kaiguy

Hi! For unrelated reasons, I decided I need to downgrade back to 6.10.3 after briefly trying out 6.11.0. The change to 6.11.0 incorporated the SSD trim plugin into the OS, so the plugin was autoremoved. After the downgrade, I attempted to reinstall the plugin (which appeared to install successfully), but there is no plugin displayed nor any setting for TRIM that can be found. I seem to recall years ago before the trim plugin, that I could setup a cron job to manually apply the trim command. Is this my best avenue for the time being? I have a few SSD cache pools, so I am unsure if this is a safe alternative. Any suggestions? Edit: Looks like I'm getting an error on the plugin install. plugin: installing: https://raw.githubusercontent.com/bergware/dynamix/master/unRAIDv6/dynamix.ssd.trim.plg plugin: downloading https://raw.githubusercontent.com/bergware/dynamix/master/unRAIDv6/dynamix.ssd.trim.plg plugin: downloading: https://raw.githubusercontent.com/bergware/dynamix/master/unRAIDv6/dynamix.ssd.trim.plg ... done plugin: downloading: https://raw.githubusercontent.com/bergware/dynamix/master/archive/dynamix.ssd.trim.txz ... failed (Generic error) plugin: https://raw.githubusercontent.com/bergware/dynamix/master/archive/dynamix.ssd.trim.txz download failure (Generic error) Updating Support Links Finished Installing. If the DONE button did not appear, then you will need to click the red X in the top right corner

Had to rollback to 6.10.3 due to the EDID block 0 issue spamming my syslog, and like another user I see that I cannot get the SSD trim plugin to reinstall (it said it installed via CA, but it does not display under plugins and there is nothing visible under settings). Any thoughts? Posting in this thread since it appears to be a byproduct of the downgrade from 6.11.0. Thanks!

Updated without issue. All seems to be working well, though I am seeing something I've never observed before. About every 20 seconds, one of my CPU cores (core 6 per the dashboard) maxes out for a bit. I was monitoring htop, sorting by CPU activity, but I can't seem to find a correlation. Anyone have a suggestion for how I can try to find out what is causing this intermittent core activity?

Your suspicion was absolutely correct. Reloading nginx properly showed a renewed cert. Thanks so much! Do you think this is something I should file a bug report for? Or could this have been an isolated incident? In any case, if I run into this again in the future, I'll know exactly what to do. Appreciate your help!

Hi there. Running 6.10.3. Searched for this error but I didn't seem to find anything in the forums. Today Firefox gave me a warning when accessing the unraid GUI that the ssl cert was invalid. I took a look and it expired 3 days ago. The Management Access settings also shows that the cert is expired, and the buttons for renew and delete are greyed out. This appears to be the relevant log items: Aug 12 08:09:38 titan nginx: 2022/08/12 08:09:38 [error] 12133#12133: OCSP response not successful (6: unauthorized) while requesting certificate status, responder: r3.o.lencr.org, peer: 184.28.78.21:80, certificate: "/boot/config/ssl/certs/certificate_bundle.pem" Aug 12 10:53:03 titan nginx: 2022/08/12 10:53:03 [error] 12133#12133: OCSP response not successful (6: unauthorized) while requesting certificate status, responder: r3.o.lencr.org, peer: 184.28.78.21:80, certificate: "/boot/config/ssl/certs/certificate_bundle.pem" Any suggestions for how to address?

Thanks for the update, @Hoopster. I've been on all the 6.10 rc's and have been experiencing the high mobo temp on just about every release (not the last few, but load has been low and the releases were coming pretty quickly), but no CPU_CATERR with turbo boost enabled for months now. I installed 6.10.0 stable last night and so far so good with no BIOS changes. I'm sure that will change, but I've sort of accepted that the high mobo temp issue is going to be an ongoing issue with this board.

Not storage related, but this community is great and full of resourceful and like-minded folks, so figured I'd check in here. I have been using the legacy free G Suite (Google Apps) for my personal email for over a decade. I don't use much else with it (Drive, Calendars, and Contacts, but those are easy enough to migrate), but losing free email with my personal domains is going to bite. I don't mind paying per se, but just wanting something commensurate with my usage (pretty much just email, but I would like more than 15G of storage). I see a few possible solutions, but all with some drawbacks: I am heavily invested in the Apple Ecosystem, so I can use my personal domain with iCloud email. I've been reading there are some quirks, so I'm probably going to use this as a last resort. Pay for Google. Probably not an option for me since they're burning so many people who have purchases associated with their free accounts, which forces them to pay or lose it (e.g., Google Play). Microsoft 365 or Exchange Online. Can't use the family plan since it would force me to move my domains to GoDaddy (no way) and no aliases, so I'm looking at something in the $8-12.50/mo range for business plans. I helped a friend's small business setup a free account when I did, and so they now have to figure out a small business solution as well. Thinking for them they should just do the hosted Exchange option... Is anyone else facing this? What are your plans for email moving forward? Thanks!

You motivated me @Hoopster to turn on Turbo Boost after my last restart. Uptime is approaching just 6 days, but no high mobo temps, no weird call traces. I haven't pushed my server much, but with typical daily use, everything is looking good here as well!

I removed the Intel-GPU-Top and GPU Statistics plugins as a test to see if that would solve my mobo temp issue like @Hoopster. 2.5 hours after reboot, with no activity on the server, the mobo temp jumped to its typical 88C reading. Running RC2. Not sure if I'm going to enable turbo boost as I wonder if there's a correlation between the high mobo temp, turbo boost, and lockups. But maybe I will...

That’s outstanding. What version of unraid are you running now, Hoop? The last stable or the current rc?

Same experience as you, though mine seems to end up in the higher-80s than lower... But everything seems to be working normally. Just a little annoying since I've had to tweak my unraid IPMI plugins to effectively ignore the mobo temps else I'll get spammed with high temp warnings. I think I might try re-enabling turbo boost this week, though I really don't want a crash and subsequent parity check.

@Hoopster is downgrading unraid a fix though? Haha. I am still running rc2 and mobo temps are low 80s like you experienced. Still haven’t re-enabled turbo boost yet… Have you?

It's not the root folder of the server, but the console of a container. Truth be told I don't even use those mappings, so I need to remove them. I believe they're from the original config years ago for that particular container.

I took a break from Firefox for a bit but decided to come back this week. Recently (since rc2, perhaps? Or maybe since the most recent MacOS Firefox 95.x release) I have the strangest font in the web terminal and docker console. It only seems to be happening with Firefox on mac (fine in Windows Firefox; fine in Edge; Safari I can't test since console doesn't seem to work at all in that browser), hence why I am not posting in the pre-release subforum. I know this probably isn't an unraid issue, but maybe someone else has run into this? Thanks!

So I've been on this new L2.35 BIOS for 5 days. I am still getting the erroneous high mobo temp report after a day or two of uptime. Haven't tried enabling turbo boost yet, but everything else seems to be in line with my experiences from the previous BIOS.

Unfortunately I just got around to flashing last night. Still a little early to tell. @nitrossub, do you have turbo boost enabled or disabled? I'd love to re-enable as long as it doesn't cause a CPU_CATRR or call traces.

Are you willing to share the link to the BIOS?I know I’d be interested in trying it.

Running L2.34 here. While I don't have regular lockups, there are times when I will catch some traces in the log and proactively reboot my server, but it's been pretty rare as of late. For the most part, as long as I have turbo boost disabled and ipvlan set for the Docker network, things run pretty stable. I wish I knew what combination of BIOS and Unraid versions/network configs are the root cause for instability--I do know there was a time when I was running with turbo boost for months without issue. I echo your frustrations that there's little consistency with these issues. Take the mobo temp issue for example--for whatever reason I haven't seen that erroneous reading all month despite seeing it constantly last month.

Oddly enough I woke up to all my docker containers shutdown and a bunch of call traces in my log (likely due to an appdata backup that crashed). Unraid wouldn't let me stop the array (it said the mover was active but it was not--super odd), so I initiated a shutdown and got hung along the way. html5 ipmi interface wouldn't actually let me login (just returned to login screen). Had to hard powerdown the server. Should have pulled the plug as well since ipmi still isn't working, but I'm back up and running. I did recently re-add a br0 container about a month ago, but it all seemed stable until this morning. So strange. Looks like I'm going into troubleshooting mode here as well.

@AlexHuang still running with Turbo disabled via Tips and Tweaks plugin. I do always get the mobo temp error so I just raised the threshold for notifications via the IPMI plugin (it's showing 87 degrees right now, which is false). I also switched the Docker custom network type to ipvlan but host access to custom networks disabled. No locks with these settings, including CPU_CATRR (which is absolutely due to turbo boost on my setup).

Here's the error I'm getting: 2021-10-21 10:54:37,876 - apscheduler.scheduler - INFO - Scheduler started 2021-10-21 10:54:37,878 - root - INFO - Setup Helper: Going though the setup helper. 2021-10-21 10:55:06,388 - app - ERROR - Exception on /posters4k [GET] Traceback (most recent call last): File "/usr/local/lib/python3.9/site-packages/flask/app.py", line 2070, in wsgi_app response = self.full_dispatch_request() File "/usr/local/lib/python3.9/site-packages/flask/app.py", line 1515, in full_dispatch_request rv = self.handle_user_exception(e) File "/usr/local/lib/python3.9/site-packages/flask/app.py", line 1513, in full_dispatch_request rv = self.dispatch_request() File "/usr/local/lib/python3.9/site-packages/flask/app.py", line 1499, in dispatch_request return self.ensure_sync(self.view_functions[rule.endpoint])(**req.view_args) File "/app/./app/routes.py", line 29, in run_posters4k return posters4k() File "/app/./app/scripts.py", line 349, in posters4k films = plex.library.section(config[0][3]) UnboundLocalError: local variable 'plex' referenced before assignment

I've been running my mobo with turbo disabled since March. Haven't had a single CPU_CATRR or mobo temperature warning since (which I was getting pretty frequently in the 6.9.x branch). Upgraded to 6.10.0-rc1 this morning and re-enabled turbo. Within an hour I got the mobo temp warning from IPMI for hitting 84 degrees. No CPU_CATRR yet... Rebooted and the temp warning happened not too long after. Are people still consistently getting this erroneous motherboard temperature reading?

Unfortunately it looks like the macvlan trace happened for me immediately after I enabled host access to custom networks. I posted diagnostics in the appropriate bug report thread.