AceRimmer Posted March 7, 2021 Share Posted March 7, 2021 (edited) Every time i reboot the machine it starts a parity check - Version: 6.9.0-rc2 I'm going to the dashboard and clicking the "reboot the system" icon. I'm not going to Main / Array Operation / Stop before rebooting... should i be doing this? I don't remember having to do this when i ran the stable release. So i am at a dilemma of do i allow it to parity check and tax my drives (the fact a parity check was triggered means an error flagged somewhere i guess) but the parity checks always come back without any errors. Any advice would be greatly appreciated. Edited March 8, 2021 by CatMilk marking title as solved Quote Link to comment
JorgeB Posted March 7, 2021 Share Posted March 7, 2021 This means an unclean shutdown, when this happens Unraid saves the diags in the flash drive, those could give a clue on what's causing it. Quote Link to comment
AceRimmer Posted March 7, 2021 Author Share Posted March 7, 2021 (edited) Not really sure where to start with this. I've uploaded the latest diag zip from the flash drive if anyone is feeling up to it. [UPDATE] Diagnostics Zip attachment deleted Edited March 8, 2021 by CatMilk removed zip attachment Quote Link to comment
JorgeB Posted March 8, 2021 Share Posted March 8, 2021 Timeout appears to be kicking sooner than it should, go to Settings - Disk Settings and change the shutdown timeout to a different value just to re-apply, set it to like 60 secs. 3 Quote Link to comment
AceRimmer Posted March 8, 2021 Author Share Posted March 8, 2021 4 hours ago, JorgeB said: Timeout appears to be kicking sooner than it should, go to Settings - Disk Settings and change the shutdown timeout to a different value just to re-apply, set it to like 60 secs. Thank you for your help. The issue was resolved by doing as you suggested. Much appreciated 1 Quote Link to comment
Garbonzo Posted January 23, 2022 Share Posted January 23, 2022 I realize this is from 6.9x and I am running 6.10 rc2 but I am running up against it doing a parity check after what I believe to be a graceful shutdown/reboot operation from the dashboard panel (or optionally through the dynamix system buttons add-on). The default (I didn't change) timeout was already set to 90. So i am not sure if I need to make it shorter or longer.. Anyway, I will research it more if it keeps happening, I am just trying to understand the order of operations of a reboot and I am very new to unraid. not sure if I should attach diagnostics, as I see the one about was removed... so any insight is appreciated... tia -G Quote Link to comment
itimpi Posted January 23, 2022 Share Posted January 23, 2022 It could be worth hitting the Stop option and time how long it takes for the array to be successfully stopped. The timeout needs to be at least as long as that plus a safety margin. at that point try a reboot. If you still get an unclean shutdown detected, then this might indicate an issue with updating the flash drive to say the array was successfully stopped before the reboot. 1 Quote Link to comment
Squid Posted January 24, 2022 Share Posted January 24, 2022 6.10-rc2 (and 6.9) also has a known issue where the system is a little too aggressive in what it terms an unclean shutdown. Those versions of the OS if any process had to be killed off to accomplish a shutdown the flag was set to indicate "unclean". Proper operation is that regardless of tasks having to be killed if the drives are successfully unmounted (essentially), then it is "clean" Fixed in the next rev. Also, the one thing that we all know is that half the time that Windows does a shutdown is that it also decides to install updates. Those updates can take significantly longer than any reasonable timeout you set. The solution on Windows VMs is to install the virtio guest tools (it's on the virtio iso image you've mounted) and set in VM Settings to hibernate instead of doing a shutdown. The result is that a Win VM will consistently shutdown within a couple of seconds without you having to worry if an update that takes half an hour to install is present -> or even if an app on Windows has a dialog box open (or "Save Changes" pops up) that effectively blocks any shutdown attempt 1 Quote Link to comment
Azriel Posted August 5, 2022 Share Posted August 5, 2022 I have a question on this. Something that has been a huge pain for a while now. Sometimes I have power outages. Every time the server is rebooted after, what seems to happen is that an automatic parity check starts running, and when I check my dockers tab none of my dockers look like they are able to start running. The server will eventually lock up and stop functioning all together until I power it down. Then reboot it, and I have to manually cancel the automatic parity check so my dockers will start and then everything seems to work fine. What is this??? it's been super annoying. Can I disable automatic parity check all together? Quote Link to comment
JonathanM Posted August 5, 2022 Share Posted August 5, 2022 13 minutes ago, Azriel said: Can I disable automatic parity check all together? Only unclean shutdowns trigger checks, so avoid unintentional shutdowns by using a properly configured UPS and verifying that the server can successfully shutdown unattended. If you can't get that done, then be sure to set the array NOT to auto start, so you can better control the bootup sequence after an event. When you shut down normally, stop the array before you power down, that way if something is hanging up you can investigate the cause and get the array properly stopped so you won't get an unclean shutdown. Quote Link to comment
trurl Posted August 5, 2022 Share Posted August 5, 2022 And you should go ahead and run a noncorrecting parity check since you didn't let it complete. Unclean shutdown can result in a (relatively) few sync errors, and exactly zero is the only acceptable result. Quote Link to comment
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.