LXC Usage Graphs in GUI Blank with Wrong Date/Time

[wespiard]

I swapped to a new motherboard and initially upon boot the usage graphs of the host were empty, showing 1969-12-31 (screenshot below). I followed some instructions in this post, which fixed the issue for the pve host.

However, I'm running into the same issue for all containers and VMs. I tried deleting the file for an individual lxc, e.g., rm -rf /var/lib/rrdcached/db/pve2-vm/102, and then systemctl restart rrdcached.service, to no avail. The date/time in the LXCs themselves seem accurate.

Is there a more sure-fire way to reset this usage history?

 

[UdoB]

1969-12-31 19:00:00

That's cool - as unix time starts at 01. Januar 1970

I really did not know that a specific timezone (don't know where you are, probably -5 hours) could drop you below zero

Obviously your clock is wrong. Try something like timedatectl set-time "2024-10-05 21:19:00" and then search for chrony to setup ntp.

See https://pve.proxmox.com/wiki/Time_Synchronization

 

[wespiard]

That's cool - as unix time starts at 01. Januar 1970

I really did not know that a specific timezone (don't know where you are, probably -5 hours) could drop you below zero

Obviously your clock is wrong. Try something like timedatectl set-time "2024-10-05 21:19:00" and then search for chrony to setup ntp.

See https://pve.proxmox.com/wiki/Time_Synchronization

Indeed, EST/EDT time zone!

Time seems correct when I run the following:

root@pve:~# date
Sat Oct  5 05:09:24 PM EDT 2024

Like I said originally, the graphs on the PVE host itself were fixed after I fixed the time and restarted rrdcached. I just don't know why the LXCs/VMs are messed up.

BIOS also seems correct (shows accurate UTC time).

 

[scottysd84]

I was having the same issue and I was able to resolve the issue by removing the entries in /var/lib/rrdcached/db/ and /var/lib/rrdcached/db/pve2-node