASH REPORT Process <0.31681.0> with 0 neighbours exited with reason: bad record client in freeswitch_outbound:outbound_ringing/2 line 201 in gen_fsm:terminate/7 line 622 | Fix | Reach |
UC-1905 | Call Recording should be included in backup archive | Fixed an issue that caused Reach Recordings to not be included in backups. Configure Reach, enable call recording for a client and perform some calls. Take a backup of the system. The backup archive should include the recorded calls. When performing a restore, in supervisor's dashboard, the call recording widget will display "Audio file not available" for all the calls that were not restored. | Fix | Reach |
UC-2902 | One way talk path between supervisors on the same barge call | Fixed an issue that caused a one-way talk path between multiple Supervisors on the same barge call. Steps to reproduce: Agent is in a call with a client. Supervisor1 barges the call. The supervisor can hear both the agent and the client, both the agent and the client can hear the supervisor. Supervisor2 barges the same call. Reported problem: Supervisor1, the agent and the client can hear Supervisor2, but Supervisor2 can hear only the agent and the client. So, Supervisor1 is not heard by Supervisor2. | Fix | Reach |
UC-2910 | Truncate Reach client recordings in MongoDB and disk | Fixed an issue similar to UC-2324, where the reach.recordings collection in mongo should also be truncated based on the "Days to Retain Recordings" value in the client configuration. This mongo collection holds information about the calls which is displayed on Call Recording gadget of the Reach dashboard. After the Reach recordings change done in UC-2051 (recordings are kept now in MongoFS), the periodic cleanup should be done for: - recordings metadata - gridfs files - disk files The algorithm is: ``` for each client with call recordings enabled do: delete disk directories older than the retain date (today - (days_to_keep)) delete old recordings from gridfs (db: reach; collections: fs.files, fs.chunks) delete old recordings metadata (db: reach; collection: recordings) ``` Log what was deleted in /var/log/sipxpbx/reach-recordings-cleanup.log | Fix | Reach |
UC-3164 | Cannot delete agent from UI, if default group for reach agents is deleted | Fixed an issue whereas the Administrator could not delete an agent from GUI if default group for reach agents is deleted. Config: openUC (14.10.20140828070812 2014-08-28EDT07:08:23 localhost.localdomain) Users already existent in openUC. Steps to reproduce: 1. Make an user a Reach agent. AdminPortal --> Features --> OpenUC Reach --> Agents --> Add New Agent --> select Agent Group : default --> select user --> OK. 2. Check Users --> User Groups. If this is the first agent you will see that a new user group is created: OpenUC-reach-agents and the user which now is an agent is added to that group, else the user is just added to this group. 3. From whatever reason the User Group: OpenUC-reach-agents is deleted. 4. Try to delete the previously added agent: OpenUC Reach --> Agents --> delete selected agent. Reported Problem: The agent cannot be deleted. An error page is presented. Comment: Issue reproduces in 14.04 too. Tested on openUC (14.04.20140430131046 2014-04-30EDT13:04:44 localhost.localdomain) | Fix | Reach |
UC-3192 | Data about Outbound Calls missing from CDR Reports | Fixed an issue with Reach CDR Report and outbound calls, no data about outbound calls could be retrieved. This information should be also accessible in CDR reports. Direction field is in this case outbound. | Fix | Reach |
UC-3204 | Reach Avatar displays wrong in Agent Manager widget | Fixed an issue caused by custom client avatars in the Agent manager. When a client avatar is defined, the agent manager displays this avatar over the top of the default handshake avatar in the agent manager. Steps to recreate: 1. Define an avatar on a client in Reach 2. Place a call into a queue such that the client with the avatar is assigned on the call 3. Answer the call with an agent 4. View the agent manager to see the answered call connected to an agent | Fix | Reach |
UC-3328 | Conference of Reach VM does NOT work when Number is selected from Transfer/Conference | Fixed an issue with conferencing a Reach Voicemail did not work properly when a number was selected from the Transfer / Conference drop-down list. Steps to reproduce: 1. Client calls to Reach line and leaves a voicemail. 2. When an agent becomes available the voicemail message is offered as a VM Callback. 3. When agent answers, the voicemail is played. 4. While listening the message the agent selects Transfer/Conference - Number. Reported problem: When Conference button is pressed nothing is triggered and the following message appears in dashboard: "Agent not available for conference" | Fix | Reach |
UC-3381 | If the client's name is longer than 18 characters the name is no longer displayed in supervisor's dashboard - Call Recordings | Fixed a display issue where if in Reach the client's name was longer than 18 characters the name is no longer displayed in supervisor's dashboard - Call Recordings. Three dots are displayed instead (...). Steps to reproduce: 1. Set in OpenUC Reach a client name longer than 18 characters. 2. Have some completed calls to Reach in order to have some recordings. 3. Check in supervisor's dashboard - Call Recordings widget that the client's name is not displayed. See picture attached. | Fix | Reach |
UC-3567 | When transferring an outbound call, "Outbound Call" is displayed instead of customer outbound caller id | Fixed a display issue when transferring an outbound call, "Outbound Call" was displayed instead of the outbound CallerID. Steps to reproduce: 1.Setup a client with Outbound CallerID - "ezuce support" 2.As an agent,login to dashboard and add the Outbound call widget 3.Use the widget to place a call to this agent (select client,answer call from Reach,dial the clients number and call) 4.Client answers call 5.Agent transfers outbound call to some other number Reported issue : "Outbound Call" is displayed at the last endpoint instead of the defined client Outbound CallerID | Fix | Reach |
UC-3572 | Only one of the agent names is displayed in the Agent Productivity report if the agent names exceed 14 characters in total | Fixed an issue where only one of the Agent names was displayed in the Agent Productivity Report if an Agent name exceeded 14 characters. Steps to reproduce: 1.Setup an user with first & last names that in total exceed 14 characters . 2.Configure Reach and add the above user as an Agent. 3.Call a queue and have this agent answer some calls so that the reach reports can be populated. 4.As a supervisor generate an Agent Productivity report and verify the Agent Name column Issue: the agent displays only one of his names - either the first name or the last name. Workaround : change the first/last name of the user to have under 15 characters (space between names is considered a character) | Fix | Reach |
UC-3591 | Problems moving data from events to call-segment_facts collection | Fixed an issue with call data collection. A customer complained about the lack of reports. During troubleshooting we discovered that this data is on events collection from reach_report but is not moved in call_segment_facts where it was needed. | Fix | Reach |
UC-3641 | Outbound call widget when agent rejects incoming ReachME call | Fixed an issue with the outbound call widget that was created when an Agent rejected an incoming ReachME call which caused the call to go to Voicemail. Steps to reproduce: 1.Agent sets his status to available Result : agent is now idle. 2.Agent adds an Outbound call widget on his dashboard Result : successfully added.Status is set to Idle. 3.Agents selects a customer from the Outbound call widget ->Select a customer drop down. Result : agent is alerted by incoming call from reach.The outbound call widget displays status - pre-call 4.Agent Rejects the incoming call Issue : the outbound call widget and the session manager widget still show "pre-call". The Outbound call widget still waits for the agent to enter a number. Workaround : You can bypass this if you refresh the dashboard page. But if you keep waiting for aprox 2 minutes after the reject - reach log will display an error : "[error] <0.21401.1> gen_fsm <0.21401.1> in state awaiting_destination terminated with reason: no such process or port in call to gen_fsm:sync_send_all_state_event(<0.21249.1>, dump_state)" Followed by : "[error] <0.21401.1> CRASH REPORT Process <0.21401.1> with 0 neighbours exited with reason: no such process or port in call to gen_fsm:sync_send_all_state_event(<0.21249.1>, dump_state) in gen_fsm:terminate/7 line 600" (sipxreach_server2.log) If you don't refresh the page and choose to go ahead and enter a number and call it,then that number is called but as soon as it answers it somehow goes to the agent voicemail.. | Fix | Reach |
UC-3649 | In conferences with outbound calls, agent and conference participant will always see caller id as "Reach" | Fixed an issue with outbound calls where in the Agent widget the Agents would see the outbound CallerID as "Reach" instead of the actual called number. Steps to reproduce: 1.Agent calls customer using the Outbound call widget : selects the customer from the drop-down. Result : after selecting the customer,the agent receives a pre-call ,his phone displays "From:Reach Reach". 2.He answers the incoming call and then dials the customer number in the outbound widget,to which the customer answers. Result : the agent keeps seeing the callerID "From:Reach" 3.The agent uses the outbound call widget to convert the current call into a conference by inviting another extension Result : both the agent and the other conference participant will see the callerID as : "From:Reach" Expected : at step 3 to see the Agent CallerID or the Outbound CallerID (if that one was setup) Not sure if on steps 1 & 2 the caller ID is the correct one. | Fix | Reach |
UC-3660 | Cannot save Reach report in PDF format from Chrome | Fixed an issue where a user could not download the PDF version of a report. After placing some calls, the user generated a Reach report from the supervisor dashboard. I was able to view the report in all available formats including PDF. Issue: i cannot download the PDF version from Chrome browser. I can download the other versions. I can download the PDF version from Firefox/IE. | Fix | Reach |
UC-3696 | Transfer on outbound calls via a voice mail call back is not working. | Fixed an issue where when an agent transferred a call that was initiated from a VoiceMail callback the callback was dropped. Steps: 1.Caller calls Line 303 where Agent 1 is Released. 2.Caller presses * to leave a VM in this queue and then presses * again to finish. 3.Agent is alerted by the newly left VM when returning to idle state. 4.Agent successfully listens the VM and then from the Dashboard clicks the Callback icon. 5.Caller is called back. 6.While on the line with caller, Agent 1 performs a call Transfer to Agent 2 from the Transfer/Conference Widget. 7.Agent 2 answers the call. Reported Problem : Agent 2 should have the called back party on the line. Actual Result : The call back is dropped, Agent 2 is hearing VM left by caller to Agent 1. After listening the routed VM, 2nd agent is capable of doing another callback to reach the caller from the Callback widget. | Fix | Reach |
UC-3817 | Outbound call widget when called party rejects outbound call | Fixed an issue where if the called party rejects an outbound call initiated by a Reach agent, then the call is not cleared correctly. The first leg of the call would still remain active, the agent must close it manually from his phone. Nonetheless, the state of the agent is correctly updated. When the same agent tries to make another outbound call, then he will receive the following error: RCV: {"jsonrpc":"2.0","error":{"code":2,"message":"existing_call"},"id":722} | Fix | Reach |
UC-3820 | Reach Call shows Talk Time but No Offered To Agent Time | Fixed an issue where a customer reported that Reach CDR shows a call that had talk time (CPT) with no Offered To Agent date/time stamp. This should not be possible. | Fix | Reach |
UC-3831 | Conferences to external numbers for outbound calls | Fixed some issues with Outbound calling for Agents. Currently, an agent has the possibility to convert an outbound call into a conference and add new parties. This feature had some issues, as follows: - When the third party rejects the conference invite call or does not answer, the dashboard will still show the agent as being on conference with that party - The agent cannot add more than a party because the transfer / conference drop down box will not be shown after the first conference invite. All those issues were found after integrating outbound call widget into the session manager widget. | Fix | Reach |
UC-3843 | Idle time not correctly computed in Reach | Fixed an issue where the Idle Time was not properly computed in the Agent Portal. We have few reports from couple offices when they press idle button on acd portal we display time under the idle/release button showing how long they have been idle. Few users are reporting that even tho when they stay idle for exact 1 hour without changing status or taking call somehow the time shows little less than 1 hour under idle." The problem appears to be timer.js module, that computes the time based on a setInterval method called each second. It increments the number of seconds by one each second, instead of using date/time differences between the current time and the started time. | Fix | Reach |
UC-3845 | Reach Call Stuck in Ringing State | Fixed an issue with agents that did not have a phone assigned to them in the system. The call would show as ringing but never be offered to another agent. This may happen in other cases but it was only really noticed when an agent did not have a phone assigned to them at all in the system. Clearly this means that there would be no registration for a phone for that user as well then. The scenario is as follows: Agent logs into the Reach portal with valid credentials but has no phone. Agent goes idle Call is offered to the agent but the agent cannot answer the call. What should happen is that the agent is set to the release mode and the call is pulled back and offered to another agent. What actually happened is that the call continued to show as being ringing both in the agent reach portal and in the queue manager widget and the call was never offered to another agent. | Fix | Reach |
UC-3850 | Call Recording Widget Icons Are Not Centered | Fixed a graphics issue in the Agent portal that caused icons to not be displayed in the middle of buttons. | Fix | Reach |
UC-3852 | Advanced Login Does Not Allow regular phone numbers | Fixed an issue caused by a data validation check that that ensured that an agent using the advanced login option enters a sip uri in the end point field caused regular phone numbers to not be accepted. This results in a user wanting to use a PSTN number (e.g. 2318232979) must enter something like 2318232979 followed by @domain. So, this workaround was required: 2318232979@domain The used to be able to simply put in the PSTN number like this: 2318232979 This is much better from a usability perspective since users dial external numbers with just a simple 10 digit or perhaps 1+10 or 91+10. We want to remove this check that was added and perhaps replace it with something that checks for only crazy special characters that would not be allowable. | Fix | Reach |
UC-3860 | Reach Callback when the agent is in a conference is not working | Fixed an issue for calling back a caller when the agent was already in a conference (this would be used if the caller got disconnected). Currently, when an agent receives a voicemail, he is allowed to conference in other parties. If he is in a conference and then tries to make a callback to the original caller, the conference would be ended and the caller would just be in a normal call with the agent. Expected behaviour: Add the caller to the conference. | Fix | Reach |
UC-3863 | Audio not cut through for Reach call transferred and then answered during Media Annc | Fixed an issue where calls transferred from one queue to another and then answered during a media announcement in the second queue were not properly connected. To recreate: call enters Sales queue call is offered to agent 2000 and is answered agent 2000 transfers the call to queue General agent 2000 is offered the call again out of that queue but did not answer (this step is not needed to recreate) caller begins hearing media announcement from recipe step on General queue at 20 seconds in the queue agent 2001 becomes available during the media announcement playback to the caller agent 2001 answers the call Note that general queue has a recipe step to play a hold media file every 20 seconds. result: Caller is left hearing in queue like treatment (MOH and media announcements) Agent is left with dead air Call seems to be connected and both legs show up in cdr and in fs_cli | Fix | Reach |
UC-3871 | VM stream can only be played once, if voicemail slider reaches the end | Fixed an issue with the voicemail slider where if it reached the end of the message the stream could not be listened to again. Reported problem: Agent receives a voicemail call, answers and hears the message. The voicemail stream can only be heard until the voicemail slider reaches the end. After that point, even if the slider is moved at the start position, pressing the play button will not resume the stream. | Fix | Reach |
UC-3872 | Issues with websocket connection keep-alive in Reach | Fixed an issue with the ReachME Dashboard and websocket connections being closed randomly. Reach dashboard opens a websocket connection on 8937 / 8936 ports when agents are logged in. Over that connection, the agents are being served realtime statistics, call queues updates and more. There is a keep-alive mechanism implemented to ensure the websocket connection is always open. Nonetheless, this mechanism is not correctly working in the latest Reach version. If a Reach agent remains logged in for a longer period of time, the websocket connection is closed (and not reopened), which makes the dashboard freezing. That agent will not receive calls any more, even if he will see himself as "available". In the browser console, the error displayed is: "Cannot read property 'status' of undefined". This issue could be caused by the embedding of Jasper report viewer in Reach dashboard. | Fix | Reach |
UC-3873 | Missing collections from Reach recordings backup | Fixed an issue with backup of Reach data, the mongo collections fs.files and fs.chunks were not added to the archive. | Fix | Reach |
UC-3874 | Trim client name in agent manager widget | Fixed an issue that caused a display problem in Agent Manager widget if the client name was too long. | Fix | Reach |
UC-3875 | Reach is not trying to reconnect to jasperserver in case the service is down and then returned to service | Fixed an issue such that if the Jasper Server failed or had not started yet, the Reach dashboard would not try to reconnect. Possible scenarios for this issue: - reach process started, but for some reason jasperserver has not started yet; if an agent would login in reach dashboard, he won't see reports, but no indication of an error would be displayed - if jasper server stops while agents are connected in the dashboard, then they won't see reports, but again, no indication of an error would be displayed. We should display the error as a notification and while jasperserver is starting, a loader instead of an empty reports drop down list. | Fix | Reach |
UC-3883 | Update CDR jrxml to account for outbound calls CDRs | Reach CDRs for outbound calls were implemented in UC-3192. This created the need to make some changes to the underlying CDR jrxml file, in order to properly display CDRs for outbound calls: - rename several columns: Caller ANI - Endpoint ANI Caller ID - Originating Caller ID Offered to Agent - Offered to Endpoint Accepted by Agent - Accepted by Endpoint Time in Queue - Time in Queue/Precall - for columns like 'Queue', 'Line' etc. specify that those fields are not applicable by NA (instead of empty entry) - update formulas timestamps: Time In Queue/Precall = Accepted by Endpoint - Date/Time OR Time in Queue/Precall = CallEnd - Date/Time Call Duration = Time in Precall OR Call Duration = Call End - Date/Time CPT = Call End - Accepted by Endpoint OR CPT = 0 | Fix | Reach |
UC-3910 | Queue page not showing inherited skills correctly | Fixed an issue with the Admin GUI where the Queue page was not showing inherited skills correctly. | Fix | Reach |
UC-3914 | Conference leg with external number fails if the third party is invited by an agent different then the first agent | Fixed an issue that caused a conference leg with external number fails if the third party is invited by an agent different then the first agent. Steps to reproduce: 1. A call is placed into a Reach line 2. The call is routed to an agent, the agent takes the call 3. The agent decides to make a conference with another agent 4. The other agent is invited to the conference and answers the conf call 5. The other agent decides to invite a third party (external number) to the conference. Issue: The external leg fails with function_clause. From Reach log: 11:10:31.445 [error] <0.1229.0>Undefined - Undefined:Undefined:Undefined - gen_fsm <0.1229.0> in state oncall terminated with reason: {{function_clause,[{agent,get_endpoint,[freeswitch_conference,"1000"],[{file,"/home/nlazar/devel/sipxecs/reach-app/apps/reach_core/src/agent.erl"},{line,287}]},{freeswitch_conference,handle_conference_to_outband,7,[{file,"/home/nlazar/devel/sipxecs/reach-app/apps/reach_freeswitch/src/freeswitch_conference.erl"},{line,191}]},{gen_media,oncall,3,[{file,"/home/nlazar/devel/sipxecs/reach-app/apps/reach_core/src/gen_media.erl"},{line,1925}]},{gen_fsm,handle_msg,7,[{file,"gen_fsm.erl"},{line,505}]},...]},...} This issue is not reproducible if the first agent (the one that creates the interaction) invites the third party to the conference. | Fix | Reach |
UC-3682 | Reach warm transfer not working correctly | There is a known issue when answering a Reach call from a Bria Softphone (Stretto for CCS and Bria 3 3.5.5) when selecting transfer/conference the hold button on the portal placed the Bria on hold, this is an issue because it's not obvious to the agent. The call does not release from hold unless manually done from the Bria. If only happens on each initial call to the agent, for example if the call is transferred back the agent, the agent can perform a transfer/conference without thee Bria going on hold. Tested on 15.05 and 14.04, 14.04 has no issue." Steps to reproduce .. : We did further testing. Behavior for Polycom is the same as Bria. Please note Bria was on my PC over VPN and Polycom was on my home network registered through the Ingate SBC. On Reach3 15.05 (see packet capture from Bria "reach3_Bria_hold.pcapng" 1. Agent "AGENT1" answer call 2. Agent places caller on hold - Bria goes on hold 3. Agent "AGENT1" conferences agent "lca" - Agent "AGENT2" answers ***-Bria for agent "AGENT1" does not release from hold*** - Agents cannot converse, (If agent "AGENT1" clicks hold button in UI, customer and agent "AGENT2" can converse). 4. Agent "AGENT1" manually takes Bria off hold 5. All parties converse. On Reach1 14.04 (see packet capture from Bria "reach1_Bria_hold.pcapng" 1. Agent "AGENT1" answer call 2. Agent places caller on hold - Bria goes on hold 3. Agent "AGENT1" conferences agent "AGENT2" - Agent "AGENT2" answers -Bria for agent "AGENT1" releases from hold - Agents can converse, 4. Agent "AGENT1" click hold in UI taking customer off hold 5. All parties converse. Additional information : " I took a closer look at each Notify hold and talk request and the "To:" Address in 14.04 is the sip uri of the agent in both hold and talk. In 15.05 the hold has the correct sip uri while talk changes the "To:" Address to the callers sip uri Multiple PCAPs are attached within the Zendesk ticket, both Bria and Polycom. Workaround is to downgrade Freeswitch to 1.4.6. Please contact TAC for assistance in doing this | Known Issue | Reach |
UC-1382 | Consultative transfer must point to the exact gateway that accepted the initial INVITE | Fixed an issue caused by multiple gateways on a single dial plan, consultative transfer needed to point to the exact gateway that accepted the initial INVITE. When there are multiple gateways configured to handle the call, there is no guaranty that the initial INVITE and the transferred INVITE with replaces will follow the same forked path. The proxy should be able to tag the REPLACEs header with the absolute URI of the gateway that accepted the initial INVITE. The registrar fallback plugin should be able to parse this tag and return it as a solitary contact. How to replicate: 1. Create 2 gateways in the same branch. Let's call them gw01 and gw02 respectively. 2. Point a dial plan to both gw01 and gw02 as tandem failover 3. Bring down gw01 4. Call a user and put it on hold. 5. Using line 2, call the dial plan you have created (gw01 is down so gw02 should accept the call). 6. Bring up gw01 7. Transfer the call. Since gw01 is now online, the transferred call will go to gw01. This will fail because the call being replaced is in gw02. This is just a manual scenario. In a real deployment, all sorts of timeouts can happen making forks failover which increases the chance that an attended transfer hits a different gw. | Fix | SIPCore |
UC-3882 | SSS should consider REFER messages | Fixed an issue with SSS and the way it handles REFER messages. For example, when call retrieved from park orbit by pressing speed dial button with presence, phone is instructed to call *78(park), in sample below *781100. So the INVITE that goes out from phone has To header *781100. When phone tries to park call second time to 1100 it will try to park the call with the To header *781100. This is done in 2 steps: INVITE to *781100 then REFER to 1100 SSS should treat properly this case (the REFER message), by looking at Refer-To header and marking that extension as busy | Fix | SSS |
UC-3892 | A monitored call park extension will show Busy status when it is actually not | Fixed an issue where a park orbit may appear to be busy but there are no calls in the orbit. Steps to reproduce: Park call, un-park, park it again but this time don't un-park the call - make the user waiting in the park extension to drop the call. Issue: the one who monitors the call park extension via presence speed dial , will see the status of the call park extension as still being busy,so it basically remains stuck - though the call park extension is now empty since the one waiting in it,dropped the call. You have to restart sss or the monitoring phone, or do another park-unpark to clear the stuck presence | Fix | SSS |
UW-168 | Allow upload of avatar from the user portal | Added an enhancement to allow users to be able to upload avatar from user portal. | Enhancement | Unite |
UW-319 | Don't Display Conference Bridge Section in Unite | Added an enhancement to not display Conference bridge section if user has no conference bridges defined. Currently Unite will show a conference bridge section even if there are no conferences defined for the user. In that section it will show what appears to be an error message stating: Feature Not Available Please Contact your Administrator This causes confusion for the users since and they do indeed "contact their administrator" even though there is nothing wrong but rather the feature is just not turned on for them. We should remove that section all together in the Unite display if there are no conferences for the user. | Enhancement | Unite |
UW-321 | In UniteWeb and Unite Lite voicemail is not sorted by date | Added an enhancement to sort voicemail messages by date instead of by name. | Enhancement | Unite |
UW-206 | Settings: change pin or password - no validation message | Fixed an issue where if a password is input with less than 8 chars the input changes border color, but there is no validation message. If a user is LDAP managed they should not be able to change their password and there should be a message as such. When changing PIN there is no validation message either. | Fix | Unite |
SIPX-424 | Yealink plugin V8X | Enhanced the Yealink plugin with new phones, add firmware version 8 model files, and correct number of lines on T4 phones. | Enhancement | Yealink |
SIPX-434 | Yealink 7x/8x misleading tooltip | For Yealink Firmware Template 7x/8x Preference > Backlight Time (seconds) Currently the config "Always" means backlight always off Configuration of 0 seconds will mean always on Updated tooltip to make it more reasonable | Fix | Yealink |