Upgraded USB HDD - buffer overruns, failure

Forum for questions and support relating to the 1.30.x releases only.
Locked
moonsky
Posts: 6
Joined: Wed Feb 08, 2017 10:58 pm

Upgraded USB HDD - buffer overruns, failure

Post by moonsky »

My Zoneminder was running pretty solid with 5 cameras and Mocord saving events to a 1TB Western Digital MyBook USB 2.0 HDD, but I wanted more space and better performance reviewing monitors. (frames would lag 5-10 seconds when seeking through the images in montage review).

I recently upgraded the external HDD to a Seagate 5TB External Desktop HDD (STEB5000100), USB 3.0. Followed the "Using a dedicated HDD" tutorial again like I did the first time.

After upgrading I have been getting abysmal performance! Regularly getting buffer overruns and 'zma -m 1' has not stopped at yy/mm/dd hh:mm:ss. Sending KILL to pid xxx, eventually the entire zoneminder service stops. I have tried reducing the FPS on the cameras from 10FPS -> 5FPS, got rid of motion detection, and disabled a camera to no avail.

I can't imagine that the performance on the Seagate HDD would be that bad in comparison to the WD drive, especially with the USB 2.0 -> 3.0 update. I have tried multiple file systems for the new drive (ext4, reiserfs, XFS) and even reinstalling ubuntu server with a fresh install of zoneminder. All efforts produce the same results. Is there anything that I am missing?

Also, if it ends up being the drive, does anyone have any suggestions for storage? I can't do an internal drive expansion due to space. Would an NAS or NFS perform better?

Thanks!

Code: Select all

2017-02-07 21:30:04.202700	zmwatch		23514	ERR	Unable to run "/usr/bin/zmdc.pl restart zma -m 2", output is "Unable to connect to server"	zmwatch.pl	
2017-02-07 21:30:04.200459	zma_m4		23484	INF	Monitor-4: 30159 - Opening new event 3069, section start	zm_monitor.cpp	1532
2017-02-07 21:30:04.200319	zma_m3		23783	INF	Monitor-3: 23334 - Opening new event 3067, section start	zm_monitor.cpp	1532
2017-02-07 21:30:04.199866	zma_m2		24418	INF	Monitor-2: 1816 - Opening new event 3066, section start	zm_monitor.cpp	1532
2017-02-07 21:30:04.196758	zmc_m4		23482	WAR	Last image read from shared memory 31 seconds ago, zma may have gone away	zm_monitor.cpp	3091
2017-02-07 21:30:04.187043	zmc_m4		23482	WAR	Buffer overrun at index 19, image 33269, slow down capture, speed up analysis or increase ring buffer size	zm_monitor.cpp	3085
2017-02-07 21:30:04.136850	zmpkg		24452	INF	Command: stop	zmpkg.pl	
2017-02-07 21:30:04.126660	zmpkg		24452	INF	Sanity checking States table...	zmpkg.pl	
2017-02-07 21:30:03.996969	zma_m5		23497	INF	Monitor-5: 31372 - Opening new event 3068, section start	zm_monitor.cpp	1532
2017-02-07 21:30:03.966590	zmwatch		23514	INF	Restarting analysis daemon for Monitor-2	zmwatch.pl	
2017-02-07 21:30:03.947400	zmwatch		23514	INF	Analysis daemon for Monitor-2 needs restarting, time since last analysis 13 seconds (1486524586-1486524573)	zmwatch.pl	
2017-02-07 21:30:03.933420	zma_m2		24418	INF	Monitor-2: 1816 - Closing event 3064, section end	zm_monitor.cpp	1512
2017-02-07 21:30:03.932340	zmdc		23413	INF	'zma -m 1' crashed, signal 8	zmdc.pl	
2017-02-07 21:30:03.928898	zma_m4		23484	INF	Monitor-4: 30159 - Closing event 3060, section end	zm_monitor.cpp	1512
2017-02-07 21:30:03.928743	zma_m5		23497	INF	Monitor-5: 31372 - Closing event 3063, section end	zm_monitor.cpp	1512
2017-02-07 21:30:03.926059	zma_m3		23783	INF	Monitor-3: 23334 - Closing event 3061, section end	zm_monitor.cpp	1512
2017-02-07 21:30:03.920900	zmdc		23413	WAR	'zma -m 1' has not stopped at 17/02/07 21:30:03. Sending KILL to pid 24415	zmdc.pl	
2017-02-07 21:30:02.907560	zmdc		23413	WAR	'zma -m 1' has not stopped at 17/02/07 21:30:02. Sending KILL to pid 24415	zmdc.pl	
2017-02-07 21:30:01.897200	zmdc		23413	WAR	'zma -m 1' has not stopped at 17/02/07 21:30:01. Sending KILL to pid 24415	zmdc.pl	
2017-02-07 21:30:00.886880	zmdc		23413	WAR	'zma -m 1' has not stopped at 17/02/07 21:30:00. Sending KILL to pid 24415	zmdc.pl	
2017-02-07 21:29:59.876900	zmdc		23413	WAR	'zma -m 1' has not stopped at 17/02/07 21:29:59. Sending KILL to pid 24415	zmdc.pl	
2017-02-07 21:29:58.861210	zmdc		23413	WAR	'zma -m 1' has not stopped at 17/02/07 21:29:58. Sending KILL to pid 24415	zmdc.pl	
2017-02-07 21:29:57.845840	zmdc		23413	WAR	'zma -m 1' has not stopped at 17/02/07 21:29:57. Sending KILL to pid 24415	zmdc.pl	
2017-02-07 21:29:56.840620	zmdc		23413	WAR	'zma -m 1' has not stopped at 17/02/07 21:29:56. Sending KILL to pid 24415	zmdc.pl	
2017-02-07 21:29:55.823910	zmdc		23413	WAR	'zma -m 1' has not stopped at 17/02/07 21:29:55. Sending KILL to pid 24415	zmdc.pl	
2017-02-07 21:29:54.810230	zmdc		23413	WAR	'zma -m 1' has not stopped at 17/02/07 21:29:54. Sending KILL to pid 24415	zmdc.pl	
2017-02-07 21:29:53.805290	zmdc		23413	WAR	'zma -m 1' has not stopped at 17/02/07 21:29:53. Sending KILL to pid 24415	zmdc.pl	
2017-02-07 21:29:52.989726	zmc_m1		23443	WAR	Last image read from shared memory 20 seconds ago, zma may have gone away	zm_monitor.cpp	3091
2017-02-07 21:29:52.980194	zmc_m1		23443	WAR	Buffer overrun at index 2, image 34502, slow down capture, speed up analysis or increase ring buffer size	zm_monitor.cpp	3085
2017-02-07 21:29:52.795820	zmdc		23413	WAR	'zma -m 1' has not stopped at 17/02/07 21:29:52. Sending KILL to pid 24415	zmdc.pl	
2017-02-07 21:29:52.747093	zmc_m5		23495	WAR	Last image read from shared memory 20 seconds ago, zma may have gone away	zm_monitor.cpp	3091
2017-02-07 21:29:52.742678	zmc_m5		23495	WAR	Buffer overrun at index 44, image 34494, slow down capture, speed up analysis or increase ring buffer size	zm_monitor.cpp	3085
2017-02-07 21:29:46.779560	zmdc		23413	INF	'zma -m 1' sending stop to pid 24415 at 17/02/07 21:29:46	zmdc.pl	
2017-02-07 21:29:46.646610	zmwatch		23514	INF	Restarting analysis daemon for Monitor-1	zmwatch.pl	
2017-02-07 21:29:46.642000	zmwatch		23514	INF	Analysis daemon for Monitor-1 needs restarting, time since last analysis 14 seconds (1486524586-1486524572)	zmwatch.pl	
2017-02-07 21:29:43.346015	zmc_m3		23469	WAR	Last image read from shared memory 11 seconds ago, zma may have gone away	zm_monitor.cpp	3091
2017-02-07 21:29:43.336696	zmc_m3		23469	WAR	Buffer overrun at index 3, image 33153, slow down capture, speed up analysis or increase ring buffer size	zm_monitor.cpp	3085
2017-02-07 21:29:43.236246	zmc_m4		23482	WAR	Last image read from shared memory 11 seconds ago, zma may have gone away	zm_monitor.cpp	3091
2017-02-07 21:29:43.226536	zmc_m4		23482	WAR	Buffer overrun at index 18, image 33168, slow down capture, speed up analysis or increase ring buffer size	zm_monitor.cpp	3085
2017-02-07 21:29:42.985792	zmc_m1		23443	WAR	Buffer overrun at index 2, image 34452, slow down capture, speed up analysis or increase ring buffer size	zm_monitor.cpp	3085
2017-02-07 21:29:42.731156	zmc_m5		23495	WAR	Buffer overrun at index 44, image 34444, slow down capture, speed up analysis or increase ring buffer size	zm_monitor.cpp	3085
moonsky
Posts: 6
Joined: Wed Feb 08, 2017 10:58 pm

Re: Upgraded USB HDD - buffer overruns, failure

Post by moonsky »

Switched back to the old drive and performance went back to normal. Definitely the Seagate HDD! :x
bbunge
Posts: 2956
Joined: Mon Mar 26, 2012 11:40 am
Location: Pennsylvania

Re: Upgraded USB HDD - buffer overruns, failure

Post by bbunge »

The only Seagate drive that worked well for me on Linux was a video surveillance drive. In fact, most Seagate drives have shown errors and failed faster than any other. Try a WD USB 3.
User avatar
iconnor
Posts: 3274
Joined: Fri Oct 29, 2010 1:43 am
Location: Toronto
Contact:

Re: Upgraded USB HDD - buffer overruns, failure

Post by iconnor »

i wonder if it is spinning down or something maybe you can tune it.
Locked