Closed jamesbeedy closed 1 year ago
Dear James,
Thank you for the bugreport! We will check it and come back. At the moment it is a first bugreport about usage vsphere
and I suspect something similar to recently addressed MaaS issue: https://github.com/canonical/mysql-operator/issues/121.
Please compare this and this:
The root cause is that the hostname name resolution is resolving to 127.0.1.1 which is not bound to any interface, and mysqld will not accepted as a valid host to configure as a Group Replication node.
VS
... my vsphere hostname resolved to loopback and my aws unit to a valid ip on a real interface.
At the moment mysql
charm is in active migration process. The stable
and edge
risks have completely different codebase (reactive charm
VS operators framework (OF)
).
Let's continue discussion here for the OF charm in edge
risk. You are correct, we have two series there focal
and jammy
. Please focus on jammy
and do NOT use focal
as we do not update focal
for a while(focal
revision 86, jammy
revision 133).
Hope this will bring you a bit more light to the situation.
Hello @jamesbeedy ,
there's a proposed fix on PR#170 In the meantime, the proposed fix is also published to a temporary charmhub branch (expire 2023-04-27), that you can deploy with:
juju deploy mysql --channel edge/vsphere --series jammy
Just great! Thank you @paulomach!
I get this message when deploying mysql from edge in a sphere cloud.
The command I used:
juju deploy mysql --channel edge --constraints "cores=2 mem=4G root-disk=20G" --series jammy
This is what I see in the logs:
Deploying on focal gives an different error
This charm seems to deploy just fine in my local lxd environment .... but fails on vsphere.
How could I investigate this further?