Seems like something is not working quite right for me with my installation.
Doesn’t appear to be able to read or get cluster data to generate capacity numbers.
Not sure if related to kubernetes service account or something not having access. Any advice on troubleshooting this would be appreciated. It keeps anything from being able to run it would seem.
bash-4.4$ awx-manage list_instances
[tower capacity=0 policy=100%]
awxstg-69b7d5c47-zqtsz capacity=0 node_type=hybrid version=19.5.0
[controlplane capacity=0 policy=100%]
awxstg-69b7d5c47-zqtsz capacity=0 node_type=hybrid version=19.5.0
[default capacity=0 policy=100%]
Something is preventing it from being able to determine capacity.
Just can’t figure out what that is yet.
In recent releases, the instance model (which you can access directly at endpoint /api/v2/instances/) has a new “errors” field. If you can identify your instance in that list and see that field, it might have a descriptive reason why it failed its periodic health check. In your screenshot I see the last time the health check was run was 12/3 10:27am. A question you want to ask is if that is sufficiently recent. For the main cluster, health checks should repeat once every 20 seconds or so. If it doesn’t, then I’d hope to see some more basic errors in the logs.
It’s also surprising to see the node_type is hybrid. For the supported install method with the operator it should be “control”. I’m curious if there’s a problem with the awx-operator, or migrating from a prior release.
I am also into the similar issue. Can someone help, what will be the fix.
bash-4.4# awx-manage list_instances
[tower capacity=0 policy=100%]
awx capacity=0 version=17.0.1
May be because of this, my redis container is also not able to start.
redis “docker-entrypoint.s…” 5 hours ago Restarting (1) 18 seconds ago
Hi,
What is your setup? are you using awx-operator or the development docker based environment? also which AWX version?
In the UI you can run a health check which should re-calculate the capacity for that instance. After running it, can you navigate to the /api/v2/instances/ endpoint and find the “errors” field for that instance? does it report anything there?
AWX Team
