K7RAID

Appreciate this follow up. Was working on this container today myself and ran into the same thing; your fix resolved the headache for me too.

Well, it has been a few hours and I couldn't just leave well enough alone, so I kept poking at it. First thing I found was that I hadn't quite cleared up my logs issue; they were still filling up so I used a different argument found on another UNRAID thread, pci=noaer, which appears to have kept the issue at bay across multiple reboots and hours. Took another stab at introducing the UPS, just attached the cable first and let the daemon read the stats from the APC for a while. No issues in UNRAID. Rebooted. Still no issues after a half hour (the UPS woes wouldn't take more than a few minutes to appear originally). Finally shutdown, attached the NAS to the UPS (though I didn't use the Master outlet this time). It's still running fine. I don't know if that Master outlet has anything to do with anything, but while the UPS was the clear cause and solution to my issues earlier today, I still can't reconcile how that was the case given how well it is acting now. Most of the rest of my efforts this evening are surrounding docker container repairs; seems the hung NAS did more damage than I thought. The ARRs stack for example couldn't write to appdata any longer for their various .db files and a few services may just need to be rebuilt, but I'm making progress. Not sure why a mere battery addition saw fit to ruin my day and deployment, but I guess this thread amounts to one large shrug as I look to clean up the pieces. I hope whatever caused it to flare up doesn't crop up again when I ultimately move the hardware to its final deployment location...

I believe I pulled these at the height of my frustrations; hoping there something there noteworthy. Currently the thing is idling along fine, though I could pull newer logs if that would be more relevant. casttower-diagnostics-20250303-1404.zip

Hi everyone. Took a stab at building out a NAS by jumping straight into UNRAID and, for the most part, I've got a baseline set up I'm happy with. Still in my trial period, but about 10 days or so of promising running and experimentation. I have UNRAID 7.0.0 set up on a Terramaster F4-424 Pro with 4 8TB Reds, 2 1TB M2 and its standard 16GB RAM. Having migrated all my ARRs and data and other odds and ends over successfully and with a week or success behind me, I took what I thought would be the final step before sitting back and enjoying this thing and added an older UPS I threw a new battery in (APC BR700G) to complete stage one of this grand deployment. Seems almost immediately after adding the UPS, I had a whirlwind of UNRAID headaches. Most frustrating of them all was the fact that the native 16GB RAM installed in the Terramaster was almost entirely consumed at this point, leaving a mere 100-300 MB remaining. This made the UNRAID UI almost entirely unusable and from the looks of it on the outside, essentially locked things up once or twice. For reference, my 'happy' state for UNRAID has about 12GB free RAM remaining when all dockers (9-10), services and drives are settled in and running. It didn't make much sense to me that the mere UPS addition would have been so detrimental, but as it was the only variable I'd added (and after a few aggressive restarts resulted in the same nearly hung server), I shut off the NAS, unplugged it from the UPS, returning it back to a surge protector and removed the RJ50 to USB cable (USB side) from the NAS entirely. Next boot up I was back to normal RAM and CPU utilization (which also suffered at times during this UPS enabled timeframe). Currently, the UPS service is running, showing an obvious Lost Communication error, but the UPS hardware is removed from the NAS. I'm wondering if there's any rhyme or reason to this aggressive use of resources when attached; perhaps something with my mostly default UPS Settings in UNRAID are to blame (picture of what I used attached). Since the removal of the UPS, the only changes I've done are to repair the damage done to my Plex container (hadn't had backups running yet; fixed that) and upgraded UNRAID from 7.0.0 to 7.0.1. I did also 'correct' an issue with my logs volume filling up within minutes of boot (this preceded the UPS install but I hadn't paid much mind to it until I started running out of options); syslog was getting spammed with 'PCIe Bus Error: severity=Corrected, type=Physical Layer'; found an older UNRAID thread that mentioned the append initrd=/bzroot pci=nommconfto workaround which I suppose solved the logs being full. Does anything jump out to anyone here? Looking closer at the UPS settings, I wonder if changing the cable (simple, smart or ether) and UPS type (this one does have an APCsmart setting that stood out to me) settings might be warranted here. The connection to the UPS did report correctly when it was connected...to the detriment of the rest of the ecosystem.