-
-
Notifications
You must be signed in to change notification settings - Fork 32
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
New puppetserver CA certs causes backup failures #165
Comments
I was going to put together a PR for this and couldn't figure out how/where these certs/key were getting deployed by the module so that I could add the appropriate |
I think it was half automatic at some point, but fragile so we had to rework it extensively 😉 Changing the transport certs should not hurt, but if you encrypted the data, beware! Changing the so called "PKI" certs will silently prevent you from restoring 😨 Testing advised! |
@smortex Thanks for that heads up. This was only transport, but I tested a restore of a backup before and after the switchover just to be sure. |
I've had a long stretch of success with this module but I believe I've uncovered a bug.So what changed? I rebuilt my Puppet Server, going from puppetserver-6.16.1-1.el7 on CentOS 7 to puppetserver-7.4.1-1.el8 on CentOS 8 Stream. I did not try to preserve my Puppet DB content nor any of the old certificates. New DB, new CA. That meant wiping the ssldir of each puppet client, new CSRs and signing. Nice and clean and not too much work because I only have a couple dozen clients. The module rebuilt all the expected configs from the exported resources like normal. I was therefore a bit surprised to see this morning that all my backup jobs failed overnight since this migration. Each job failed with:
Examining the certificates in /etc/bacula/ssl on scooby-f34 (my director, running on Fedora 34 with puppet-7.9.0-1.fc34 and bacula-director-11.0.5-1.fc34) didn't reveal anything and the following also looked good (as run on scooby-f34):
Digging further, I discovered the bacula-fd.service did not get restarted on the clients, nor did the bacula-sd.service.
The text was updated successfully, but these errors were encountered: