possible bus loading problem during resync
I'm working on 2 systems that are mainly for running vdr. I've had these running somewhat for awhile with raid. But a couple nights ago as I was quitting for the night, I noticed one of the computers drive light staying on. I had just made some changes to xine and didn't know if something had crashed. Turned on the TV and found the video was freezing for 10-20secs every 10-20secs. Logging in using putty and winscp I found it very sluggish to respond.Starting top I found it was doing the regular array check/resync. The process was using about 64% cpu and cpu was staying at idle speed (1000Mhz). These computers use Athlon64 x2 cpu's. A problem with the AN2 socket systems is that when the cpu is throttled back, it also slows the bus. This has been found to be a problem on boards with integrated graphics when using nvidia's vdpau for hardware video decoding because they use system ram. The fix is to set the lower speed limit to 1800Mhz and/or change the up_threshold to ~50% . However, I am using PCIe video cards and so up till now have not had a problem.
I stopped vdr, but putty and winscp where still sluggish. This tells me that it is loading the bus so much that both the video card and the network is effected. it would also effect any tuner cards interfering with any recording that may be going on at the time. I change the up_threshold from the default 95% to 50% which should kick the speed up when it's syncing. But I'm not sure that will be enough. Could there be some other setting that is wrong raising the priority of the process? Seems like this would be a problem for any system having raid maintenance bring the system to its knees. The eta to finish was 75 minutes.
Quoted 16 lines · Timothy D. Lenz
I'm working on 2 systems that are mainly for running vdr. I've had these running somewhat for awhile with raid. But a couple nights ago as I was quitting for the night, I noticed one of the computers drive light staying on. I had just made some changes to xine and didn't know if something had crashed. Turned on the TV and found the video was freezing for 10-20secs every 10-20secs. Logging in using putty and winscp I found it very sluggish to respond.Starting top I found it was doing the regular array check/resync. The process was using about 64% cpu and cpu was staying at idle speed (1000Mhz). These computers use Athlon64 x2 cpu's. A problem with the AN2 socket systems is that when the cpu is throttled back, it also slows the bus. This has been found to be a problem on boards with integrated graphics when using nvidia's vdpau for hardware video decoding because they use system ram. The fix is to set the lower speed limit to 1800Mhz and/or change the
up_thresholdto ~50% . However, I am using PCIe video cards and so up till now have not had a problem.
Quoted 10 lines
I stopped vdr, but putty and winscp where still sluggish. This tells me that it is loading the bus so much that both the video card and the network is effected. it would also effect any tuner cards interfering with any recording that may be going on at the time. I change the
up_thresholdfrom the default 95% to 50% which should kick the speed up when it's syncing. But I'm not sure that will be enough. Could there be some other setting that is wrong raising the priority of the process? Seems like this would be a problem for any system having raid maintenance bring the system to its knees. The eta to finish was 75 minutes.
--
Sorry about the incredibly brief answer: Not to dismiss other issues, but that behavior seems like exactly what I've seen when a disk has been failing.
-- Kristleifur
Quoted 13 lines · Kristleifur Daðason
Quoted 9 lines · Timothy D. Lenz
I'm working on 2 systems that are mainly for running vdr. I've had these running somewhat for awhile with raid. But a couple nights ago as I was quitting for the night, I noticed one of the computers drive light staying on. I had just made some changes to xine and didn't know if something had crashed. Turned on the TV and found the video was freezing for 10-20secs every 10-20secs. Logging in using putty and winscp I found it very sluggish to respond.Starting top I found it was doing the regular array check/resync....... --
Quoted 4 lines
Sorry about the incredibly brief answer: Not to dismiss other issues, but that behavior seems like exactly what I've seen when a disk has been failing.
If that is true, how does that happen, the driver is hung? But anyway, how can such things happen when there is more than one CPU-core?
try disabling NCQ by echo 1 > /sys/block/sdX/device/queue_depth for all drives. After doing this, at most 1 request can be issued to one drive until the drive has serviced such request.
After doing this, firstly I'd say the sluggishness should disappear, at least on SSH when not touching the disks. And then you can look with "iostat -x 1": probably the bad drive will have a service time (svctm) or await much worse than the others.
Just guesses, correct me if I'm wrong
--
To unsubscribe from this list: send the line "unsubscribe linux-raid" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Quoted 14 lines · Asdo
Quoted 13 lines · Kristleifur Daðason
Quoted 9 lines · Timothy D. Lenz
I'm working on 2 systems that are mainly for running vdr. I've had these running somewhat for awhile with raid. But a couple nights ago as I was quitting for the night, I noticed one of the computers drive light staying on. I had just made some changes to xine and didn't know if something had crashed. Turned on the TV and found the video was freezing for 10-20secs every 10-20secs. Logging in using putty and winscp I found it very sluggish to respond.Starting top I found it was doing the regular array check/resync....... --
Quoted 7 lines
Quoted 4 lines
Sorry about the incredibly brief answer: Not to dismiss other issues, but that behavior seems like exactly what I've seen when a disk has been failing.
If that is true, how does that happen, the driver is hung? But anyway, how can such things happen when there is more than one CPU-core?
A drive produces an error, the whole controler hangs and resets all ports, all drives have to finish being reset before any IO can continue. Hapens easily enough.
Quoted 10 lines
try disabling NCQ by echo 1 >
/sys/block/sdX/device/queue_depthfor all drives. After doing this, at most 1 request can be issued to one drive until the drive has serviced such request.After doing this, firstly I'd say the sluggishness should disappear, at least on SSH when not touching the disks. And then you can look with "iostat -x 1": probably the bad drive will have a service time (svctm) or await much worse than the others.
Just guesses, correct me if I'm wrong
What I would start with is check the resync/check speed of the raid and kernel messages. If it is running at high speed and there are no kernel messages about IO errors then it is probably just a case of the IO subsystem being busy. I got similar sluggish behaviour when I increased the stripe cache to 16384 for a reshape.
If there are no hardware problems on the disks causing this then try setting the max speed for the resync lower. That way the resync will leave pauses where other IO and bus activity can happen. The raid should slow down automatically if there is normal IO pending but in my experience that doesn't always work.
MfG
Goswin
--
To unsubscribe from this list: send the line "unsubscribe linux-raid" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Quoted 9 lines
Quoted 4 lines
If that is true, how does that happen, the driver is hung? But anyway, how can such things happen when there is more than one CPU-core?
A drive produces an error, the whole controler hangs and resets all ports, all drives have to finish being reset before any IO can continue. Hapens easily enough.
Ok but this is a multi-core CPU and he said Putty and WinSCP were hung. Ok for WinSCP... but Putty?
Timothy is Putty hung on array check even on NCQ disabled?
What is the resync speed? If it is very high it could be a CPU starvation but it's strange with only 2 drives. If it is very low I am not sure. Are disks write caches enabled?
I am not able to spot any problem from your iostat or dmesg.
Note: yes the iostat -x 1 was supposed to be captured during resync.
Trigger a resync manually for the test. You can start it with echo check and stop it with echo idle > /sys/block/mdX/md/sync_action
Quoted 15 lines · Asdo
Quoted 9 lines
Quoted 4 lines
If that is true, how does that happen, the driver is hung? But anyway, how can such things happen when there is more than one CPU-core?
A drive produces an error, the whole controler hangs and resets all ports, all drives have to finish being reset before any IO can continue. Hapens easily enough.
Ok but this is a multi-core CPU and he said Putty and WinSCP were hung. Ok for WinSCP... but Putty?
Timothy is Putty hung on array check even on NCQ disabled?
Quoted 9 lines
What is the resync speed? If it is very high it could be a CPU starvation but it's strange with only 2 drives. If it is very low I am not sure. Are disks write caches enabled?
I am not able to spot any problem from your iostat or dmesg.
Note: yes the iostat -x 1 was supposed to be captured during resync. Trigger a resync manually for the test. You can start it with echo check and stop it with echo idle >/sys/block/mdX/md/sync_action
I find "gnome-disk-utility" A.K.A. "palimpsest" to be a very good heuristic to tell whether any drives are giving me physical trouble. If there is a high remapped-sector-count or if Palimpsest otherwise thinks a drive is suspect, there is a good chance that any unexplained slowdowns in the machine are due to that drive.
If nothing more, it's a cheap way to get more information.
(My way of getting Palimpsest to check out drives in a machine that doesn't have the program available in its installed distro repositories is to run Ubuntu 9.10 from an USB stick. Boot, and up comes Palimpsest.)
Hope this helps. Best of luck.
-- Kristleifur