Dashboard missing vms occasionally

[agit8or1]

I have pulse installed on a few clusters and every now and then, the vms on the dashboard wont display. Sometimes a refresh will reload and display and sometimes I just have to wait for them to reappear. No communication issues from pulse to hypervisors.

[rcourtman]

When the dashboard drops VM rows briefly it typically means Pulse could not pull GetVMs from a Proxmox node during that refresh. The UI replaces that node list with nothing until the next successful poll, so a transient API hiccup appears as missing VMs. Please grab backend logs right after the rows disappear and search for Failed to get VMs by running journalctl -u pulse-backend -n 400 | grep 'get_vms'. Seeing those entries would confirm a single node call is timing out or returning a 5xx. Please also share the Pulse version (cat /opt/pulse/VERSION) and whether it affects every cluster or just a particular one. Engineering is working on keeping the previous VM data when a node poll fails so the dashboard stays populated, and the log evidence above will help confirm that this is the issue you are hitting.