[asterisk-bugs] [JIRA] (ASTERISK-21235) Call Quality Degrades When Second Network Interface is Brought Online

[Daniel McElveen (JIRA)]

Daniel McElveen created ASTERISK-21235:
------------------------------------------

             Summary: Call Quality Degrades When Second Network Interface is Brought Online
                 Key: ASTERISK-21235
                 URL: https://issues.asterisk.org/jira/browse/ASTERISK-21235
             Project: Asterisk
          Issue Type: Information Request
      Security Level: None
          Components: General
    Affects Versions: 1.8.11.1
         Environment: Linux version 2.6.32-5-amd64 (Debian 2.6.32-46) (dannf at debian.org) (gcc version 4.3.5 (Debian 4.3.5-4) ) #1 SMP Sun Sep 23 11:00:33 UTC 2012
            Reporter: Daniel McElveen


Customer Description:

We have remote agents attempting to connect over a VPN network to our hosting facility which has a secondary Asterisk server that connects via IAX over a dedicated public internet link to our local primary Asterisk sever. Our local agents connect to the local primary Asterisk server over the LAN. There are no call quality issues with the local LAN agents when only a single network interface (eth0) is up and running. Upon enabling the second network interface (eth1) and without connecting any remote agents, 30 seconds into every call local LAN agents experience call quality issues with every other or third word dropping, a choppy conversation. As soon as the secondary network interface (eth1) is disabled, calls go back to normal quality.

Technical Support Action:

I've notified the customer that they are sharing an IRQ between a Digium TE121 and one of the network interfaces (IRQ 10). I've also notified the customer that XWindows and Asterisk historically don't play very nice with each other, and recommended that X be removed. Customer has performed the following troubleshooting steps over the course of a few days:

- eth1 is configured to different subnet from eth0 
- Routing tables are set up to properly route traffic; SIP traffic is bound to the 1.198 subnet 
- Call degradation is variable: Test Session 1 reports 80% of calls have bad quality. Test Session 2 has 20%. Test session 3 10%. 
- Bringing down eth1 absolutely removes call quality issues 100%

As of this posting, the customer is generating a new autosupport dump that is a snapshot of the system when the issue is present; current autosupport dumps are taken before the second network interface is brought up. Calls over the PRI do not experience call quality degradation, it is only the SIP calls for the agents on the local LAN.

I've gone through everything I know to do on this one; anyone else have any ideas?

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.asterisk.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira