Welcome Guest! To enable all features please Login. New Registrations are disabled.

Notification

Icon
Error

Andrew
#1 Posted : Wednesday, October 16, 2013 9:54:56 AM(UTC)
Andrew



Joined: 10/16/2013(UTC)
Posts: 0

Hello,
While looking into why my front camera appears to be dropping flames, I noticed the following errors in the log. They appear to occur every 10 minutes or so.

Can you pls. confirm the actions I need to take (if any) to address this "failure"?


Quote:

2013-10-16 17:40:24 MOTION front Motion detected
2013-10-16 17:40:33 FAILURE Lounge Database queue overflow: 1780 record(s) discarded.
2013-10-16 17:40:33 FAILURE Back Yard2 Database queue overflow: 1044 record(s) discarded.
2013-10-16 17:40:34 FAILURE front Database queue overflow: 553 record(s) discarded.
2013-10-16 17:40:34 FAILURE Garage Database queue overflow: 144 record(s) discarded.
2013-10-16 17:41:43 MOTION front Motion detected


Thanks.
Torsten
#2 Posted : Wednesday, October 16, 2013 7:01:11 PM(UTC)
Torsten



Joined: 2/11/2013(UTC)
Posts: 451

Thanks: 28 times
Was thanked: 127 time(s) in 119 post(s)
Hej Andrew,

"overflow" means that your hard disk is not fast enough to receive the video which the Recording Server wants to write to it.

If your disk is generally fast enough (compared to the number of cams, their frame rate and resolution), then you might want to chase Windows routines causing harddisk I/O. This includes: Virus scans, defragmentation, Windows indexing, data transfer, etc. The can slow down your disk and cause overflow.

On the Milestone knowledge base there is an article #329 which is written for XProtect Corporate, but it might give you some ideas what to consider.

Good luck,
Torsten
Andrew
#3 Posted : Thursday, October 17, 2013 4:07:36 AM(UTC)
Andrew



Joined: 10/16/2013(UTC)
Posts: 0

Thanks for that. I have followed all the (applicable tips).

I initially though indexing, but it's 2008 R2, so no service unless you add it explicitly....so that's not it (however I have unticked all drives anyway). No virus checker or other software is running. Mobile server is even on another machine. Note : I am running this server as a vm under esxi.

The issue seems strange as I am using a Sandisk 250G ssd (with decent IOPS) for recording storage (only ~40GB is in use before it's archived to a dedicated 2TB drive)....but I am still seeing these faults.

It seems to by the system process (as well as the RecordingServer.exe) are hitting the .pic files at the same time. No idea why system is hitting all these files.

I just restarted Milestone and it performed an archive on startup and I have not seen a "Failure" message yet.

When it happens, I can see in Windows Resource Monitor the queue length grow to 10 and the Disk IO hits 100MB/s for a few seconds...then it fauls...hmmm moer investigation required here.
Chris
#4 Posted : Friday, April 29, 2016 2:15:25 AM(UTC)
Chris



Joined: 4/8/2016(UTC)
Posts: 0

Originally Posted by: Andrew Go to Quoted Post
I just restarted Milestone and it performed an archive on startup and I have not seen a "Failure" message yet.

When it happens, I can see in Windows Resource Monitor the queue length grow to 10 and the Disk IO hits 100MB/s for a few seconds...then it fauls...hmmm moer investigation required here.


Did you ever find a solution? I had the same problem with overflows exactly every 10 minutes so I upgraded from a slow laptop hard drive to an 850 EVO SSD and I get the exact same problem, even about the same number of overflows as with the slower drive. I noticed ImageImportService spikes the disk use right before the overflows so I stopped "Image Import Service" but the problem remains. Disabled "Windows Search" and problem remains. "Volume Shadow Copy" was never running. Stopped "SQL Server VSS Writer" and disabled Windows Defender real-time protection and problem remains.

Then I noticed that each "Database queue overflow" error showed the same camera to its left. I disabled that camera and the errors stopped (I'm running two other cameras). When I enabled the camera, it got 20 overflow immediately, then exactly 10 minutes and 1 second later it got 376 records overflowed. The problem camera is an HIKvision DS-2CD2632F-IS. I tried connecting to its console and watching 'top' while the overflow occurred but didn't notice anything spike in usage. Looked in /var/log and there is nothing really logged there. Checked through the web UI for anything that might be running every 10 minutes but couldn't find anything. I'm not sure what else to do at this point.

I'm also wondering what dropping 376 records actually means. It can't mean dropping 367 frames of video because the camera is only running at 10fps so that would be losing 36.7 seconds of video and I'm not seeing any such huge gaps in recordings. In fact I see no gap at all in a recording anywhere near the time the records were dropped.

Edited by user Friday, April 29, 2016 2:27:03 AM(UTC)  | Reason: Not specified

Users browsing this topic
Guest
Forum Jump  
You cannot post new topics in this forum.
You cannot reply to topics in this forum.
You cannot delete your posts in this forum.
You cannot edit your posts in this forum.
You cannot create polls in this forum.
You cannot vote in polls in this forum.