-
Notifications
You must be signed in to change notification settings - Fork 233
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
puppetdb fails to start after switching to v8.1.0 #412
Comments
|
@vchepkov If the service fail to restart, there are corrective changes right? What are they? Also the message indicate that the server cannot reach the database in 3s and timeout. Have you tried to measure how long it take to connect? Maybe some changes delay the connection and a raising the value way workaround the issue? |
That's a good question. I only used it for bootstrap and it didn't start. I will try to see, what correction v8.0.1 will do |
PuppetDB runs on the same server as database, so, there is no any delay |
From what I can tell, only permissions on configuration files are changing and puppetdb is still operational
But if you start with v8.1.0, puppetdb service doesn't start, strange |
@smortex , this code has a logic flaw
Since puppetdb is not installed yet during initial run, puppetdb_version is nil, and split probably provides not the result one would expect P.S. never mind, it's a variable too |
Could be a missing ordering dependency? Configure the database and only then attempt to start PuppetDB that need it? |
I am not sure, wasn't able to find a culrpit yet. You can easily reproduce it using a vagrant project:
|
v8.1.0 breaks PuppetDB: puppetlabs/puppetlabs-puppetdb#412
In case anyone else messes up their system with v8.1.0 of puppetlabs-puppetdb, I was able to recover (on Debian 11) by setting the puppetlabs-puppetdb version to 8.0.1 in
|
I have compared the catalogs produced by 8.0.1 and 8.1.0
|
@vchepkov nice, I'd guess the issue is in the changes to the SQL code, I'm assuming the file mode changes are innocuous. |
Right, so the command create database fails and that fact doesn't get detected by puppet
It makes an assumption that locale is installed, but it's not, by default
So, if module wants to use this locale, it should make sure glibc-langpack-en is installed before postgresql started |
INFO: the responsible commit is 0c5c270 And a question, why is not possible to use C as locale? Adding the package in question does not seem to be enough, the puppetdb still hangs
|
I'm observing the same problem on a rhel-8 where the locale en_US.UTF-8 is not installed. Changing the locale to C.UTF-8 in source of the module fixes it. The en_US.UTF-8 was arbitrary chosen in commit 0c5c270 I guess C.UTF-8 should be a better match, as it's supported since glibc 2.53 and should be more widely available. |
Describe the Bug
Upgrading to module v8.1.0 renders puppetdb unoperational , service doesn't start
The following log entries are observed in puppetdb.log
Reverting back to v8.0.1 restores functionality
Environment
AlmaLinux 9.4
The text was updated successfully, but these errors were encountered: