Important note about SSL VPN compatibility for 20.0 MR1 with EoL SFOS versions and UTM9 OS. Learn more in the release notes.

This discussion has been locked.
You can no longer post new replies to this discussion. If you have a question you can start a new discussion

Intermittent RTP audio after SIP connects ok

We are running an XG135 with SFOS 17.5.6 MR-9 installed. The system connects to a fibre internet connection of 1GbS bidirectional. We have around 30 computers on a single subnet lan that is primarily Windows 10 based. We have a Cisco Unified Messaging VOIP system which has been working since July 2019.

For the majority of the time everything is working fine, however every so often (i.e. intermittently during a normal work day) we are unable to get VOIP audio after a phone call connection (SIP) is made. The audio is lost both ways despite the SIP connection being made. This happens at random times during the day and for periods of up to 30 minutes.

At other times phone calls are great and audio quality is perfect. We have had our fibre provider check the line for any issues however the have provided reports that show no runts, retransmissions, session errors, jitter etc. Line latency is constant and inline with the physical distance between the office and the provider.

We have done the usual things on the Sophos XG 135 list UDP timeout increase, we have loaded / unload H.323 and SIP modules etc. nothing resolves this issue.

 

Can anyone tell me if they have seen similar issues with this XG series and firmware version with respect to VOIP/SIP/RTP please.

Alternatively should I be creating specific firewall rules (bidirectional or otherwise) for the RTP voice traffic? All the Sophos articles I have read tell me as long as the internal device completes the SIP connection there should be an open channel between receiving / calling parties and RTP traffic should just work (which it seems to do 99% of the time during the day)



This thread was automatically locked due to age.
Parents
  • Hi Graham,

    check the CISCO logs for the number of SIP sessions and see if that is the limit of the PABX, also check the XG for the maximum number of SIP sessions that it can connect at the same time.

    Where to look, I would suggest you search hither KBAs for documentation.

    Ian

    XG115W - v20.0.3 MR-3 - Home

    XG on VM 8 - v21 GA

    If a post solves your question please use the 'Verify Answer' button.

  • In addition to what Ian suggests, make sure to prioritise VoIP traffic by applying a proper qos policy.

    Also check from reporting, wan traffic consumption.

    If you cannot get nothing from the pbx, switch and any other voip components, a tcpdump is suggested.

Reply
  • In addition to what Ian suggests, make sure to prioritise VoIP traffic by applying a proper qos policy.

    Also check from reporting, wan traffic consumption.

    If you cannot get nothing from the pbx, switch and any other voip components, a tcpdump is suggested.

Children
  • Thanks for the information - much appreciated.

    I have enabled 'Enforce guaranteed bandwidth' in the Traffic Shaping settings screen under system services.  have then created a schedule for the built-in default 'Voip Guarantee' rule under Traffic Shaping (the second last menu option) to cover my users for business hours.

    Question though, do I need to load or unload the SIP module in the system console? I see lots of people saying unload the module and others who say set it to loaded. Currently I have two modules unloaded these are: H.323 and SIP

    I have scheduled a meeting with a Cisco UC boffin in the morning so we shall see what transpires on the Cisco side. But for now I would like to know the best setting for those 2 system modules.

    I have not been able to find the SIP session limit command in XG documentation (yet) I shall keep looking.

    Thanks

    GrahamG

  • Both modules should be on. Try to increase the timeout as suggested here: