sydsick

Precisely, I did just downgrade to 6.11.5, and rebuilt the docker.img, and that took me from about 90% downtime to 5% downtime. The issue isn't presenting itself currently, but if it does again I'll give your fix a try. Thanks!

I've noticed that if I simply restart the docker service, Internet comes back. Though I don't see any issues with any specific docker, and I've tried to go one by one enabling and disabling, and can't replicate or solve the issue in that way. The issue isn't presenting itself currently, but if it doesn't the future I will give your method a try, thanks!

Thanks for responding JorgeB, you a real one. This was actually one of the earlier posts that I made regarding a series of issues that I've had. Memtest came up clear, switching to ipvlan (at your suggestion in another post) stopped the crashing that I was getting later on (powered on but unresponsive). Eventually I ended up having to just downgrade back to 6.11.5 and recreate the docker.img. I am currently having issues where unRAID will periodically lose all internet connectivity (can't ping out, doctors have no internet, disconnected from connect, etc) until I restart the docker service (settings > docker > no, settings > docker > yes). If anyone ends up here from googling their own problem, apparently ipvlans can cause these sorts of problems when upgrading to a new version of unRAID, for me particularly they popped up when going from 6.11.5 to 6.12.0, allegedly for others they have come up when upgrading too far earlier releases. I am rapidly figuring out how to look at the logs and diagnostics myself, but I'm still a moron so any pointers would be greatly appreciated. It isn't happening at the moment, but I will post current diagnostics once I get back home.

I have been getting this daily since fixing other issues brought up by upgrading to 6.12.x seems resolve on its own after 10 - 60 minutes, or after a reboot. no dockers have internet access, can't see apps page, cant ping out, etc. If I use "unraid-api restart", I get a different error message, something about "cant connected to mothership", but the issue persists until it decides to go away. I am running pfsense and pfblockerng, but there is nothing in the logs there pertaining to anything from the unraid server being blocked, and firewall rules are set to pass the traffic. This setup has worked flawlessly with unraid for the last 8 months. At least as far as I can tell, I obviously am doing something wrong since I'm having an issue. syslog diagnostics after reboot.zip diagnostics before reboot.zip

I have been getting this daily since fixing other issues brought up by upgrading to 6.12.x seems resolve on its own after 10 - 60 minutes, no dockers have internet access, can't see apps page, cant ping out, etc. If I use "unraid-api restart", I get a different error message, something about "cant connected to mothership", but the issue persists until it decides to go away. I am running pfsense and pfblockerng but there is nothing in the logs there pertaining to anything from the unraid server being blocked, and firewall rules are set to pass the traffic. At least as far as I can tell, I obviously am doing something wrong since I'm having an issue. syslog towerofdoom-diagnostics-20230626-2116.zip

Almost 48 hours of uptime (with other whack a mole issues popping up), Thanks boss!

Thanks for assisting JorgeB. I did that and reinstalled dockers that broke. Any reason in particular that that might suddenly cause a problem? I know I didn't recently switch to macvlan. I'll update with status.

Lots of problems recently after being up for the better part of a year, memtest shows no errors, temps look good. The past couple days Unraid becomes unresponsive seemingly at random (to me). No connection to docker containers, web UI cant connect, files inaccessible, but the machine stays powered on. Was occurring of 6.12.0, so I downgraded to 6.11.5, lost all docker containers, and still locked up. upgraded to 6.12.1, got all the docker containers back, and its still locking up. The only issues I could really find with search pertained to Ryzen CPU's, but I have Intel. I am hoping one of y'all chad brains can help me track down the issue. attached syslog and last diagnostic syslog towerofdoom-diagnostics-20230621-2358.zip

memtest passed, no errors. I did find some other troubleshooting steps based on the clues you gave me now that the subreddit is open, thanks! Please let me know if there is anything else I should be aware of or try. I'll update once I know more.

No, at the moment I was using Brave. I'll run memtest and post back here. I am now getting "DOWNLOAD OF APPFEED FAILED." on the apps page, and lots of stuff not working because of that dns issue, can't change network settings until I reboot unraid, after reboot everything is fine it really feels like I have had a new problem every day since upgrading.

upgraded to 6.12.0 2 days ago? yesterday, while using krusader, all dockers disappeared: "docker failed to start". restarted docker, no dice, rebooted unraid, everything came back. maybe 15 minutes ago, while messing with sonarr it freezes, notice that all disks are unmountable (except parity). stopped the array, started in maintenance mode, try to run file system check: don't remember verbatim, said something like 'disk contains mountable and readable file system, fatal error could not start xfs check'. half panicked, half wtf, I rebooted unraid, and everything came back... obviously everything isn't fine, please help. attached diagnostics from after docker crashed, after I noticed the unmountable disks, and after the reboot where everything is back (named accordingly) towerofdoom-diagnostics-20230616-0238- after reboot all disks back.zip towerofdoom-diagnostics-20230616-0216 all disks unmountable.zip towerofdoom-diagnostics-20230615-1440 -docker failed.zip

Hello! While working in Krusader, VNC disconnected. Find that docker page has title error, attempt to restart docker, no dice. Rebooted Unraid, and everything came back. Everything appears fine, but I eat crayons and don't know how to get anything useful from the diagnostic files. I am hoping that the brilliant folks here could tell me if this is just a transitory event, or if any further troubleshooting is needed. Diagnostic attached, please let me know if you need anything else. Thanks in advance! towerofdoom-diagnostics-20230615-1440.zip

I'm guessing its a bad NIC since I can swap it with a single port mcx311a-xcat with the same cables/transceivers and it get a solid connection. Thanks!

I am at the limit of my understanding here, and I'm probably googling the wrong keywords. I was successfully able to get two mellanox Connectx-3's (mcx311a-xcat) to connect (windows 11 <-> Unraid) and transfer data at 10gb. This tells me the basic network setup should be correct. I swapped in a dual port Connectx-3 pro (MCX312B-XCCT) into the Unraid server, but I cant get it to work. I have seen one other Unraid user with this nic in their server, so I know it should be compatible, but idk what I'm doing wrong. I rebooted Unraid many times, reflashed the current firmware using the Mellanox CA plugin, swapped the nic to eth0 and back, but no dice. diagnostics-20221012-1341.zip