Hi All

 

I have been forced to cut the power 3 times since I started using Unraid instead of primary Synology server.

I know this could be caused by me running the Version: 6.7.0-rc7 release

 

And every time its been in relation to a docker application that wouldn't stop (Plex) or UAD not finding mounted drives

 

I typically see a stopped on the UI like this:

 

But the Array is still mounted, and it's just keeps stating:

 

Logfiles: Can't get them when the server is in this "state"

 

I do have console:

root@TOWER:~# tail -f /var/log/syslog
Apr 21 14:23:21 TOWER emhttpd: shcmd (4745): umount /etc/libvirt
Apr 21 14:23:21 TOWER root: Stopping docker_load
Apr 21 14:23:21 TOWER emhttpd: shcmd (4747): /etc/rc.d/rc.docker stop
Apr 21 14:25:11 TOWER nginx: 2019/04/21 14:25:11 [error] 12777#12777: *1477951 upstream timed out (110: Connection timed out) while reading upstream, client: 192.168.0.176, server: , request: "POST /plugins/unassigned.devices/UnassignedDevices.php HTTP/2.0", upstream: "fastcgi://unix:/var/run/php5-fpm.sock:", host: "e7591da5b3c213bd7e17dc48d29fc9dcd1283498.unraid.net", referrer: "https://e7591da5b3c213bd7e17dc48d29fc9dcd1283498.unraid.net/Main"
Apr 21 14:26:12 TOWER nginx: 2019/04/21 14:26:12 [error] 12777#12777: *1477951 upstream timed out (110: Connection timed out) while reading upstream, client: 192.168.0.176, server: , request: "POST /update.htm HTTP/2.0", upstream: "http://unix:/var/run/emhttpd.socket:/update.htm", host: "e7591da5b3c213bd7e17dc48d29fc9dcd1283498.unraid.net", referrer: "https://e7591da5b3c213bd7e17dc48d29fc9dcd1283498.unraid.net/Main"
Apr 21 14:36:03 TOWER login[32238]: ROOT LOGIN  on '/dev/pts/0'
Apr 21 14:36:44 TOWER nginx: 2019/04/21 14:36:44 [error] 12777#12777: *1479468 upstream timed out (110: Connection timed out) while reading response header from upstream, client: 192.168.0.176, server: , request: "POST /webGui/include/DashboardApps.php HTTP/2.0", upstream: "fastcgi://unix:/var/run/php5-fpm.sock", host: "e7591da5b3c213bd7e17dc48d29fc9dcd1283498.unraid.net", referrer: "https://e7591da5b3c213bd7e17dc48d29fc9dcd1283498.unraid.net/Dashboard"
Apr 21 14:41:53 TOWER nginx: 2019/04/21 14:41:53 [error] 12777#12777: *1481156 upstream timed out (110: Connection timed out) while reading response header from upstream, client: 192.168.0.176, server: , request: "POST /webGui/include/DashboardApps.php HTTP/2.0", upstream: "fastcgi://unix:/var/run/php5-fpm.sock", host: "e7591da5b3c213bd7e17dc48d29fc9dcd1283498.unraid.net", referrer: "https://e7591da5b3c213bd7e17dc48d29fc9dcd1283498.unraid.net/Dashboard"
Apr 21 14:51:06 TOWER nginx: 2019/04/21 14:51:06 [error] 12777#12777: *1481156 upstream timed out (110: Connection timed out) while reading response header from upstream, client: 192.168.0.176, server: , request: "POST /webGui/include/Download.php HTTP/2.0", upstream: "fastcgi://unix:/var/run/php5-fpm.sock", host: "e7591da5b3c213bd7e17dc48d29fc9dcd1283498.unraid.net", referrer: "https://e7591da5b3c213bd7e17dc48d29fc9dcd1283498.unraid.net/Tools/Diagnostics"
Apr 21 14:54:15 TOWER login[8725]: ROOT LOGIN  on '/dev/pts/0'

Any input to what I could do when this happens next time?

 

Br

Casperse

 

 

 

Instead of grabbing the diagnostics via the UI, log in to the terminal (either via the UI or at the command prompt), and enter in

diagnostics

It *will* eventually get them, and will not have the timeouts associated by the UI's hard limit of 120 seconds for any request

Thanks I will try this next time this happens. Do you think this could be Plex/docker causing this?

I couldn't even stop the Plex docker before trying to stop the whole server

Hard to say.  For a couple of the 6.7-rc's I was having a ton of trouble with some apps refusing to stop (and intermittently they would crash).  Problem eventually went away.  Whether it was a RC causing this, or a hardware issue (I've subsequently upgraded my mb / cpu), or a bad update to one or more apps I can't 100% say.

My advice give this an install. 

 

 

When ever I reboot or shut down I normally always shutdown all my dockers first and have this installed. 

 

I normally use my PC to activate the shutdown. If my Array hangs I fire up that Plugin with another tab or on my phone and see if one of my drives is being connected too. Sometimes it’s a stuck MC (MidnightCommander) process and kill it. Of course it can be anything but I quickly check and walla Array shuts down and all is well.