ZMDC restarting every 10 seconds

jetheller · Post by **jetheller** » Wed Jun 15, 2005 8:18 pm

For some reason my capture dameon is restarting every 10 seconds. I just installed Fedora Core 4 and then installed 1.21.1. My network video capture is working correctly, but only updates once every 10 seconds. I installed 1.21.0 just for kicks and it works like a charm, so the problem is particular to 1.21.1.

Post by **zoneminder** » Wed Jun 15, 2005 10:02 pm

Do you have anything from /var/log/messages or anywhere that might give an indication of any errors?

Phil

jetheller · Post by **jetheller** » Thu Jun 16, 2005 12:01 am

Here are the interesting logs:
==> zmdc.log <==
'zmc -m 1' stopping at 05/06/15 19:59:35
'zmc -m 1' died at 05/06/15 19:59:35
'zmc -m 1' started at 05/06/15 19:59:35
'zmc -m 1' starting at 05/06/15 19:59:35, pid = 17157

==> zmwatch.log <==
Restarting capture daemon ('/usr/local/bin/zmdc.pl restart zmc -m 1'), time since last capture 9 seconds (1118879974-1118879965)
'zmc -m 1' started at 05/06/15 19:59:35
'zmc -m 1' starting at 05/06/15 19:59:35, pid = 17157

and then /var/log/messages that coinicides with the event

Jun 15 20:00:16 localhost zmc_m1[17587]: INF [Got TERM signal, exiting]
Jun 15 20:00:16 localhost zmc_m1[17587]: ERR [Select error: Interrupted system call]
Jun 15 20:00:16 localhost zmc_m1[17723]: INF [Debug Level = 0, Debug Log = <none>]
Jun 15 20:00:16 localhost zmc_m1[17723]: INF [Starting Capture]

That's all I get. That and when I view the monitor I see 0.00fps and the one shot per ten seconds in the montage.

jetheller · Post by **jetheller** » Thu Jun 16, 2005 12:03 am

After watching the logs scroll by for about 5 minutes I noticed this in /var/logs/messages, a continueing stream of these messages:
Jun 15 20:02:40 localhost zmc_m1[17157]: ERR [Unable to extract HTTP status from header]
Jun 15 20:02:41 localhost zmc_m1[17836]: ERR [Unable to extract HTTP status from header]
Jun 15 20:02:41 localhost zmc_m1[18121]: ERR [Unable to extract HTTP status from header]
Jun 15 20:02:42 localhost zmc_m1[17157]: ERR [Unable to extract HTTP status from header]
Jun 15 20:02:42 localhost zmc_m1[17836]: ERR [Unable to extract HTTP status from header]
Jun 15 20:02:43 localhost zmc_m1[18121]: ERR [Unable to extract HTTP status from header]
Jun 15 20:02:43 localhost zmc_m1[17157]: ERR [Unable to extract HTTP status from header]

Post by **zoneminder** » Thu Jun 16, 2005 9:13 am

Can you increase the debug level of zmc and rerun it from the command line? It seems that it's getting an invalid response from your camera.

Phil

jetheller · Post by **jetheller** » Thu Jun 16, 2005 12:03 pm

zmc -h doesn't show how to set the debug level and I didn't see it anywhere in the options of the application. How do I run zmc in a higher debug level?

SyRenity · Post by **SyRenity** » Thu Jun 16, 2005 12:59 pm

Hi jet.

Just define a higher debug level via enviroment variable, like this:

export ZM_DBG_LEVEL_zmc=4

Phil, I have seen in the manual, that you can set the debug level via signals - does it mean you can kill -(something) zmc, and it will change the debug level on fly? What is the correct command form of "kill" then it should be?

jetheller · Post by **jetheller** » Thu Jun 16, 2005 1:23 pm

Okay zmc -m 1 (my only monitor) at debug level 4 puts a fast stream of this into the zmcd.log file
Jun 16 09:20:42 localhost zmc_m1[28486]: DB3 [Expecting 1460 bytes]
Jun 16 09:20:42 localhost zmc_m1[28486]: DB3 [Read 1460 bytes]
Jun 16 09:20:42 localhost zmc_m1[28486]: DB3 [Unable to extract subheader from stream, retrying]
Jun 16 09:20:42 localhost zmc_m1[28486]: DB3 [Expecting 134 bytes]
Jun 16 09:20:42 localhost zmc_m1[28486]: DB3 [Read 134 bytes]
Jun 16 09:20:42 localhost zmc_m1[28486]: DB3 [Unable to extract subheader from stream, retrying]

I'm using a DCS-900 which I've never had problems with before and still works fine with 1.21.0

Post by **zoneminder** » Fri Jun 17, 2005 1:22 pm

These look fine. Are you saying the previous ERR messages are no longeer displayed?

Phil

jetheller · Post by **jetheller** » Fri Jun 17, 2005 2:45 pm

The original errors still happen, but that last log was just with the zmc running alone. I'm not sure how to run zmc in error mode and still run all the other processes? Do I manually start up zmc with the env set for debugging, and then start the other processes through the web gui?

Post by **zoneminder** » Fri Jun 17, 2005 4:10 pm

zmc is the primary process for capture so it doesn't actually need any other processes to run, to does whether any are affect what errors it might produce.

Phil

jetheller · Post by **jetheller** » Fri Jun 17, 2005 4:41 pm

With zmc started manually with debug level 4 and the rest of the processes started via the web gui here is an except from tail -f /tmp/*.log

==> zmdc.log <==
'zmc -m 1' stopping at 05/06/17 12:31:18
'zmc -m 1' died at 05/06/17 12:31:18
'zmc -m 1' started at 05/06/17 12:31:18
'zmc -m 1' starting at 05/06/17 12:31:18, pid = 10508

==> zmwatch.log <==
'zmc -m 1' starting at 05/06/17 12:31:18, pid = 10508
'zmc -m 1' started at 05/06/17 12:31:18

==> zmdc.log <==
'zmc -m 1' stopping at 05/06/17 12:31:29
'zmc -m 1' died at 05/06/17 12:31:29
'zmc -m 1' started at 05/06/17 12:31:29
'zmc -m 1' starting at 05/06/17 12:31:29, pid = 10526

==> zmwatch.log <==
Restarting capture daemon ('/usr/local/bin/zmdc.pl restart zmc -m 1'), time since last capture 9 seconds (1119025888-1119025879)
'zmc -m 1' started at 05/06/17 12:31:29
'zmc -m 1' starting at 05/06/17 12:31:29, pid = 10526

The logs from /var/log/messages are the same as before. I'm not sure what more debugging info I can provide. But if this doesn't help, I'll just switch back to 1.21.0 and wait for the next release

Thanks again for the awsome application.