July 19, 201510 yr Hello all ! I have recently upgraded from 5.06 to 6.01 and have noticed that my server has been starting parity checks on it's own. The scheduler is " Disabled " for the parity check option, so i am assuming that's not it. Any insight would be greatly appreciated as my box seems to be running them whenever it feels like it.... sometimes one a week, sometimes every other day. Thanks in advance !! Loving 6.0 aside from this though... awesome job ! EDIT: It would appear that my server shut down maybe ? The server itself is on a UPS and i had no power outages last night, any thoughts ? Jul 19 02:46:18 Beafy su[2052]: Successful su for unraid-plex by root Jul 19 02:46:18 Beafy su[2052]: + root:unraid-plex Jul 19 02:46:20 Beafy pms: Plex Media Server IS running Jul 19 02:46:20 Beafy emhttp: unclean shutdown detected Jul 19 02:46:20 Beafy kernel: mdcmd (41): check CORRECT Jul 19 02:46:20 Beafy kernel: md: recovery thread woken up ... Jul 19 02:46:20 Beafy kernel: md: recovery thread checking parity... Jul 19 02:46:20 Beafy kernel: md: using 1536k window, over a total of 3907018532 blocks.
July 19, 201510 yr Community Expert Let the parity complete. Then shutdown and put your flash in your PC and let it checkdisk.
July 19, 201510 yr Author Let the parity complete. Then shutdown and put your flash in your PC and let it checkdisk. I will do that, thanks - hope it helps ! Do you figure something is wrong with my flash drive ?
July 19, 201510 yr Community Expert Let the parity complete. Then shutdown and put your flash in your PC and let it checkdisk. I will do that, thanks - hope it helps ! Do you figure something is wrong with my flash drive ? Sometimes it can get corruption, especially with an unclean shutdown. unRAID will always start a correcting parity check after an unclean shutdown. If it is doing that even when you think you have shutdown cleanly, then it's possible that unRAID was not able to write the stopped status to config/super.dat because of an issue with the flash. Go to Tools - Diagnostics and post the results and we can look to see if that might be happening. Another thing you can check is the uptime which is displayed in the upper right of the webGUI. If it seems to be less than you expect, then maybe it is shutting down on its own. Is all of the cooling working OK? What is the exact model of your power supply?
July 19, 201510 yr Author Let the parity complete. Then shutdown and put your flash in your PC and let it checkdisk. I will do that, thanks - hope it helps ! Do you figure something is wrong with my flash drive ? Sometimes it can get corruption, especially with an unclean shutdown. unRAID will always start a correcting parity check after an unclean shutdown. If it is doing that even when you think you have shutdown cleanly, then it's possible that unRAID was not able to write the stopped status to config/super.dat because of an issue with the flash. Go to Tools - Diagnostics and post the results and we can look to see if that might be happening. Another thing you can check is the uptime which is displayed in the upper right of the webGUI. If it seems to be less than you expect, then maybe it is shutting down on its own. Is all of the cooling working OK? What is the exact model of your power supply? The server shutdown on its own, so im not exactly sure what happened, but all of this never seemed to be an issue before the swap to 6.0.1 The PSU is a Corsair CX600 and the server is plugged into an APC UPS ES-750 Thank you for your time, it's appreciated. Diag zip attached, thanks again. beafy-diagnostics-20150719-1349.zip
July 19, 201510 yr Community Expert I do see this in your syslog Jul 19 02:44:35 Beafy kernel: FAT-fs (sda1): Volume was not properly unmounted. Some data may be corrupt. Please run fsck. so it wouldn't hurt to do the checkdisk I mentioned. I didn't see anything about it being read-only though so it is not clear there is really an issue there. If you know it is shutting down when it isn't supposed to then that is probably the issue. Maybe the UPS is faulty. If you normally have good power you could try without the UPS for a while. If the CPU or mobo is overheating it can also cause a shutdown which is why I asked about the cooling. You don't have many drives so that PSU should be OK unless it is failing.
July 19, 201510 yr Author I do see this in your syslog Jul 19 02:44:35 Beafy kernel: FAT-fs (sda1): Volume was not properly unmounted. Some data may be corrupt. Please run fsck. so it wouldn't hurt to do the checkdisk I mentioned. I didn't see anything about it being read-only though so it is not clear there is really an issue there. If you know it is shutting down when it isn't supposed to then that is probably the issue. Maybe the UPS is faulty. If you normally have good power you could try without the UPS for a while. If the CPU or mobo is overheating it can also cause a shutdown which is why I asked about the cooling. You don't have many drives so that PSU should be OK unless it is failing. Sounds good, as soon as the parity check is done, i will do the check. The cooling all seems to be fine, that box is in the basement and generally stays very cool. Just to confirm, the possible corruption on the flash would have no bearing on the server rebooting though, right. Or were you saying that it would be the main culprit ? Thanks again so much !!
July 19, 201510 yr Community Expert Just to confirm, the possible corruption on the flash would have no bearing on the server rebooting though, right. Or were you saying that it would be the main culprit ? No - the effect of corruption might be to cause an unexpected parity check to happen after rebooting when you think the system closed down tidily.
July 19, 201510 yr Author No the flash should not cause a reboot. Ok, thanks. I just let windows run a check on the flash and it says all is normal " no problems were found on this device " I guess maybe either my PSU or UPS is failing ? I'll try having it straight into the wall as much as it pains me i suppose, to see if that rectifies the issue... or maybe ill just go buy a new one seeing as that particular unit is no spring chicken. Thanks yet again !
July 19, 201510 yr Author No the flash should not cause a reboot. I've been getting these as well.... Event: unRAID Disk 5 SMART health [188] Subject: Warning [bEAFY] - command timeout is 1 Description: ST3000DM001-1CH166_Z1F1A012 (sdg) Importance: warning Could this have anything to do with it or is it anything to worry about ?
July 19, 201510 yr Author Just to confirm, the possible corruption on the flash would have no bearing on the server rebooting though, right. Or were you saying that it would be the main culprit ? No - the effect of corruption might be to cause an unexpected parity check to happen after rebooting when you think the system closed down tidily. Thank you very much for your input !
July 19, 201510 yr Community Expert No the flash should not cause a reboot. I've been getting these as well.... Event: unRAID Disk 5 SMART health [188] Subject: Warning [bEAFY] - command timeout is 1 Description: ST3000DM001-1CH166_Z1F1A012 (sdg) Importance: warning Could this have anything to do with it or is it anything to worry about ? SMART attributes
July 19, 201510 yr Author No the flash should not cause a reboot. I've been getting these as well.... Event: unRAID Disk 5 SMART health [188] Subject: Warning [bEAFY] - command timeout is 1 Description: ST3000DM001-1CH166_Z1F1A012 (sdg) Importance: warning Could this have anything to do with it or is it anything to worry about ? SMART attributes Grrr.... apparently while i was out my server restarted again, however... i had a browser window open when i left, and when i came back it said " System going for a shutdown " or something along those lines at the top of the browser window - do you think that would indicate more of a UPS issue than a PSU issue ? Thanks
July 19, 201510 yr Community Expert No the flash should not cause a reboot. I've been getting these as well.... Event: unRAID Disk 5 SMART health [188] Subject: Warning [bEAFY] - command timeout is 1 Description: ST3000DM001-1CH166_Z1F1A012 (sdg) Importance: warning Could this have anything to do with it or is it anything to worry about ? SMART attributes Grrr.... apparently while i was out my server restarted again, however... i had a browser window open when i left, and when i came back it said " System going for a shutdown " or something along those lines at the top of the browser window - do you think that would indicate more of a UPS issue than a PSU issue ? Thanks That indicates that unRAID knew it was shutting down, as opposed to the power just being killed. I'm not entirely sure what all could cause that though. Did it start another parity check when it restarted? Do you have the powerdown plugin installed? I think it will write the syslog to your flash when it shuts down so we can look at it later if this happens again. Are you still using the UPS? If so, then depending on how you have the UPS settings in unRAID it could have signaled it to shutdown. Another possibility, perhaps unlikely, is someone else shutting it down. I think ctrl-alt-del on an attached keyboard might do it, and also pushing the power button briefly (unlike holding down the power button, which just kills the power). Do you have cats?
July 19, 201510 yr Author No the flash should not cause a reboot. I've been getting these as well.... Event: unRAID Disk 5 SMART health [188] Subject: Warning [bEAFY] - command timeout is 1 Description: ST3000DM001-1CH166_Z1F1A012 (sdg) Importance: warning Could this have anything to do with it or is it anything to worry about ? SMART attributes Grrr.... apparently while i was out my server restarted again, however... i had a browser window open when i left, and when i came back it said " System going for a shutdown " or something along those lines at the top of the browser window - do you think that would indicate more of a UPS issue than a PSU issue ? Thanks That indicates that unRAID knew it was shutting down, as opposed to the power just being killed. I'm not entirely sure what all could cause that though. Did it start another parity check when it restarted? Do you have the powerdown plugin installed? I think it will write the syslog to your flash when it shuts down so we can look at it later if this happens again. Are you still using the UPS? If so, then depending on how you have the UPS settings in unRAID it could have signaled it to shutdown. Another possibility, perhaps unlikely, is someone else shutting it down. I think ctrl-alt-del on an attached keyboard might do it, and also pushing the power button briefly (unlike holding down the power button, which just kills the power). Do you have cats? It did start another parity check, yes. The UPS is still being used at this time,yes. The UPS settings are set to shutdown the server at 10 min remaining on the battery. As far as I know, I have no powerdown plugin... just the default unraid one - is this something I should acquire ? There is no keyboard attached to the server currently, and I live alone ( no cats lol ) I'm guessing I should be removing the UPS and seeing how it goes from there..... As I mentioned before, this did not happen until I did the upgrade, not even once - so my gut tells me its software related ? or I suppose it could just be a coincidence. Thanks again !
July 19, 201510 yr Community Expert Well, UPS support is built-in to v6. Did you have UPS plugin installed when you were using v5? I think the powerdown plugin is worth having and it might help provide more information for your problem. See search tips in my sig if you have trouble finding it.
July 20, 201510 yr Author Well, UPS support is built-in to v6. Did you have UPS plugin installed when you were using v5? I think the powerdown plugin is worth having and it might help provide more information for your problem. See search tips in my sig if you have trouble finding it. I feel like I remember adding something along the lines of " clean powerdown " or something like that in my old build. I will have a look at that plugin. Thanks yet AGAIN !
July 20, 201510 yr Event: unRAID Disk 5 SMART health [188] Subject: Warning [bEAFY] - command timeout is 1 Description: ST3000DM001-1CH166_Z1F1A012 (sdg) Could this have anything to do with it or is it anything to worry about ? Ignore it, nothing to worry about.
July 20, 201510 yr Author Event: unRAID Disk 5 SMART health [188] Subject: Warning [bEAFY] - command timeout is 1 Description: ST3000DM001-1CH166_Z1F1A012 (sdg) Could this have anything to do with it or is it anything to worry about ? Ignore it, nothing to worry about. I thank you sir.
July 20, 201510 yr Community Expert Just thought I would add a recent UPS experience I had. During a thunderstorm recently we lost power while I was out of the house, and it stayed off for a while. Everything shut down like it was supposed to, and later that evening when I returned home I started it all back up. Not long after that there was a very brief brownout, and since my UPS had not had a chance to recharge enough, it told my systems to shutdown again, even though power didn't really go out again. Probably no matter what settings I chose there would still be edge cases like this. No harm done.
July 20, 201510 yr Author Just thought I would add a recent UPS experience I had. During a thunderstorm recently we lost power while I was out of the house, and it stayed off for a while. Everything shut down like it was supposed to, and later that evening when I returned home I started it all back up. Not long after that there was a very brief brownout, and since my UPS had not had a chance to recharge enough, it told my systems to shutdown again, even though power didn't really go out again. Probably no matter what settings I chose there would still be edge cases like this. No harm done. Makes sense, thanks ! I've had a few power outages but nothing substantial that has caused a shutdown. its been good since last night, i'm giving the UPS one last try before i leave my sever naked into the wall plug. I'm still confused on why it's only been happening since my upgrade to 6.0.1 but I may just chalk it up to coincidence :\ As always, thanks again !
July 21, 201510 yr Popple- How old is the UPS and/or the batteries? Not sure i saw that in the thread. I have been having somewhat similar issues. Quick brownouts that i don't think my UPS is handling all that well (when it used to just fine) which is leading to the server losing power. I have the apc software and the clean shutdown script and all worked perfectly up until maybe 4 months ago. Generally you will get 3-5 years out of the batteries on a decent UPS. My UPS was bought new about 3.5 years ago, so I am replacing the batteries and will see what happens.
July 21, 201510 yr Author Popple- How old is the UPS and/or the batteries? Not sure i saw that in the thread. I have been having somewhat similar issues. Quick brownouts that i don't think my UPS is handling all that well (when it used to just fine) which is leading to the server losing power. I have the apc software and the clean shutdown script and all worked perfectly up until maybe 4 months ago. Generally you will get 3-5 years out of the batteries on a decent UPS. My UPS was bought new about 3.5 years ago, so I am replacing the batteries and will see what happens. . To be completely honest, I dont remember... if I had to guess, I'd say about 2-3 years. I went to costco today to buy another, but they didnt have the one I was looking for. I would love to know if your new battery solves your issues ! Thanks for your input !
Archived
This topic is now archived and is closed to further replies.