mirror of
https://github.com/fosrl/pangolin.git
synced 2026-07-16 06:08:33 -05:00
[GH-ISSUE #2968] Health is Unknown but healthcheck is healthy! #17265
No Assignees
Notifications
Due Date
No due date set.
Dependencies
No dependencies set.
Reference: github-starred/pangolin#17265
Reference in New Issue
Block a user
Delete Branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Originally created by @itsbhanusharma on GitHub (May 2, 2026).
Original GitHub issue: https://github.com/fosrl/pangolin/issues/2968
Describe the Bug
I am trying to make some sense of what's exactly happening here or if I am missing something here, My healthchecks are passing for public sites but health stays at unknown.
Environment
To Reproduce
Visit Resources > Public page.
Expected Behavior
Health Column should reflect the health status, uptime bars should show up as expected.
@svillar commented on GitHub (May 2, 2026):
I think the problem is that in the resource listing the healthcheck status is not properly updated. In my case I stopped newt in one of my resources, the healthcheck changed to unhealthy at some point, but when newt reconnected, the status didn't change. What's more, if you open the dropdown from healthcheck status it shows the site is healthy but the listing shows unhealthy.
@AstralDestiny commented on GitHub (May 2, 2026):
If you remove the target, save and re-add the target does the issue persist?
@itsbhanusharma commented on GitHub (May 2, 2026):
@AstralDestiny I deleted the target, Clicked save, then added the target again, enabled health check, saved again, the situation remains the same, Healthcheck works but health stays unknown, not sure if this has something to do with CE vs EE? (I am on CE and a lot of features are unable there)
@itsbhanusharma commented on GitHub (May 2, 2026):
@svillar I had thought of that possibility but I restarted everything just to be sure, mine stayed stuck at unknown even after everything was rebooted.
@oschwartz10612 commented on GitHub (May 2, 2026):
This should work just fine on the CE. I would expect deleting and recreating the target to work - but perhaps for some reason it is not and there is a bug there.
If you create a new resources does it work?
@itsbhanusharma commented on GitHub (May 2, 2026):
@oschwartz10612 I tried creating a new resource and it is still stuck on the same "unknown" status. if I click the drop down next to unknown it shows the resource as healthy. Check the attached screenshot.
Next I can try setting up a fresh newt and trying to spin up a new site using that. Just curiously asking, would the install method for newt make a difference? all my newt instances are dockerized and I haven't tested with one installed on the host os.
@svillar commented on GitHub (May 2, 2026):
That's exactly what I see and I use the binary distributed newt, so it does not depend on that.
@AstralDestiny commented on GitHub (May 2, 2026):
Like on EE,
vs on CE,
Even after many refreshes on the CE part, Yes CE is behind the EE but newt can connect just fine directly to the CE edition.
Healthcheck despite being unknown shows up in
docker exec pangolin curl http://localhost:3001/api/v1/traefik-config | jqthat it's in the routing config for traefik.@oschwartz10612
@oschwartz10612 commented on GitHub (May 2, 2026):
Ahh okay yes after looking into the code this is a CE issue. Will get a fix out in this next patch release today
@itsbhanusharma commented on GitHub (May 3, 2026):
Thanks @oschwartz10612
After updating to 1.18.2 I had to remove and re-add all the resources but once that was done, statuses seem to be reflecting correctly now.
@benello commented on GitHub (May 3, 2026):
I also had the same problem. Disabling and enabling the health check did the trick as well!
@Nanianmichaels commented on GitHub (May 4, 2026):
Had to do the same thing. Which, with over 20 targets, was a bit of a slog. It would be nice not to have to do that.
Weirdly enough, I had a couple of targets that showed proper health status without me needing to touch them, so I wonder if there's something else going on in the code that causes this kind of behavior...
@oschwartz10612 commented on GitHub (May 4, 2026):
Unfortunately this effects people who used the original 1.18.0 migration before it was patched in the 1.18.1 version. I added a migration to the 1.18.3 version that will come out earlier this week to help fix this and the mismatch between the health check status and resource status issue.
@fpillajo commented on GitHub (May 8, 2026):
I upgraded from version 1.16.2 CE to version 1.18.3 CE and no information appears on the sites or in the resources.
Screenshot Site
Screenshot Resources
@oschwartz10612 commented on GitHub (May 8, 2026):
Yeah this is a known bug will be fixed in the next patch
It is excluding data for "today"