mike20865

Hi, My system seems to have randomly completely locked up in the middle of last night. I woke up this morning to find it completely unresponsive, to the point that caps lock on a directly connected keyboard didn't even work. The system was still "on", with all drives and fans spinning but as I said completely unresponsive (even showed as offline in my network). I ended up having to do a hard restart as I could not interact with it at all. There were seemingly no warning signs before I went to bed, and I don't think it was doing anything particularly intensive, nor had I installed any new dockers/plugins or made any major setting changes immediately preceding this. I am on 6.12.8 but I upgraded around when it released so it had been running fine for at least a few weeks since then. I did not enable syslog saving before this (I have done so since) so I don't have the log from before the crash. Just posting for any insight anyone might have and to see if anyone else has had any similar issues recently.

Only dockers I have installed currently are debain bookworm (running) and the binhex plexpass (stopped). I have the app data backup, community applications, dynamix system temp, Intel GPU TOP, NVTOP, Unassigned devices normal and preclear plugins installed. The issue appears to predate all of the things mentioned except Unassigned devices. The issue is that my write speed seems to fall over time to about half what it should be. The first time I noticed it was when I was first making the array and the parity sync was running. At first it was writing at the full 250MB/s that my drives support, but over time it fell to about half that. The average reported speed after the parity sync completed was 202MB/s. The earlier diagnostics file was from a point in time toward the end of the parity sync when the speed was down to about 150 or so. I am currently transferring the data from a couple of old drives using the debian docker. I am doing it this way since they are formatted without a partition so unassigned devices can't mount them. I have about 20TB of data to transfer so you can see why I want as much write speed as possible. When I started the copy it was again writing at about 250MB/s, now it is down to about 120 after a couple hours. I don't think this is a file quantity issue as these are all >1GB movie files. I already set the array to reconstruct write so I don't think that's the bottleneck. Furthermore, my CPU is nowhere near maxed out as none of the cores are going above 60% or so, and even then for brief moments. It is not a temp issue either as my hottest drive is 34C. The later diagnostics file is from during this ongoing transfer. I am also aware I was getting the nginx memory bug, but restarting nginx fixed it for now yet did nothing for transfer speed. Any help or ideas as to how I can improve my write speed would be greatly appreciated. tower-diagnostics-20231026-1635.zip tower-diagnostics-20231026-1134.zip