July 9, 20232 yr I´m having a hard time finding the culprit of this issue atm. My Unraid box will reboot unexpectedly/sporadically. It has done this at least 4 times over the course of the last 4 weeks. Diagnostics attached. Would appreciate some help to drill down on this issue. skynet-diagnostics-20230709-1724.zip Edited July 9, 20232 yr by Tetsuo
July 9, 20232 yr Community Expert Install the Syslog Server. Upload the resulting file as an attachment in a new post in this thread after you have had a reboot.
July 10, 20232 yr Author Already had that running fortunately. The last unexpected reboot occured some time between the 8th and 9th of July. I noticed the sd 10:0:2:0: attempting task abort!scmd errors that seem to be related to some issue with the HDD Temp Docker (according to another support thread). Things i´ve changed since the last unexpected reboot: * disable the HDD Temp Docker * disable spin down (for now) syslog-10.10.1.10.log
July 10, 20232 yr Community Expert 3 hours ago, Tetsuo said: The last unexpected reboot occured some time between the 8th and 9th of July. It looks to me that it happened at this point: Jul 9 00:00:30 Skynet emhttpd: spinning down /dev/sdc Jul 9 17:22:43 Skynet cache_dirs: Arguments=-i Backups -i Data -i Downloads -i Media -i RAW\ Vault -i Transfer -i nextcloud -l off Jul 9 17:22:43 Skynet cache_dirs: Max Scan Secs=10, Min Scan Secs=1 Jul 9 17:22:43 Skynet cache_dirs: Scan Type=adaptive Jul 9 17:22:43 Skynet cache_dirs: Min Scan Depth=4 Jul 9 17:22:43 Skynet cache_dirs: Max Scan Depth=none Jul 9 17:22:43 Skynet cache_dirs: Use Command='find -noleaf' Jul 9 17:22:43 Skynet cache_dirs: ---------- Caching Directories --------------- Jul 9 17:22:43 Skynet cache_dirs: Backups Jul 9 17:22:43 Skynet cache_dirs: Data Jul 9 17:22:43 Skynet cache_dirs: Downloads Jul 9 17:22:43 Skynet cache_dirs: Media Jul 9 17:22:43 Skynet cache_dirs: RAW Vault Jul 9 17:22:43 Skynet cache_dirs: Transfer Jul 9 17:22:43 Skynet cache_dirs: nextcloud I am no Guru at figuring out what was going on on your server prior to Jul 9 at 17:22:43 but that syslog does not look normal. Let me ping @JorgeB as he can often help in these situations.
July 10, 20232 yr Community Expert 19 hours ago, Tetsuo said: will reboot unexpectedly/sporadically. This is usually hardware related, and nothing relevant being logged points to the same.
July 10, 20232 yr Author So this essentially could be anything really? Probably a Memtest would be the first course of action then.
July 10, 20232 yr Community Expert 30 minutes ago, Tetsuo said: Probably a Memtest would be the first course of action then. It is non-ECC memory so you can use the supplied Unraid memory tester. Look in the boot menu. Run for 24 hours. Zero errors is the only acceptable result. Make sure that the case, fans and heat sinks are reasonably clean of dust and air intakes and exhausts are not blocked. (EDIT: That 17:00 reboot time would make me suspicious of a heat related issue...) Power Supplies have been involved in many of these shutdown/reboot problems in the past. Edited July 10, 20232 yr by Frank1940
July 10, 20232 yr Author Im confident that temperature is not an issue as i cleaned everything when i added 2 more disks just a few weeks ago. CPU heatsink is a beefy Noctua NH-U12S and disk fans are Noctua PPC 140s in a Fractal R5. CPU and disk temps are always in the mid 30s. Maybe low 40s when it´s boiling hot in the summer. Memtest it is i guess.
July 10, 20232 yr Community Expert 6 minutes ago, Tetsuo said: Im confident that temperature is not an issue as i cleaned everything when i added 2 more disks just a few weeks ago. This would to a pointer to me to suspect the PS even more. Modern PS are not just a power brick as they once were. They are fairly complex electronic devices and have very fast acting protection circuits to protect themselves when the rating of any of the various voltage busses are exceeded. The +12V buss(es) are often involved in this situation as server use is not the intended application that most of them are designed for. (Ideally, one should look for a PS with a single rail +12V buss for use in Unraid servers.)
July 12, 20232 yr Author Will Change the powersupply tomorrow. Just ordered a new one for my main rig (that i wanted to upgrade anyway) and will transfer the "old" one over to the Unraid server. Will report back if that helped.
July 17, 20232 yr Author FWIW i removed the Nvidia Quadro P600 that was still in, but which i wasn´t using for anything anymore, and didn´t have an unexpected reboot since. Either the Card went south or this was some unexpected behavior after the Update to 6.12.2 as i never had any issues on 6.11. Will still change the power supply for a bigger more efficient one.
July 17, 20232 yr Community Expert 4 hours ago, Tetsuo said: Nvidia Quadro P600 A Quick Google showed this card has a max power requirement of 40W. Removing this card (and its load) could have stabilized the PS...
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.