Domains VM folder lost after upgrade to 6.10.3

[r8z0r1]

Hi, I upgraded without any issue yesterday to 6.10.3 and had previously run a stable environment for many years without incident. However today without any warning the path to mnt/user/domains was lost causing my VM's to be lost. A reboot of unraid has since resolved he issue however I wanted to check if this is known and likely to happen again?

[JorgeB]

It's not known and if it happens again grab the diagnostics before rebooting so we can see what happened.

[r8z0r1]

Hi it has happened again, which file from the diagnostics is useful, I have uploaded the syslog file for now as this shows issues start with the nvme where the folder is located: 

syslog.txt

[r8z0r1]

Here is the full zip folder from the diagnostic, something else I have noted is that after a reload it takes around 10 mins for the Unraid login page to be available despite being able to ping the server almost instantly, prior to the update the login page was available within a few seconds of being able to ping the server.

tower-diagnostics-20220804-1146.zip

[trurl]

nvme device disconnected

[JorgeB]

This sometimes helps with dropping NVMe devices:

 

Some NVMe devices have issues with power states on Linux, try this, on the main GUI page click on flash, scroll down to "Syslinux Configuration", make sure it's set to "menu view" (on the top right) and add this to your default boot option, after "append initrd=/bzroot"

nvme_core.default_ps_max_latency_us=0

e.g.:

append initrd=/bzroot nvme_core.default_ps_max_latency_us=0

Reboot and see if it makes a difference.

[r8z0r1]

Cheers, I'll give it a go although I've had the NVMe drive in for quite a while.

 

[r8z0r1]

Unfortunately that didn't work, just lost the NVMe again today, this has only started happening since I installed this Stable release?

[JorgeB]

You can try reverting to the previous release or upgrading to v6.11.0-rc3, it might not like the current kernel.

[r8z0r1]

Thanks for the advice. I've been down the road of release candidates too many times and each one comes with its own issues (as well as fixes) but ultimately leaves me in an unknown state. Likewise rolling back may have unknown consequences at this point. At least with where I am now, I understand the issue, the symptoms and how to resolve, might be a pain for a while but I can cope with it.

 

The only other thing I would point out and ask is regarding the time it takes for the Unraid login page to become available after a reload, used to be pretty much instantly after the server was pingable, it is now taking up to 10 mins for the login page to appear which coincides with the Ngix service starting when watching the syslogs. Any thoughts or is this normal for the new release?

Edited August 12, 2022 by r8z0r1

[JorgeB]

20 hours ago, r8z0r1 said:

Any thoughts or is this normal for the new release?

It's not normal, diags might show something.

[r8z0r1]

this is the diagnostic from the latest reload, seems to be required every 5 days. Hoping to understand why it takes so long for the Ngix server deamon to start and the login page to become available.

 

Aug 17 10:48:27 Tower unraid-api[3643]: ✔️ UNRAID API started successfully!
Aug 17 10:48:27 Tower emhttpd: shcmd (35): /etc/rc.d/rc.nginx start
Aug 17 10:48:45 Tower nmbd[3501]: [2022/08/17 10:48:45.888608,  0] ../../source3/nmbd/nmbd_become_lmb.c:398(become_local_master_stage2)
Aug 17 10:48:45 Tower nmbd[3501]:   *****
Aug 17 10:48:45 Tower nmbd[3501]:
Aug 17 10:48:45 Tower nmbd[3501]:   Samba name server TOWER is now a local master browser for workgroup HOME on subnet 172.16.1.1
Aug 17 10:48:45 Tower nmbd[3501]:
Aug 17 10:48:45 Tower nmbd[3501]:   *****
Aug 17 10:48:45 Tower nmbd[3501]: [2022/08/17 10:48:45.888695,  0] ../../source3/nmbd/nmbd_become_lmb.c:398(become_local_master_stage2)
Aug 17 10:48:45 Tower nmbd[3501]:   *****
Aug 17 10:48:45 Tower nmbd[3501]:
Aug 17 10:48:45 Tower nmbd[3501]:   Samba name server TOWER is now a local master browser for workgroup HOME on subnet 192.168.50.253
Aug 17 10:48:45 Tower nmbd[3501]:
Aug 17 10:48:45 Tower nmbd[3501]:   *****
Aug 17 10:55:57 Tower root: Starting Nginx server daemon...
Aug 17 10:56:09 Tower root: nginx: [warn] "ssl_stapling" ignored, host not found in OCSP responder "r3.o.lencr.org" in the certificate "/boot/config/ssl/certs/certificate_bundle.pem"
Aug 17 10:56:12 Tower root: Starting [email protected]
Aug 17 10:56:13 Tower emhttpd: shcmd (36): /etc/rc.d/rc.flash_backup start
Aug 17 10:56:13 Tower unraid-api[4210]: ✔️ UNRAID API started successfully!
 

tower-diagnostics-20220817-1109.zip

[JorgeB]

Please try booting in safe mode to see if there's any difference.