As the title says, i cant open my vm tab anymore. No config changes lately that i can remember. tried rebooting to no avail.

 

I see

Jul 19 11:36:57 HP nginx: 2021/07/19 11:36:57 [error] 7158#7158: *26475 upstream timed out (110: Connection timed out) while reading upstream, client: 192.168.0.100, server: , request: "GET /VMs HTTP/1.1", upstream: "fastcgi://unix:/var/run/php5-fpm.sock:", host: "hp.local", referrer: "http://hp.local/Apps"

in the logs when i try.

 

All other tabs work fine. I can start the VMs via the dashboard and they run. The also autostart on Reboot.

 

What is a little interesting: when i disable vms and try to reenable, i cant because libvrt.img is in use. i have to stop docker, start vm service and then start docker to work.

 

diagnostics is attached. any help would be greatly apreciated, i have googled for hours without any help. i think my problem is very specific...

hp-diagnostics-20210719-0751.zip

Edited July 19, 20214 yr by Michael H

little update after a day of going nowhere:

 

i can eliminate libvrt.img issues. even with a new file i get the tab to load, but then it hangs on vm creation (only icon is shown, then timeout again).

also while trying this i noticed that the unmount process does not seem clean. that might also be the reason why i cant reboot cleanly. it always hangs on terminal "rebooting." and then i have to cut the power.

 

Jul 19 21:38:11 HP emhttpd: shcmd (1426): /usr/local/sbin/mount_image '/mnt/user/system/libvirt/libvirt.img' /etc/libvirt 2
Jul 19 21:38:11 HP root: mount: /etc/libvirt: mount(2) system call failed: File exists.
Jul 19 21:38:11 HP root: mount error

 

i have to delete /etc/libvirt (is empty) and mkdir /etc/libvirt to get vm service to start up again.

 

my nginx starts also very, very late. about 10 minutes after i can ssh into the machine. seems strange to me, but maybe its just something i havent paid much attention to as i never had to reboot unraid that often.

 

very frustrating stuff...

 

can anybody help me how i can see what the upstream sock would do so that i can look into that further? request: "GET /VMs HTTP/1.1", upstream: "fastcgi://unix:/var/run/php5-fpm.sock:", host: "hp.local"

 

i think it has something to do with the passed through network card, as the iommu groups also dont show up on the sysdev page:

Jul 19 22:04:48 HP nginx: 2021/07/19 22:04:48 [error] 7005#7005: *36174 upstream timed out (110: Connection timed out) while reading response header from upstream, client: 192.168.0.169, server: , request: "POST /webGui/include/SysDevs.php HTTP/1.1", upstream: "fastcgi://unix:/var/run/php5-fpm.sock", host: "hp.local", referrer: "http://hp.local/Tools/SysDevs"
Jul 19 22:04:52 HP nginx: 2021/07/19 22:04:52 [error] 7005#7005: *36144 upstream timed out (110: Connection timed out) while reading upstream, client: 192.168.0.169, server: , request: "POST /webGui/include/SysDevs.php HTTP/1.1", upstream: "fastcgi://unix:/var/run/php5-fpm.sock:", host: "hp.local", referrer: "http://hp.local/Tools/SysDevs"

 

or is there any way to "clean install" the libvirt (and maybe nginx) packages? any other pointers that i could check out?

Edited July 19, 20214 yr by Michael H