Cupid

I found the issue. You're right. Hardware related Fractal Design AIO Cooler Recall Thanks, everyone.

Would you suggest I install and run cinebench?

Perhaps. but I don't know what else to do. I used a different cooler as well and it was still the same result with all dockers closed. Should a plugin really be taking up that much CPU power?

I reapplied it twice. Everything seems back to normal now that I removed cache directory and net data. htop said those were taking up 140% of my cpu power. Has anyone else had something like this before?

It's the CPU overheating. Crashing because of it. Right now it seems to have been fixed after I removed that cache directory and netdata. If it happens again, I'll do that. Thank you.

It did it again after removing the directory cache plugin. syslog-192.168.0.103.log

I did some troubleshooting on my own and found when I removed my second HBA card every thing was fine temperature wise. Obviously I had some missing disks. I looked at the CPU usage with htop and it said SHFS was using 100%. I don't know what that is, but other posts talked about the dynamix directory cache plugin. I have disabled it and everything seems to be much better. Less usage and normal temperatures. syslog-192.168.0.103.log

Hello, helpful community. I have noticed the past few days my machine's CPU fan (AIO cooler) has been on high at times when I feel it shouldn't. The only demanding thing it runs is a Plex docker. It was even running high when only one stream was playing and also running on high when nothing was playing. It also had an unclean shutdown last night. I can't find it in the settings to save syslogs. Has that setting moved with the update? I can only get the syslogs after I restarted. What do you think could be the problem? Hardware or software? Thank you for all your support. tower-diagnostics-20240114-1157.zip

This seems to have fixed the problem. Thank you! Is there anything else I should do? I have attached my updated diagnostics. tower-diagnostics-20230725-0043.zip

I've added me diagnosis, but it's after restoring to my previous version. When I updated and when I restored, unraid wouldn't shut down my system by itself and I had to do a hard reset both times. Could that be a clue? tower-diagnostics-20230704-2127.zip

I recently updated UNraid to 6.12.2 from 6.11.5 and ever since it's been very slow in several areas. Mainly the GUI. Even logging in takes about 60 seconds for everything to load and up to 120-240 seconds to see my drives. Plex is no longer playing movies either. Accessing shares through the network has also been slowed down a lot. Should I restore to the old version? I've been trying to download the diagnosis for 10 minutes and it still isn't doing it. Has anyone else experienced this?

I just tested that. I unplugged the UPS and ran a few dockers to put a load on the CPU to run down the battery faster. I was logged into the gui and it said the UPS is below threshold and is shutting down. I now have it set to 65% batter and 35 minutes. 0 for the seconds on battery. I restarted it and there was no parity check. So that's good. Maybe I had the settings wrong? Could it be that the UPS/Unraid is being tricked during a power outage since the power usually flickers for a few seconds before actually cutting out?

I will try that.

Yes. Everything is fine if I just shut it down myself or restart it from the gui

Hello. What have I configured wrong? My power goes out about 4 times a month. It's horrible. I have an APC UPS model: BR1000G. Every time the power goes out, when I power things back on, a parity check is always started. Today when the power went out, I checked the server and it was still on because of the UPS. Then shut off like 10-15 minutes later. I assumed it did a proper shut down. I was wrong I guess. Edit: The "Time left on battery before shutdown" is set to 15 here, but in the past it's been disabled (set to 0) and the same thing occurs. Thank you. tower-diagnostics-20230401-1251.zip

Thank you so much for answering that! Peace of mind.

Hello, helpful community. I don't have any problems with my UnRaid system at the moment. I just have a couple questions about the parity process. Right now this is how I have my parity check set up. This will start the parity check at midnight on Tuesday of the second week of every other month and not stop until completed. If I use the cumulative parity check feature and set it up as follows, what will change? My understanding is it will start just like before at midnight on Tuesday of the second week of every other month. With these settings will it start at 12:00am and continue for 10 hours and stop at 10:00am and resume Wednesday at 12:00am and continue that way until finished? Is there any downside to this other than taking multiple days to finish? Could it hurt the integrity of the parity check? My other two questions are related to my fear of something bad always happening since I don't know a whole lot about computers. When a parity check is in process, is the process robust? If I write to my disk array during the parity check, is it possible for it to mess up the check? I assume the performance drop during the parity check is for a good reason, so it doesn't mess up anything. What about during a disk rebuild? (if I take out 8TB disk 4 and replace it with 12TB) Being emulated, it would be preferred to not write to the array, but what if you do? Any chance of messing up the rebuild or does UnRAID know how to handle it? It seems like so much work for the system to write and read from parity plus read from the remaining disks and write to a disk or even write to an emulated disk. Not sure if that's how it works. Right now when I rebuild a drive, I disable the mover and only allow Plex to run. During a parity check, I don't disable the mover. Should I? Thanks for all your support. Sorry for so many questions. I just don't want to stress the system if it's not necessary.

This a great example of how not to treat a community that provides free help.

I am in the United States. What made you think Canada? Thanks for all the information. I'll be looking at them today. I'm wanting a 16i card.

I just finished my parity check. zero errors. I guess that's good.

Thanks. I'll do that.

Thanks for helping. If it's just. SATA problem, I guess it could have been much worse. I have a budget lsi hba card from eBay. So who knows if it's actually legit. What is the next step right now? I have a spare card and cables. I could swap everything out. But it's also an eBay card. I would be willing to pay much more for a quality one. I'm just not sure where to look. Am I safe to clear those stats? Will any of this affect parity? Thanks for all the help on this by the way. I'm sorry I can't contribute more.

So I did a data rebuild Monday upgrading a 10TB old disk to a new 18TB. I haven't had any problems with it all week. This morning I got a UDMA error count of 143 and 4,162 errors on the main array page under disk 4. The notifications said they were read errors. Any thoughts on this? tower-diagnostics-20221126-0736.zip