[asterisk-users] Any thoughts on Asterisk 16.17.0 outputting FRACK refcount related messages

[Joshua C. Colp]

On Thu, Sep 23, 2021 at 1:59 PM Dan Cropp <dan at amtelco.com> wrote:

> We have an extremely busy/large customer.  They run fine most of the time,
> but periodically asterisk will output FRACK refcount related messages.  It
> doesn’t seem to be related to the volume, because it’s not breaking during
> their peak times.
>
>
>
> When this happens, the system becomes unstable and they have to restart to
> get things resolved.
>
> To give an idea of the instability, we have seen INVITE/Trying responses
> in SIP messaging logs.
>
> We tell Asterisk to answer via AMI, but Asterisk never sends the OK (even
> 24 seconds later it hasn’t sent).
>
> Eventually the other send CANCEL of the call.
>
>
>
>
>
> We’ve now captured 4 different days where something like the following
> occurs.
>
> 1) Is there a good way to tell if this may be fixed in Asterisk 16.20.0
> (short of upgrading)?
>

You could examine the changes between the two and see if any issues seem
relevant.


> 2) Would this be something I should submit as an asterisk issue?
> Unfortunately, site is so busy capturing the debug will be very difficult
> (if not impossible) due to amount of data.
>

Without a backtrace, or debug logs it'd be really hard to look at. The
message just means that an object was referenced a lot. Why that is - who
knows. It could be that the system can't keep up for some reason, or
there's an off-nominal path, or something is deadlocked, or something else.

-- 
Joshua C. Colp
Asterisk Technical Lead
Sangoma Technologies
Check us out at www.sangoma.com and www.asterisk.org
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.digium.com/pipermail/asterisk-users/attachments/20210923/55027448/attachment.html>