Viewing live stream stops without reason.

[ihr]

I have a zoneminder ver. 1.28.1 running og CentOS7.
Zoneminder was installed from zmrepo, and it generally runs fine, but from time to time it stops the live streaming of all camera's, but the input streams from the cameras apparently continues, as the recognition and saving of events continues normally.
I have tried to export the log-file, but it results in just a file with zero content. Instead of this, I have attached an image of the latest log-file contents.
The only way to remedy this, is a reboot of the server. A restart of zoneminder won't do it!

Does anybody have a clue how to fix this problem.

Regards - IHR

[ihr]

I can add a little more information to the above problem:
The mentioned socket (zms-611021s.sock) does indeed exist, and it's owned by apache:apache.
Is it normal with hundreds of socket-files in the /var/lib/zoneminder/sock directory ?
I don't thimk the problem has anything to do with the pc watching the ZM-console, as I have tried 2 different pc's with 2 different browsers.

IHR

[bbunge]

Which web browser are you using? Are you leaving the "main" Zoneminder page open when you view the montage? Is your PATH_SWAP set to /dev/shm?

[ihr]

I'm using Mozilla Firefox and Google Chrome. The main Zoneminder page is left open when viewing the montage, and PATH_SWAP is poiting to /dev/shm.

[ihr]

Today I updated zoneminder to version 1.29.0 from zmrepo to see if this brings any change to my problem.
Immediately after upgrading, it wasn't possible to view any camera - I simply got no camera-picture at all.
A reboot of the OS brought it back to normal, and everything is working now. I just have to wait, maybe weeks, to see if it stays that way!

[dazed]

I have this problem also...I thought the problem was related to mysql (I even tried to tune it to no avail).
The problem usually appears during/after viewing it remotely. I even tried turning down view settings and bitrate.

I'm running 12.04 and am waiting for the 1.29 to be out of beta...maybe it will fix this issue....So far my temporary fix is daily cron reboot.

Please keep me posted if it fixed your problems...my Zoneminder page is also left open (dint know this would be an issue) when viewing the montage, and PATH_SWAP is pointing to /dev/shm.

[ihr]

@dazed - Just a status update!

It has been 10 days since I updated zoneminder to 1.29.0. I did it the "official" way ie. "yum update" which means not only zoneminder but also the whole system was updated. As I wrote ealier it might be weeks between incidents, so there is nothing definate yet, but it looks good.

I think the combination of CentOS7 and Zoneminder 1.29 is very successfull.

[ihr]

It has now been 4 weeks since I updated Zoneminder to version 1.29.0, and I havn't hat a glitch since. So I consider the problem as solved.
As I have said before: CentOS7 and Zoneminder 1.29 seems to be a very good combination.

IHR

[bbunge]

It has now been 4 weeks since I updated Zoneminder to version 1.29.0, and I havn't hat a glitch since. So I consider the problem as solved.
As I have said before: CentOS7 and Zoneminder 1.29 seems to be a very good combination.

IHR

Keep an eye on the database. Mariadb seems to like to use a fair amount of memory for innodb_buffer_pool_size as the number of events builds up. I'm seeing about 65K events over 30 days. My innodb_buffer_pool_size is just under 1 GIG and I have the value set to 2 GIG just-in-case. Use mysqltuner.pl to check if you need to tweak your database settings. Google for how to do this.

bb

[dazed]

Mine has' not been acting up either (1.28.1)...but It may due to the fact I upgraded my internet speeds (logs used to say throttling)
...and a daily cron reboot (although I was still having issued until I upgraded my upload speed).

Im gonna take a stab and say something in the throttling down the send stream was causing an issue.