I started to have a strange issue today. Every share I try to open from my Windows 10 computer on my unraid server is taking minutes to open. Once open the folders inside open fast and without issue. If I back out of the share, then try to open it again, it takes minutes again. This just started happening today. Performance on my Windows computer shows low cpu, memory, disk, and network utilization, and everything else works great from the computer. Opening SMB shares on my other NAS works quick and without issue, so this seems to be an unraid issue, but I'm not sure what could be causing it.

Try running Dynamix Cache Dirs

6 minutes ago, Squid said:

Try running Dynamix Cache Dirs

What is that and how do I run that?

Its a plugin.  Install it and enable it and see if it makes any difference.  If all of your drives are spun down it should improve the responsiveness

3 minutes ago, Squid said:

Its a plugin.  Install it and enable it and see if it makes any difference.  If all of your drives are spun down it should improve the responsiveness

I have my drive spindown set to never.

1 hour ago, Squid said:

Its a plugin.  Install it and enable it and see if it makes any difference.  If all of your drives are spun down it should improve the responsiveness

I have something randomly pegging one of my cores in my cpu. I've run a top and ps aux on the console and I can't see anything running "hot" on my system. I even rebooted the whole server and it's still acting the same with the core pegging and the SMB share taking forever to load.

Edited September 12, 20214 yr by Squirreljester

And htop view doesn't show the 100% core like what's on the unraid gui. Makes no sense to me what's going on or why it's happening all of a sudden.

In the share setting (has to be set for each share) change "Case-sensitive names" to Yes.

43 minutes ago, SelfSD said:

In the share setting (has to be set for each share) change "Case-sensitive names" to Yes.

Done, same behavior.

4 hours ago, Squid said:

Its a plugin.  Install it and enable it and see if it makes any difference.  If all of your drives are spun down it should improve the responsiveness

I installed this plug-in, set my share folders as included, and enabled the plug-in. It's sitting with the unraid wavy icon showing it's doing something, but it's been like that for a few minutes now. How long does it take to start normally?

If the spinner is up for more than exactly 120 seconds it will never disappear.  reload the page.

5 hours ago, Squid said:

If the spinner is up for more than exactly 120 seconds it will never disappear.  reload the page.

Thanks, refreshing the page cleared up the spinner. The server is still behaving oddly even after the plugin, shares take minutes to open and there's some mystery thing using one of my cores pegging it at 100%. Everything was running fine up until this morning. This is really frustrating.

I hare errors like this in my syslog:
Error: shell_exec(/bin/df '/mnt/remotes/SQUIRRELSTOR_UnraidBackup' --output=size,used,avail | /bin/grep -v '1K-blocks' 2>/dev/null) took longer than 5s!

nginx: 2021/09/12 20:38:19 [error] 12606#12606: *2990054 upstream timed out (110: Connection timed out) while reading response header from upstream, client: 192.168.15.178, server: , request: "POST /webGui/include/Boot.php HTTP/1.1", upstream: "fastcgi://unix:/var/run/php5-fpm.sock", host: "192.168.15.187", referrer: "http://192.168.15.187/Main"

 

My SMB mounts to my other NAS are all working except one, that UnraidBackup mount. I'm trying to unmount it, and it's been unmounting for 10 minutes now.

So, it appears as though Appdata Backup/Restore v2 was causing all of my issues, the cpu core spiking, the shares taking forever, everything. I don't know why though. Because of the above errors I looked into my share on my other NAS that I had mounted in unraid, and nothing was showing up when I looked at it in unraid, but it looked fine on the other nas. I couldn't unmount the share in unraid either. I left it unmounting overnight, and aborted the running backup from Appdata Backup/Restore v2, and over night it looks like everything cleared up. The unmount timed out again like before, but the backup abort finally took, and freed up the share so content is visible again. My other shares open fine now, and there's no core spiking to 100% anymore. It also looks like it finally rebooted too, as my other reboots just didn't happen. It appeared to reboot, losing connection to my dockers and the gui, but it always had my 15 day uptime in the top right. Now it says uptime 7 hours.

 

The first time my backup ran it was fast and it ran fine. I'm not sure why it freaked out this time, but if it happens again I'm removing the plug-in and looking for something else.

Edited September 13, 20214 yr by Squirreljester

Did the backup complete?

7 hours ago, Squid said:

Did the backup complete?

Very first time it ran, yes. After that it seemed like it was just stuck and it wouldn't abort. I left it overnight and sometime during the night it finally aborted.