[asterisk-users] Asterisk mem leak behavior?

[Mark Greene]

So here is my setup.

Hardware:
Intel P3 1.2 Ghz
1 GB RAM
36 GB Drives Mirrored

Software:
 CentOS 5
2.6.18 Kernel

Asterisk 1.4.14
Zaptel 1.4.7 (redfone)
LIbpri 1.4.2

I'm using TDMoE with my PRI using a product called fonebridge from a company
called redfone. They require that I use their own build of zaptel and I am
trying to figure out if the problem is with them or something else. The
TDMoE traffic is running over a dedicated interface on the asterisk server.
Nothing but TDMoE traffic goes over this interface, it does not even have an
IP assigned to it.

After a about 6-7 days asterisk will stop passing calls through the PRI. It
will continue to accept calls from extension to extension or to voicemail,
etc. But nothing over zap channels. From the CLI if I check the status of
the channels it shows all is well. If I use zttool it shows all OKs. And the
lights on the fonebridge itself are green across the board as well.

If I issue the command from the CLI to restart asterisk "restart now" it
drops me to another CLI prompt as if it ran the command, but it didn't
asterisk continues to run. It takes me dropping out of the CLI and issueing
an init command to restart "/etc/init.d/asterisk restart". Once asterisk
restarts all is well and things keep moving, but 5 or so days later the same
thing happens.

What's the best way to trouble shoot this? Any and all comments are welcome.
Let me know if you need more details.

- Mark
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.digium.com/pipermail/asterisk-users/attachments/20080129/7e09d38b/attachment.htm