ObviousViking

Every so often my largest unraid server locks up. I lose ssh, webui, unraid connect and pressing the power button doesnt initiate shutdown. However, ping works and plex can still access the media..for a time. The only way back is a hard reboot. Logs attached but even though syslog server was active before lockup - i dont seem to have muchsyslog-10.10.10.10.log - and any help is appreciated hercules-diagnostics-20250805-0033.zip

Turns out the disk dropping offline this time made my BIOS redo my boot order so it was trying to boot my appdata disk. Ive just made the amendment again and giving it a restart. Once up ill try transferring the files i was doing previously to see if the drop off occurs still

i i made this change, let the rebuild complete but after a reboot it now no longer boots. Im just getting a flashing cursor in the top left of the screen. I edited the syslinux config file on the usb by plugging into windows and copying the default that i have on my other unraid boxes but still nothing. Im still on with troubleshooting

Hi all I finally remembered to grab my diags before a reboot. I find at times my cache nvme drops into an offline state. It doesn't come back until i fully power down and reboot, a restart does not fix it. I find it mainly happens when data is hitting the drive in bulk. This time i was copying about 9GB of image files from my windows machine to the unraid so they were hitting the cache. The downside now is i was in the middle of a data rebuild so i cant power off without starting that again. Any help as to why it drops would be great. This has happened with the previous motherboard i had, also with my current mobo but my previous nvme hercules-diagnostics-20250219-2001.zip

So ive had a look around but other posts are either hard drives or sata ssds and usually its a power issue. So every so often my server will become buggy, shares disappear, cant mount remote shares, dockers arent working. When i check it appears my cache drive has fallen offline. I try to add it back in but its not detected, however i can see it in unassigned devices. A restart doesnt fix it, only a full shutdown and reboot does. This happened with an old nvme drive i had also before i upgraded to a newer, faster one. While the device is in unassigned devices the disk log shows As far as i know the other drives were mostly spun down so there was virtually no activity other than myself streaming a show from plex (plex on a separate server). i finally remembered this time to grab the diagnostics before rebooting so any help is appreciated hercules-diagnostics-20241112-2232.zip

ah noted, thanks ill get it put on

Hi all So ive been having a few lock ups recently on my main data server. Today, my cache had hit 90% but the hour hadn't arrived so i manually hit move. It then just started to lock up, no webui, no unraid conenct, no ssh no nothing. I even tried ping but got no response even though i could see network activity and my smart plug monitoring the energy suggests it was not idling. Ive recently moved a lot of my dockers off this box to other smaller unraid machines so the only thing on it now that uses anything resources is unmanic due to the GPU. I have two instances of that and both only had 1 worker active but from what i remember no active transcodes. At a bit of a loss as to whats causing this. Any help is appreciated, im getting tired of hard restarts. Attached are the diagnostics from after i hard restarted hercules-diagnostics-20240617-2255.zip

yes when all dockers are stopped it calms way down. hmmm i wonder if the nvme that came with it isnt as fast as the new one i put in. ill move everything to the new nvme and see if that does anything

Hi all, so this is my 4th unraid box ive spun up. Its a HP mini prodesk - 8500T, 8GB RAM and 2 nvme drives. Its purely to run the arrs - sab - qbit. My main unraid server gets the new files pushed to it via a mounted share. My CPU seems to be forever pinned but when i check my docker page and top it tells a different story but im thinking this is whats contributing to lock ups and sab just sitting there at 0B/s. Attached are the diagnostics and a snip of the top and dashboard athena-diagnostics-20240530-0910.zip

Ah ok, ive now swapped that manually to this speed, ill see how it goes

Yeah came across this. Ive now done the BIOS changes. I set power supply idle to current idle and disabled c states globally. Not sure if both are needed or one of the other but quick change back. As for RAM left it on auto rather than manually set it to 2667 which i think it suggests for 4 dimms in use on a Ryzen 5 3600. Is there anything else worth checking or logging? Im assuming id see less lock ups with less stuff running on the server?

So ive had this a couple of times, and it may just point back to a ryzen stability issue that i see dotted around everywhere. However, there are logs and im not an expert in looking at them so there may be another issue. So the server was just doing its thing about 25 dockers running just the arrs, a minecraft server (low resource) a Win111 VM (idling) etc. I set off unmanic doing a scan for new files as id just messed with some locally and uploaded them and about 5 minutes later it was dead. no web ui, no pings, no docker pages, minecraft server down etc etc. Thought id try initiate a graceful shutdown by tapping the power button the case, nothing just 2 drives in the case with solid activity lights. So, a hard power off it was. Attached are the logs from after the reboot. Other things maybe of note? CPU is set to power save using tips/tweaks and so is limited clock speed I have an NFS share to another unraid bow but that just handles plex and some other bits - no scan running there Any help would be appreciated hercules-diagnostics-20240401-0027.zip

Hi all not sure if this is a bug but its happened a few time lately so thought id capture it and post. randomly when i load the dashboard i get a spam of notifications some of which are days old and even when i go into the menu and acknowledge them they play out until done. as per the time in the screenshot i let it keep going and at 13:11 it was still loading them

so i seem to have fixed mine. i was using a nord proxy set up inside prowlarr. moved over to tunnelling prowlarr through the nordlynx docker and been fine since. i spoke to Nord and the rep told me their proxy servers dont get the attention they need so connections can be hit and and miss which seems to match with my experience. im looking at trying to get the hotio image working as it now comes with a built in vpn but i need to do the linux trick to get the wireguard config for nord

Did you find a fix for this? Im on virgin and just got the issue starting. My prowlarr has a socks5 proxy for nord enabled so virgin blocking stuff shouldnt be an issue