markiii Posted January 8, 2022 Share Posted January 8, 2022 One of my drives started having the odd read error last week. As its one of my older 4tb drives I shrugged and ordered a replacement drive. Shortly after Unraid disabled the drive. No problem, iIhave dual parity and a replacement is on its way. before the arrival of my new drive another drive became disabled. My replacement drive is now here and I'm trying to rebuild the first of the disabled drive onto this new 8tb disk. However the rebuild gets to somewhere in the high 40gigs of rebuild and consistently hangs. Plex and LMS non responsive, Unraid web interface unresponsive and neither will the server respond to a simple ping. A hard power off and restart brings the box back into the same state, array comes up and the rebuild begins again. hits 40 something gig into the rebuild and repeat. Unraid Version is 6.8.2 , i was planning to upgrade this week but issues got in the way Not sure if related but the week previously I had an issue with the logs hitting 100% utilisation and the CPU display on the summary page not working. Since the server hadn't been rebooted in over a year, I rebooted and both issues disappeared. thoughts appreciated as I'm running on borrowed time with no protection at the moment Quote Link to comment
trurl Posted January 8, 2022 Share Posted January 8, 2022 attach diagnostics to your NEXT post in this thread. Also setup syslog server Quote Link to comment
markiii Posted January 8, 2022 Author Share Posted January 8, 2022 here you go deepspace9-diagnostics-20220108-1827.zip Quote Link to comment
markiii Posted January 8, 2022 Author Share Posted January 8, 2022 syslog server enabled Quote Link to comment
trurl Posted January 8, 2022 Share Posted January 8, 2022 You have a lot of (very full) disks. Are you sure you don't have a power problem? Why are you running such an old version of Unraid? Quote Link to comment
markiii Posted January 8, 2022 Author Share Posted January 8, 2022 version of unraid is simply i haven't got around to upgrading for a while you are right i have a lot of rather full disks, is that likely to be a contributing issue? what makes you ask ref power? the number of disks? Quote Link to comment
trurl Posted January 8, 2022 Share Posted January 8, 2022 3 minutes ago, markiii said: what makes you ask ref power? the number of disks? yes Quote Link to comment
markiii Posted January 8, 2022 Author Share Posted January 8, 2022 i guess its possible, they wouldn't usually be all up and hammered as hard as when doing a rebuild i could reduce the disks as the new one is a larger capacity anyway, the issue is I'd need to fix the rebuild issue beforehand hmm Quote Link to comment
Squid Posted January 8, 2022 Share Posted January 8, 2022 Yeah, you'd need a PS that can handle ~70A on the 12V rail at minimum (ignore the total wattage of the powersupply as it's an irrelevant rating, and excluding any power hungry video cards etc ) and add in the considerations that as they age, Power Supplies cannot handle the rated power (on average they lose ~10% a year) Thanks @trurl I didn't think of that.... Additionally if it's possible, avoid using the Marvel controllers. Marvel is one of those companies that since they realize that they can't compete don't even bother with keeping their drivers functional, and random drop offs are one of the nagging issues that they have. Disabling VT-d in the BIOS (if you're not running a VM that requires passthrough) may help Quote Link to comment
markiii Posted January 10, 2022 Author Share Posted January 10, 2022 thanks for the help so far chaps. I've rigged up a second PSU to power 6 of the disks. However its not made any difference. rebuild still fails at the high 40gigs of progress. Quote Link to comment
trurl Posted January 10, 2022 Share Posted January 10, 2022 On 1/8/2022 at 1:31 PM, markiii said: syslog server enabled Post Quote Link to comment
markiii Posted January 10, 2022 Author Share Posted January 10, 2022 there you go Quote Link to comment
trurl Posted January 11, 2022 Share Posted January 11, 2022 Reboot is between these 2 entries? Jan 10 14:23:02 DeepSpace9 root: error: /plugins/unassigned.devices/UnassignedDevices.php: wrong csrf_token Jan 10 14:32:08 DeepSpace9 kernel: microcode: microcode updated early to revision 0x27, date = 2019-02-26 That wrong csrf_token is cluttering your syslog, not the reason for hang, but easy to fix so you get a cleaner log. Quote Link to comment
trurl Posted January 11, 2022 Share Posted January 11, 2022 On 1/8/2022 at 4:50 PM, Squid said: if it's possible, avoid using the Marvel controllers. Marvel is one of those companies that since they realize that they can't compete don't even bother with keeping their drivers functional, and random drop offs are one of the nagging issues that they have. Disabling VT-d in the BIOS (if you're not running a VM that requires passthrough) may help Quote Link to comment
markiii Posted January 12, 2022 Author Share Posted January 12, 2022 (edited) having read the wrong csrf token errors above in the log, i decided to remove my plugins and see if that helped. all disks then rebuilt as per normal operation. Everything's now back up. I've updated to 6.9.2 and reinstalled my plugins. so far all seems good. thanks for the help chaps Edited January 12, 2022 by markiii 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.