craigr

ipmifan fan is also getting perpetually larger, but at a much slower rate.

I think it might have been the FIle Activity addon log. The log bar graph was getting higher as I sat here again. I disabled File Activity and deleted its log and now it seems normal again. Has something changed with unRAID 12 that may have made that plugin incompatible? Because I've had it set the same way for a very long time.

After going to 6.12.2 I'm back to this: I guess I will have to wait to see if anything explodes in size in /var/log because right now there is nothing abnormal and no error messages.

Well I just went from 6.12.1 to 6.12.2 and my server crashed again just like when I went from 6.12.0 to 6.12.1. Exact same thing: I now have a new IPMI event that matches: These two past upgrades are literally the only time this server has ever crashed.

I've noticed more memory being used and was not sure why. Still, I'm not so sure I want to revert to a much older patched Docker version. I just did a lot of work rebuilding and renaming the cache pool that holds my dockers and libvert when I converted it to ZFS. I'm not in the mood for something to go sideways after rebuilding my Dockers. I also intentionally nuked my libvert and started from scratch, though I suppose this update should not matter for VM's? I mean, is Docker going to go back to v23.x in unRAID 6.12.3? I'm always positive on upgrades, but this one has me concerned. That said, on 6.12.1 I do have this log issue that nobody answered about in the 6.12.1 thread. Will 6.12.2 address this? All that said, I guess I'm leaning towards upgrading to the downgraded docker. Until a couple weeks ago I only had 64GB of RAM and it was usually only using 60% or so unless I was doing intense things. Now I'm using 64GB at steady state without extra activities.

Why is my log at 100% and RED? Help please!

Sold the old RAM on eBay for $200. Minus shipping and fees, I should clear $160. I guess ~$150 to go from 64GB to 128GB isn't so bad. My server really needed the memory badly.

Naturally you were correct. It's back to the way it was again.

It's in a weird place IMHO:

You've seen this thread right: Read at least through my post on Feb 20th. EDIT: Never mind, it's in the PLEX menu now. How did you disable tone mapping? I do have tone mapping enabled. Obviously, you can't really watch HDR as SDR without tone mapping unless you like really awful looking video.

Same CPU as you, no issues here:

You could try stoping the array. Set Enable bonding to "no". Then see if you get any "Extra Interface" that you can assign to your eth0 interface.

Do you think there could be a driver issue for my hardware with the new unRAID / Linux distribution? I just don't understand what caused my crash and I really want to understand.

Well I ran memtest all last night with no errors. Rebooted unRAID just now without issues. One thing, AFTER I updated unRAID from 6.12.0 to 6.12.1, the bar at the top of the screen (don't know what you guys call it) notified me there was an update for Unassigned Devices. Without thinking I clicked update. Then, I rebooted unRAID for the 6.12.1 upgrade. Could this have caused the issue? It seems unRAID crashed on me the first time while it was loading apps after reboot.

I'm trying to figure out how to get the log file, but it's late I just got back from a business trip, and I'm exhausted.

Thoughts please: This time would have been about when I had to do a hard shutdown. Thanks

Just noticed this error in IPMI when I rebooted: I replaced all my RAM on Jun 13th and have had zero problems. I kinda think it's probably not the RAM because this seems too coincidental?

Also, big thanks for maintaining many of the apps that you do. Especially IPMI, but also many others that I use.

I just upgraded to 6.12.1 and suddenly (at least for now) the Power Draw graph is working perfectly. It goes up and down with respect to the amount of power being used and then back to zero when transcodes stop. Go figure? Does my IGPI support Blitter Load? This has never reported anything but zero. I just removed it from the display.

Just updated from 6.12.0. First reboot everything seemed fine. Discs seemed to mount, array seemed to start, and some apps (maybe all) loaded. I know this only because IPMI must have loaded; the fans spun down to proper speed. Then I sat and waited for my Windows 11 VM screen to come up, but it didn't. No pretty unRAID logo instead of OVMF BIOS screen, no web GUI on a different computer. Went back to the CLI on the IGPU of the unRAID machine and the cursor for login was not blinking. Could not get any response from the machine. Not even a BEEP when I pressed the power down button. No SSH. It was totally frozen. Had to do a hard reset. After that, rebooted and everything started perfectly so a parity check began. Seems fine now. Could not get logs because there was no way to interact with the machine, but fans were running using IPMI settings the whole time. All seems fine now.

Wait. Sorry, but won't that only persist until reboot?

I'll give it a go when I get back from my business trip Thursday. Thank you.

Spoke too soon. I tried a transcode and no matter how much power draw (0.1 - 10 w) the bar stayed at 100%. Now with zero draw it's at 100% again.

My power draw bar graph is suddenly working correctly. Nothing that I know of has changed. I have not even rebooted the server.

It looks like it's been updated according to the thread now. I didn't know about this script. I may give it a try. I hate that PLEX is constantly thrashing my SSD's. However, I did calculate it out by the increase in LBA's over a 24 hour period and it only comes out to about 50 GB a year. I don't know if the scrip really makes a difference for SSD life, at least in my applications.