Hello all, I have a problem and I don’t know where to turn, so of course I turn to you guys! J

 

The problem is that I often found my server in a stalled state, this is an Gentoo server running 2.16.16.20 and python 2.4 with all the four python transports (nice 10) and ejabberd jabber server. I long has thought it to be a memory problem, now I did a memory scan test (memtest+ I think it was called) and no errors. The symptoms are that I can alt+f2 alt+f3 etc to change the logon screen, but no text can be entered. Ctrl+alt+del does not work either, and no network traffic is getting trough.

 

So I installed a new server, ubuntu server, with ejabberd, erland and the four python transports as the only software. And again it happened, this morning I found my server stalled, exactly the same error state. No Kernel panic, no input accepted, and this is an ESX server that is hosting other servers, no hw failures 100% secure.

 

So now I need help to find the problem. I must admit I suspect one of the transports eating cpu away to the point where I cannot use my server no more. How can this happen, and how can I control it? Ejabberd is hosting much larger sites, I think if this was a ejabberd bug it would have been known by now. My only guess is you guys. PS!, on the ubuntu server I do not nice the processes.

 

Any ideas? Anyone else experience similar things? The only thing I can think of right now is to try and disable one and one transport until the problem goes away, the only thing is that it can take a week before the problem happens.

 

About versions, ejabberd 1.1.2, pymsn-t latest svn, pyyimt latest cvs, pyicqt and pyaimt is 0.8.

 

Best desperate regards

Stian B. Barmen

www.jabber.no

 

 

 

_______________________________________________
py-transports mailing list
[email protected]
http://www.modevia.com/cgi-bin/mailman/listinfo/py-transports

Reply via email to