WCS service crash's

[cheinan]

Hi Max
After updating from WCS 5.0.3233 to WCS 5.1.3467
We get a lot of WCS service crash's
Our server is cantos 7
It have 16G ram and 20 processors.
And its seems that always one cpu is on 100%
and all the others are around 6%.
Is there any WCS configuration witch determines the allocation of resources for all processors?
What else can cause this?

Thanks
Cheinan

 

[Max]

Hello

Server can be crashed
1. On transcoding.
You can find crash logs in WCS_HONE/logs/error[PID].log
2. On system Out Of Memory.
In such a case you should see corresponding messages in /var/log/messages

So please send us:
1. Crash log files WCS_HONE/logs/error[PID].log
2. Log files a hour before the crash WCS_HONE/logs/server_logs
3. System messages /var/log/messages
to support@flashphoner.com
4. Configuration at WCS_home/conf

 

[Max]

Regarding CPU,
Try to find thread that utilizes CPU.

1. Dumping threads

jstack 12223 >> 12223.jstack

Here 12223 is PID of WCS process.
Each thread has ID in hex form 0x2213121

2. Then you can run 'top' and press Shift+H to switch thread view.
Here you can see threads, CPU load, and thread ID in decimal form.

So you can match a thread by ID.
Please send us jstack file, top screenshot and matched thread name.