syslog went crazy and appears to have corrupted my install can i recover "easily"?

[perfessor101]

Hello,

my syslog went unresponsive a couple days ago it looks like.

I was using CA Backup set for weekly backup retaining them for thirty days (Appdata and USB).

 

I rebooted this afternoon and the array wouldn't autostart ...

checked the interface and found a few things off but started the array

then I found it had wiped a bunch of settings and they went back to defaults ...

 

I have a single drive in my btrfs cache pool that got kicked out of the array

I guess I want to know how to safely add it back in?

do I have to mount it and copy everything off it to another drive before adding it back into the array as a btrfs cache drive?

 

I would like to restore the appdata and USB backup ... but no cache drive currently.

It looks like it was only retaining the latest unraid USB backup which looks corrupted also,

but I won't know for sure until I attempt to restore it.

 

All my other unraid backups of this server appear to have disappeared ... except for some really really old ones.

 

thanks for any time and help,

Bobby

 

ps - 6.8.3 Stable

[JorgeB]

Sounds like flash drive problems, you should post the diags.

 

2 hours ago, perfessor101 said:

I have a single drive in my btrfs cache pool that got kicked out of the array

I guess I want to know how to safely add it back in?

You just need to re-assign it.

[perfessor101]

I cleared a few TB of space moving things around and made a copy of the drive before assigning it, but when trying to update sabnzbd i get errors like this

 

Configuration not found. Was this container created using this plugin?



Warning: DOMDocument::load(): Document is empty in /boot/config/plugins/dockerMan/templates-user/my-binhex-emby.xml, line: 1 in /usr/local/emhttp/plugins/dynamix.docker.manager/include/DockerClient.php on line 251

Warning: DOMDocument::load(): Specification mandates value for attribute D in /boot/config/plugins/dockerMan/templates-user/my-sabnzbd.xml, line: 72 in /usr/local/emhttp/plugins/dynamix.docker.manager/include/DockerClient.php on line 251

Warning: DOMDocument::load(): attributes construct error in /boot/config/plugins/dockerMan/templates-user/my-sabnzbd.xml, line: 72 in /usr/local/emhttp/plugins/dynamix.docker.manager/include/DockerClient.php on line 251

Warning: DOMDocument::load(): Couldn't find end of Start Tag Config line 72 in /boot/config/plugins/dockerMan/templates-user/my-sabnzbd.xml, line: 72 in /usr/local/emhttp/plugins/dynamix.docker.manager/include/DockerClient.php on line 251

Warning: DOMDocument::load(): EndTag: '</' not found in /boot/config/plugins/dockerMan/templates-user/my-sabnzbd.xml, line: 72 in /usr/local/emhttp/plugins/dynamix.docker.manager/include/DockerClient.php on line 251

Would the 'working' backups from CA Backup possibly help with this?

 

I have multiple docker directories as they got moved once or twice by limetech from when they were first added

 

if I go into CA Apps and 'reinstall' these dockers  would that work without damaging all the docker data that is already there?

 

[Squid]

This means that the template stored on the flash drive is corrupted.  

 

Easiest solution is to make a note of all the paths, ports etc from the docker tab for emby and sabnzbd (use advanced view, and make sure to show everything), and then reinstall them (but not from the previous apps section) and make all the necessary adjustments.

 

Or, if you have a backup of the flash drive, then restore /config/plugins/dockerMan/templates-user from the backup

[perfessor101]

thanks I'll try the backups from three weeks ago I'm not sure but it seems like a few dockers are giving similar errors like you've listed emby and sabnzbd

 

Warning: DOMDocument::load(): Document is empty in /boot/config/plugins/dockerMan/templates-user/my-binhex-emby.xml, line: 1 in /usr/local/emhttp/plugins/dynamix.docker.manager/include/DockerClient.php on line 251

Pulling image: gogs/gogs:latest

IMAGE ID [1549052435]: Pulling from gogs/gogs.
IMAGE ID [cbdbe7a5bc2a]: Already exists.
IMAGE ID [28c462924ad8]: Already exists.
IMAGE ID [b0f38349e8a8]: Already exists.

 

Rambling about why I made the mistake of mirroring the syslog to the flash (set it to rotate in a folder on the array, but never saw any files there)

 

Not Important information follows

 

I was getting random crashes every 12-24 hours when I tried the first beta after 6.8.3 ... reported it via feedback.

I tried Syslog Watcher 4 to remotely monitor the syslogs of both servers but nothing was coming up in the syslogs on a crash via remote monitoring.
I tried adjusting the syslog settings to save them to a share on the array and rotate four 100 MB syslog files ... nothing ever appeared in that share.

I set the syslog settings to mirror the syslog to my flash drive ... nothing appeared in the logs to suggest a reason for a crash.

While the server was crashed (non-responsive?) all ports for various services were open when scanned remotely and the server responded to pings, but I couldn't access anything (UI or files) ... so the monitoring software would say the server was up and I wouldn't know that it was non-responsive to file or UI access.

I put both servers back to 6.8.3

I left the syslog mirrored to the flash drive expecting another crash, but something went wrong with that and it filled the flash drive and anything that needed writing out before a reboot ... it looks like got damaged.