Coindance

Just an update, Looks like the server just stayed on all night after Disabling global C states in bios and moved the flash drive to a usb 3.0 port, no crashing as of yet with no parity check running. I'll leave it on for the rest of the week, updated to latest unraid build after it was stable for an hour and has been stable for 4+ hours since. Going to give it a few days just to make sure. Thanks again for the help too!

So this time when the server went unresponsive it actually gave some errors on the attached monitor Looks like the usb drive might be failing? in which case I do have a spare I can try to transfer this one to. Added syslog just in case too. syslog_12_3_23

Thanks JorgeB I set C states to auto and set idle control to typical idle, which I have done in the past but this time I also set my ram speed to 2400 Just booted up the server going to see if it will stay running without a parity check and will update later. Will disable C state next if that doesn't work but I'll post updates here. Thanks again for the help!

Hi all, I've had my Unraid server for a few years now, originally had it running on my old desktop build but after a year or so I started getting many system freezes and just complete loss of power to the server when it does crash. I ended up getting a used build from a friend + some new parts (new liquid cooler incase overheating was the issue and new power supply incase it was causing the sudden loss of power to the server). With the new hardware the server no longer loses power completely but it still does go unresponsive with or without docker apps running. (server will stay up and work normally if a parity check is running, when it's done it's normally less than an hour before the server stops responding, no video output on the server side and web client won't connect at all. Single pressing power button does nothing (left for 10 mins after single press) have to hold down power button to turn off server and reboot. Current specs are: (Unraid 6.12.4 2023-08-31) Seasonic GX-1000 PSU B350 TOMAHAWK (MS-7A34) AMD Ryzen 7 1700X Eight-Core @ 3400 MHz G.Skill RipJaws V Series 64GB (2 x 32GB) 288-Pin SDRAM DDR4 3600 2x 16TB WD drives (1 parity, 1 data) 1x 14TB WD drive (Data) 1x 500gb NVME M.2 (Cache) XFX DD R9 280X Black Edition GPU CyberPower CP1500PFCLCD PFC Sinewave UPS System, 1500VA/1000W Previous build was only using the same drives and flash drive for boot (haven't tried a new flash drive yet either now that I think about it) but parity checks never return any errors and system goes unresponsive in maintenance mode too. I enabled mirroring syslog to flash and will attach the latest syslog along with diagnostics. I wasn't able to get a syslog server working on my desktop so I can give that another shot if needed. At this point I bought an 18TB external drive and backed up all my data (took several days because the sever kept crashing during transfers) using Krusader and unassigned devices. If I can't get this working before next year the plan is to wipe the drives and recreate the server from scratch. Would just prefer to have someone confirm I'm stupid and missed something obvious so wanted to check here first. I've been lurking the forums for pretty much the whole time this has been happening, tried changing cpu power idle options in BIOS, disabling global C-states just to try out, but the server has the same issue no matter what I set (tried all 3 of the idle options in bios with no change and currently set bios back to defaults). The only docker apps I have currently are the official Plex media server app and Krusader. Doesn't matter if docker apps are running or not, system will go unresponsive regardless. If a parity check is running the server won't go unresponsive, can use all docker apps normally but ~1 hour after a parity check completes the server is down again. Any help with this issue would be awesome, sorry I didn't have more detailed error info but I will try and get whatever we need to resolve this. Attached 2 syslogs, I think the smaller sized one is from a crash in maintenance mode and the other is a crash in normal mode. Normally has a 1-2 hour uptime when not running a parity check. Thanks in advance for the help too! Edit: New syslog from when I started up the server while making this post, Docker service was disabled when I started for some reason but enabled it and plex was working normally, about an hour later server went unresponsive, new syslog attached, leaving server off for now. gibson-diagnostics-20231202-1159.zip syslog_11_28_23 syslog_11_30_23 syslog_12_2_23