Okay, if you give a user access to the proxmox panel, he obviously can change all settings he has access too.
Including if you give him permissions to the disk settings, which you need for the boot order.
Means, he can remove the I/O limits, you set, to prevent abuse or limit abuse on a host...
I just said why, If you set I/O limits, you may not want to give the user permission to change these.
However you may want to give the user permission to change the boot order.
As you give the user edit permissions, so he can change the boot order.
The user can also edit the I/O limits you had set.
Which would be not ideal, for just giving the user the option to change the boot order.
If something like VM.Config.AuditDisk would be available, so at least you could change the boot order.
Despite that, the thing I don't get is, you can change the boot order if you don't have permission to change the network device, but you need Modify on the disk to change the boot order.
Any solution to prevent someone to change the Bandwidth limits of the Disk itself, when giving the permissions mention above to allow for boot order change?
Okay, but do you got any explanation why I sometimes get connection refused using this "workaround" ?
Despite it seems to be running.
edit:
tcp 0 0 127.0.0.1:5900 0.0.0.0:* LISTEN 0 905292292 3753452/vncterm
tcp6 0 0 :::5900...
I don't run a Proxmox Cluster, these are all single instances.
However, I tried it anyway.
Last login: Tue Feb 7 13:10:26 2023 from ::1
root@prox:~# export LC_THIS_IS_A_TEST=success
root@prox:~# echo $LC_THIS_IS_A_TEST
success
root@prox:~# ssh root@localhost
Linux prox.bla.rocks 5.15.83-1-pve...
The thing that is even more wired, is for some VM's on the same Node, VNC works via
As said above, I already checked them.
I did not find any difference within these configs when I compared the machine that doesn't have the error and these who do.
I forgot to mention, that a noVNC session via the WebUI works without issues as it has been before.
pveversion -v output
proxmox-ve: 7.3-1 (running kernel: 5.15.83-1-pve)
pve-manager: 7.3-4 (running version: 7.3-4/d69b70d4)
pve-kernel-5.15: 7.3-1
pve-kernel-helper: 7.3-1
pve-kernel-5.13: 7.1-9...
Same issue with pvesh. @Moayad
pvesh create /nodes/prox/qemu/100/vncproxy
Runs fine, however when connecting, it results in the same error.
Feb 4 08:31:51 prox pvesh[8575]: <root@pam> starting task UPID:prox:00002180:0005B88F:63DE5E47:vncproxy:100:root@pam:
Feb 4 08:31:51 prox pvesh[8576]...
It seems like I did implement it wrongly, since it was behaving buggy.
I ended up passing websocket=1 to
https://pve.proxmox.com/pve-docs/api-viewer/#/nodes/{node}/qemu/{vmid}/vncproxy
Which made the vncproxy work.
However, I never intended to use websocks, I just wanted the VNC to work...
Hey,
Oh sorry, I should have mention that.
I am using as per documentation: https://pve.proxmox.com/pve-docs/api-viewer/#/nodes/{node}/qemu/{vmid}/vncproxy
However, the code also appends ?websocket=1, using this since Proxmox 5.x.
Removing websocket=1, despite not using websockets at all, just...
Hey,
When I request a VNC session from Proxmox using the API, the timeout is pretty short with less than 5 seconds.
Anybody got an idea if its possible to adjust it? So far I didn't find any way yet....
Regards,
Ne00n
This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
By continuing to use this site, you are consenting to our use of cookies.