Public bug reported:
Edge Prometheus snap refreshed on a few managed clouds and saw them all loose
communication with the exporter. The primary alert we received was the
following:
HTTP CRITICAL - No data received from host
jhebden rolled beta, edge, and candidate channels back to revision 16 to fix.
recreating the bug was more difficult than expected.
Once the edge snap had been rolled back, I ran 'snap refresh' on all 4 clouds
and saw the majority of alerts change to:
connect to address 127.0.0.1 and port 9183: Connection refused
I see nothing listening on port 9183 and I see multiple exporter processes
running (this might be normal but not sure):
systemctl list-units|grep openstack
run-snapd-ns-prometheus\x2dopenstack\x2dexporter.mnt.mount loaded active
mounted /run/snapd/ns/prometheus-openstack-exporter.mnt
snap-prometheus\x2dopenstack\x2dexporter-10.mount loaded active
mounted Mount unit for prometheus-openstack-exporter
snap-prometheus\x2dopenstack\x2dexporter-16.mount loaded active
mounted Mount unit for prometheus-openstack-exporter
snap-prometheus\x2dopenstack\x2dexporter-18.mount loaded active
mounted Mount unit for prometheus-openstack-exporter
jujud-unit-prometheus-openstack-exporter-0.service loaded active
running juju unit agent for prometheus-openstack-exporter/0
** Affects: prometheus-node-exporter (Ubuntu)
Importance: Undecided
Status: New
--
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/1736742
Title:
New Prometheus snap refresh brakes exporter
To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/prometheus-node-exporter/+bug/1736742/+subscriptions
--
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs