This message was deleted.
# rke2a
adamant-kite-43734
04/30/2024, 3:57 PMThis message was deleted.
c
creamy-pencil-82913
04/30/2024, 4:13 PMsounds like it’s not scraping the etcd metrics properly. If your etcd cluster didn’t have a leader you would know about it very quickly because your cluster would be non-functional.
a
acoustic-bird-95130
04/30/2024, 4:37 PMThats what I figured.
acoustic-bird-95130
04/30/2024, 4:38 PMAny advice or reference on how I can work through it?
acoustic-bird-95130
04/30/2024, 4:38 PMI was going to start teaching myself alerts, but I dont want to get a million ETCD errors.
c
creamy-pencil-82913
04/30/2024, 4:48 PMI’m not sure what metric that field pulls from, you might poke at the dashboard and compare that to the metrics you’re actually scraping.
creamy-pencil-82913
04/30/2024, 4:49 PMI don’t muck about with the reporting and monitoring stuff much
p
powerful-librarian-10572
04/30/2024, 5:31 PMI have exactly the same "issue" but i figured out it was just a scraping bug
a
acoustic-bird-95130
04/30/2024, 6:14 PM@powerful-librarian-10572 Were you using the integrated Grafana? Mine is just that way from the Apps/Helm, Install Rancher Monitoring package. I didnt configure any scraping.
p
powerful-librarian-10572
04/30/2024, 6:34 PMYep, rancher-monitoring, all stock.
powerful-librarian-10572
04/30/2024, 6:35 PMWhat i mean by "i figured out it's a scraping bug" i just ignore it
a
acoustic-bird-95130
04/30/2024, 6:39 PMlol
c
calm-coat-56369
05/01/2024, 6:28 AMI have the same call from my cluster ^^.
3 nodes each Control Plane, Etcd v1.29.2+rke2r1
a
acoustic-bird-95130
05/01/2024, 12:57 PMis this just like something, one of us should open an Issue for?
p
powerful-librarian-10572
05/01/2024, 6:24 PMProbably @calm-coat-56369 as he's running kubes 1.29 (assuming then it's rancher 2.9), which is more bleeding edge than my 2.8.3 rancher with kubes 1.28
2 Views