[OpenSIPS-Devel] [opensips] Dispatcher strange balancing (#387)

[Rik Broers]

Re-opening.

1.10.4 release, 2 april 2015 tar.gz from website.
Case: 2 sets of 30 destinations. None inactive.
Algorithm 0, so hash over call-id.

Observed behaviour after around 36 hours.
First destination had at some point 600 channels,
second destination 300 - 400
third 200
fourth 100
fifth 60
and this keeps going till the twelfth, fourteenth destination, all destinations after that didn't seem to handle any calls.

All calls initially get dispatched to the first location until this one can no longer handle it, then the failure_route kicks in and dispatches to the following location. I have logs to back up this behaviour.

My development environment is not large enough to simulate with this many destinations.

As this is 1.10.4 maybe remove the backport from it? I don't know if this is 1.10.4 only or also exists in 11 and 2.1.

---
Reply to this email directly or view it on GitHub:
https://github.com/OpenSIPS/opensips/issues/387#issuecomment-96975851
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.opensips.org/pipermail/devel/attachments/20150428/399e06be/attachment.htm>