zmc -m 3' exited abnormally, exit status 11

[voronwe]

Hi all

Iam gettin g this funny error on a custemer machine, and the most funny "haha" part is that i had this server running at home for 3 days with no error at all, then today i installed it a the custemer site and there the error showed up, and it tells me nothing at all.(But then again iam no programmer).

Server Config Info :

Motherboard = Asus P5WD2-E Premium
CPU = P4 3.2GHz Dual Core
Ram = 2048
Linux = FC4 i386
ZM Version : 1.22.1
Cams = Axis 221 Firmware 4.3

zmdc.log Errors :

05/03/06 15:51:16.259442 zmdc[4095].ERR ['zmc -m 3' exited abnormally, exit status 11]

Messages Error :

May 3 15:51:16 video zmc_m3[4197]: ERR [Got signal (Segmentation fault), crashing]
May 3 15:51:16 video zmc_m3[4197]: ERR [Signal address is 0xdfbe859c, from 0x805aa77 ]
May 3 15:51:16 video zmc_m3[4197]: ERR [Backtrace: /usr/libexec/zm/bin/zmc [0x805aa77]]
May 3 15:51:16 video zmc_m3[4197]: ERR [Backtrace: /usr/libexec/zm/bin/zmc [0x805aa77]]
May 3 15:51:16 video zmc_m3[4197]: ERR [Backtrace: /usr/libexec/zm/bin/zmc [0x805c8d4]]
May 3 15:51:16 video zmc_m3[4197]: ERR [Backtrace: /usr/libexec/zm/bin/zmc [0x805db69]]
May 3 15:51:16 video zmc_m3[4197]: ERR [Backtrace: /usr/libexec/zm/bin/zmc [0x805fd5b]]
May 3 15:51:16 video zmc_m3[4197]: ERR [Backtrace: /usr/libexec/zm/bin/zmc [0x804b203]]
May 3 15:51:16 video zmc_m3[4197]: ERR [Backtrace: /lib/libc.so.6(__libc_start_main+0xdf) [0x174d5f]]
May 3 15:51:16 video zmc_m3[4197]: ERR [Backtrace: /usr/libexec/zm/bin/zmc(__gxx_personality_v0+0xa1) [0x804aa89]]
May 3 15:51:16 video zmc_m3[4197]: INF [Backtrace complete]

Does anyone have an idea what could be the problem.

To solve the problem ill go install zm 1.21.4 tommorow, and ill see if i can reproduse the error here on my test servers.

[zoneminder]

Can you do the addr2line thing here as well using 0x805aa77 as the address this time.

[voronwe]

Hi Phil

Ill see if the problem is still there tommorrow and then post a reply then, i dont have ssh access to the machine.

[voronwe]

Hi Phil

I have run the command, and the responce follows.

Can you do the addr2line thing here as well using 0x805aa77 as the address this time.

Responce :

Code: Select all

 ??:0

Sadly i havent been able to reproduse this error on my test machine yet, but ill have a go at it when i get home.

[bkortleven]

I have a similar issue, but I cannot find any help here...

This is my setup:
asrock pci-express board, Celeron D 2.8GHz - 1GB DDR400
bttv 878 card, id below
Debian Etch - amd64, backports enabled
zm 1.22.3 running.

Everytime I 'restart' zm (both from console or init.d or so), I get about 2 minutes (if even) of 'normal use' on the console, and then the device goes 'orange'...


Any idea?


These are the lines I get in my /tmp/zmdc.log:

06/24/07 23:00:17.638378 zmdc[27820].INF [Server shutdown at 07/06/24 23:00:17]
06/24/07 23:00:18.785882 zmdc[32368].INF [Server starting at 07/06/24 23:00:18]
06/24/07 23:00:22.047547 zmdc[32368].INF ['zmc -d /dev/video0' starting at 07/06/24 23:00:22, pid = 32391]
06/24/07 23:00:22.406955 zmdc[32368].INF ['zma -m 1' starting at 07/06/24 23:00:22, pid = 32393]
06/24/07 23:00:22.759170 zmdc[32368].INF ['zmfilter.pl' starting at 07/06/24 23:00:22, pid = 32395]
06/24/07 23:00:23.542954 zmdc[32368].INF ['zmaudit.pl -c' starting at 07/06/24 23:00:23, pid = 32397]
06/24/07 23:00:24.171195 zmdc[32368].INF ['zmwatch.pl' starting at 07/06/24 23:00:24, pid = 32400]
06/24/07 23:00:24.766950 zmdc[32368].INF ['zmupdate.pl -c' starting at 07/06/24 23:00:24, pid = 32402]
06/24/07 23:00:27.571207 zmdc[32368].ERR ['zma -m 1' exited abnormally, exit status 11]
06/24/07 23:00:27.572221 zmdc[32368].INF [Starting pending process, zma -m 1]
06/24/07 23:00:27.575400 zmdc[32368].INF ['zma -m 1' starting at 07/06/24 23:00:27, pid = 32411]
06/24/07 23:00:32.859461 zmdc[32368].ERR ['zma -m 1' exited abnormally, exit status 11]
06/24/07 23:00:37.046840 zmdc[32368].INF [Starting pending process, zma -m 1]
06/24/07 23:00:37.051371 zmdc[32368].INF ['zma -m 1' starting at 07/06/24 23:00:37, pid = 32415]
06/24/07 23:00:40.231550 zmdc[32368].ERR ['zma -m 1' exited abnormally, exit status 11]
06/24/07 23:00:50.047102 zmdc[32368].INF [Starting pending process, zma -m 1]
06/24/07 23:00:50.055358 zmdc[32368].INF ['zma -m 1' starting at 07/06/24 23:00:50, pid = 32419]
06/24/07 23:00:54.283467 zmdc[32368].ERR ['zma -m 1' exited abnormally, exit status 11]
06/24/07 23:01:14.099482 zmdc[32368].INF [Starting pending process, zma -m 1]
06/24/07 23:01:14.111729 zmdc[32368].INF ['zma -m 1' starting at 07/06/24 23:01:14, pid = 32426]
06/24/07 23:01:17.536164 zmdc[32368].ERR ['zma -m 1' exited abnormally, exit status 11]




My bttv settings from dmesg

bttv: driver version 0.9.16 loaded
bttv: using 8 buffers with 2080k (520 pages) each for capture
bttv: Bt8xx card found (0).
bttv0: Bt878 (rev 2) at 0000:00:09.0, irq: 17, latency: 32, mmio: 0xff6ff000
bttv0: detected: (Askey Magic/others) TView99 CPH06x [card=38], PCI subsystem ID is 144f:3000
bttv0: using: Askey CPH06X TView99 [card=38,autodetected]
bttv0: gpio: en=00000000, out=00000000 in=00ffeff7 [init]
bttv0: using tuner=1
bttv0: i2c: checking for MSP34xx @ 0x80... not found
bttv0: i2c: checking for TDA9875 @ 0xb0... not found
bttv0: i2c: checking for TDA7432 @ 0x8a... not found
bttv0: i2c: checking for TDA9887 @ 0x86... not found
bttv0: registered device video0
bttv0: registered device vbi0
bttv0: PLL: 28636363 => 35468950 .. ok
input: bttv IR (card=38) as /class/input/input2

[akupsta]

Could be zone related.. Try this:

Disable all zones in your monitor.

Then start enabling one zone at a time. I had this problem a few times. In my case it was related to the zone config.

When you find your problematic zone, delete it and recreate. If it doesn't solve your problem, placement might also be an issue.

e.g. What I noticed:

Imagine you are capturing a patio door from a camera mounted just under the roof.

Two zones are defined, one for the area covering the steps and one for the patio door (different levels of sensitivity). When both zones are enabled (active) ZMA crashes. I don't know why this happens.

[bkortleven]

Don't know what exactly I did, but I changed the zone setup and the orientation of the camera (mounted it onto another wall), and redid config.
It seems to work now, without any errors... No idea what happened

Weird thing is: The physical setup of the camera never changed in the past 4 years. I just moved the software from a seperate server to a shared server (some media stuff running on it, and asterisk), and it started doing this after three or four months, in which I think I hadn't changed anything. Must be some fighting rules or zones, as you proposed.

Bottom line: problem solved!

[akupsta]

Sounds like a bug to me in the zone / motion detect engine. Looks like it has something to do with adjacent zones (on monitors with multiple zones defined), the angles at which zones touch each other, etc.. Geometric in any regard. Perhaps a fix can be added in the next release.


Adam.

[cordel]

We need to be able to reproduce the problem so we can track it down, so the question is how can we produce this error consistantly?
One thing that come to mind is maybe the entry in the database got miffed somehow....
Or the monitor diminsions got changed and the zones were not changed making them larger than the captured image.

Also, Please don't double post. Thx

[akupsta]

Try this to reproduce:

Create two adjacent zones that are perpendicular.

(e.g.) Roof mounted camera looking down on patio door and stair landing.
Zone 1 encompasses the landing, Zone 2 encompasses the door only.

Or you can create one zone outlining just the door and landing.

oMy monitors are always orange when I do this.

[cordel]

I have monitors with several zones but no issues. I really need actual setting as a description just does not do enough to suffice. I need you to create a reproducible error. Unless of course your zones are a secret.

[akupsta]

I'd be more than happy to submit a screenshot. This forum does not accept uploads, is there an email address I can use to submit the screenshots / pics?

Thanks.

[MJN]

I've been able to reproduce this error (or at least zmc failures as per the OP) so hopefully the following may be of some use... (please shout for more info if/as required)

My setup consists of a single USB camera (Philips PCVC680K) and two IP cameras (Panasonic BL-C1's) all running at 3fps.

Over the past couple of days I noticed the logs reporting zmc crashing - I hadn't noticed any ill effects in terms of capturing etc (but that's not to say there weren't any).

An example error being:

Code: Select all

Aug 12 10:43:27 localhost zmc_m1[9611]: ERR [Got signal (Segmentation fault), crashing]
Aug 12 10:43:27 localhost zmc_m1[9611]: ERR [Signal address is 0xdfa53b4c, from 0x805bcc9 ]
Aug 12 10:43:27 localhost zmc_m1[9611]: ERR [Backtrace: /usr/local/bin/zmc [0x805bcc9]]
Aug 12 10:43:27 localhost zmc_m1[9611]: ERR [Backtrace: /usr/local/bin/zmc [0x805bcc9]]
Aug 12 10:43:27 localhost zmc_m1[9611]: ERR [Backtrace: /usr/local/bin/zmc [0x805d016]]
Aug 12 10:43:27 localhost zmc_m1[9611]: ERR [Backtrace: /usr/local/bin/zmc [0x805f69a]]
Aug 12 10:43:27 localhost zmc_m1[9611]: ERR [Backtrace: /usr/local/bin/zmc [0x8060bee]]
Aug 12 10:43:27 localhost zmc_m1[9611]: ERR [Backtrace: /usr/local/bin/zmc [0x804ad71]]
Aug 12 10:43:27 localhost zmc_m1[9611]: ERR [Backtrace: /lib/tls/i686/cmov/libc.so.6(__libc_start_main+0xd2) [0xb7b35ea2]]
Aug 12 10:43:27 localhost zmc_m1[9611]: ERR [Backtrace: /usr/local/bin/zmc(__gxx_personality_v0+0x91) [0x804a521]]
Aug 12 10:43:27 localhost zmc_m1[9611]: INF [Backtrace complete]
Aug 12 10:43:27 localhost zmdc[4703]: ERR ['zmc -m 1' exited abnormally, exit status 11]
Aug 12 10:43:27 localhost zmdc[4703]: INF [Starting pending process, zmc -m 1]
Aug 12 10:43:27 localhost zmdc[4703]: INF ['zmc -m 1' starting at 07/08/12 10:43:27, pid = 9628]
Aug 12 10:43:27 localhost zmdc[9628]: INF ['zmc -m 1' started at 07/08/12 10:43:27]
Aug 12 10:43:27 localhost zmc_m1[9628]: INF [Debug Level = 0, Debug Log = <none>]
Aug 12 10:43:27 localhost zmc_m1[9628]: INF [Starting Capture]

As per the FAQ here is the corresponding addr2line output:

Code: Select all

addr2line -e /usr/local/bin/zmc 0x805d12d 0x805d016 0x805f69a 0x8060bee 0x804ad71 0x804a521
/usr/local/src/ZoneMinder-1.22.3/src/zm_remote_camera.cpp:931
/usr/local/src/ZoneMinder-1.22.3/src/zm_remote_camera.cpp:974
/usr/local/src/ZoneMinder-1.22.3/src/zm_remote_camera.cpp:1079
/usr/local/src/ZoneMinder-1.22.3/src/zm_monitor.cpp:2465
/usr/local/src/ZoneMinder-1.22.3/src/zmc.cpp:249
../sysdeps/i386/elf/start.S:122

Something which soon became apparent was that the zmc crashes were only happening with the IP camera monitors and that it was occuring whether they were in monitor or modect mode. Typically, I've tweaked a number of things recently however one of the perhaps most notable was changing from static JPEGs (source: snapshotjpeg?Resolution=320x240&Quality=Clarity) to MJPEGs (source: nphmotionjpeg?Resolution=320x240&Quality=Clarity). So, last night I changed one of the cameras back to static JPEGs and lo and behold it has not suffered any crashing since. To summarise: only the camera monitor capturing in nphmotionjpeg mode is prone to crashing in this manner.

Is there any more info required to support this report?

Mathew

[cordel]

Alright, I assume that you have restarted the zm services after all these changes are done... correct me if I am mistaken.
With it in the current mode and using any monitor you seen this fault on lets verify the image is truely the same. We will do this once in each type (jpeg and mjpeg). Use zmu to capture one frame from both types.

[MJN]

Alright, I assume that you have restarted the zm services after all these changes are done... correct me if I am mistaken.

Hmm... It's so tempting to just say 'yes, of course' but hand on heart I must admit that I often forget to restart particularly as many settings appear happy to take immediate effect. I tell you what - I've just restarted now and I'll monitor the logs. It seems to be happening with enough regularity that if the problem persists it'll show up soon enough. At least this way I won't be wasting your time.

With it in the current mode and using any monitor you seen this fault on lets verify the image is truely the same. We will do this once in each type (jpeg and mjpeg). Use zmu to capture one frame from both types.

...and it'll give me time to look up how to use zmu to capture a frame!

Will get back to you (and thanks for the quick response and interest).

Mathew