cPanel & WHM update failure in upcp script

Hello guys, I am running WHM Version: 102.0.1 and AlmaLinux v8.5.0 STANDARD kvm. Recently migrated from CentOS. I have the below problem that I cannot seem to solve: The cPanel & WHM update process failed for the following reason: Maintenance ended; however, it did not exit cleanly (256). The following events were logged: "scripts/findphpversion". Review the update logs to determine why the update failed.

...
...
[2022-01-31 22:05:27 +0200] E [/usr/local/cpanel/scripts/findphpversion] "/usr/local/cpanel/bin/python-packman" reported error code "1" when it ended: Errors during downloading metadata for repository 'BaseOS':
[2022-01-31 22:05:27 +0200] E [/usr/local/cpanel/scripts/findphpversion] - Status code: 404 for http://asi-fs-d.contabo.net/centos/8/BaseOS/x86_64/os/repodata/repomd.xml (IP: 213.136.72.159)
[2022-01-31 22:05:27 +0200] E [/usr/local/cpanel/scripts/findphpversion] Traceback (most recent call last):
[2022-01-31 22:05:27 +0200] E [/usr/local/cpanel/scripts/findphpversion] File "/usr/lib/python3.6/site-packages/dnf/repo.py", line 574, in load
...
[2022-01-31 22:05:31 +0200] Processing: Updating plugins data cache
[2022-01-31 22:05:31 +0200] - Processing command `/usr/local/cpanel/bin/refresh_plugin_cache`
[2022-01-31 22:05:31 +0200] - Finished command `/usr/local/cpanel/bin/refresh_plugin_cache` in 0.015 seconds
[2022-01-31 22:05:31 +0200] Processing: Ensuring SSL certificate information for CCS is up to date.
[2022-01-31 22:05:31 +0200] - Processing command `/usr/local/cpanel/scripts/ccs-check --run --ssl`
[2022-01-31 22:05:31 +0200] - Finished command `/usr/local/cpanel/scripts/ccs-check --run --ssl` in 0.006 seconds
[2022-01-31 22:05:31 +0200] Processing: Ensure cpanel-plugins yum repo exists
[2022-01-31 22:05:31 +0200] - Finished in 0.508 seconds
[2022-01-31 22:05:31 +0200] Processing: Checking Addon Licenses
[2022-01-31 22:05:31 +0200] - Finished in 0.002 seconds
[2022-01-31 22:05:31 +0200] Processing: Updating Public Suffix List
[2022-01-31 22:05:31 +0200] Processing: Checking End Of Life for current version.
[2022-01-31 22:05:31 +0200] - Finished in 0.094 seconds
[2022-01-31 22:05:31 +0200] Processing:
[2022-01-31 22:05:31 +0200] Maintenance complete.
=> Log closed Mon Jan 31 22:05:31 2022
----------------------------------------------------------------------------------------------------
=> Log opened from cPanel Update (upcp) - Slave (1631720) at Mon Jan 31 22:05:31 2022

When I perform the "dnf makecache", I get the following:

CentOS-8 - Base                                                                                       81 kB/s | 282  B     00:00
Error: Failed to download metadata for repo 'BaseOS': Cannot download repomd.xml: Cannot download repodata/repomd.xml: All mirrors were tried

When I do "System Update" from WHM, I get the following:

System update process has started.

"/usr/bin/dnf" reported error code "1" when it ended:  w/ /usr/bin/dnf upgrade -y
CentOS-8 - Base                                  84 kB/s | 282  B     00:00  
Error: Failed to download metadata for repo 'BaseOS': Cannot download repomd.xml: Cannot download repodata/repomd.xml: All mirrors were tried

Often errors like this can be resolved by running `dnf makecache`

System update process has finished

Any ideas how to fix this?

cPRex Jurassic Moderator

February 01, 2022 13:19

Hey there! I see that you mention running AlmaLinux, but the output you've provided certainly looks like the system thinks it is running CentOS, which is no longer supported and would explain those errors:

JohnATK

February 01, 2022 13:55

Hey there! I see that you mention running AlmaLinux, but the output you've provided certainly looks like the system thinks it is running CentOS, which is no longer supported and would explain those errors: cPanel's guide to convert from CentOS to AlmaLinux, and followed the same steps. When I LS /etc/yum.repos.d, here is what I get: almalinux-ha.repo almalinux.repo CentOS-Base.repo cpanel-plugins.repo epel-playground.repo epel-testing.repo almalinux-plus.repo almalinux-resilientstorage.repo CentOS-Linux-PowerTools.repo.rpmsave EA4.repo epel.repo kernelcare.repo almalinux-powertools.repo CentOS-AppStream.repo cPAddons.repo
Which ones should I delete? and do I do it using the rm command, or I need to do something else/extra?

February 01, 2022 14:19

I wouldn't *delete* any of them, but I would move some out of the way for testing. I'd do this to move the files out of the way, update the Yum cache, and test an update, and see if that works well:

mkdir /root/old_repos
mv /etc/yum.repos.d/CentOS-Base.repo /root/old_repos/CentOS-Base.repo 
yum clean all
yum update

That way if something does go wrong you can still move it back.

February 01, 2022 14:36

That's probably the wisest to do, I agree. So I did this, and for the following after yum update:

EA4 ( EasyApache 4 ) 1.8 MB/s | 2.7 MB     00:01
cPanel Addons Production Feed 17 kB/s | 8.9 kB     00:00
cPanel Plugins project 395 kB/s | 423 kB     00:01
CentOS-8 - AppStream 77 kB/s | 282  B     00:00
Error: Failed to download metadata for repo 'AppStream': Cannot download repomd.xml: Cannot download repodata/repomd.xml: All mirrors were tried

It's also worth mentioning that "yum clean all" resulted in "0 files removed". What are we missing? to move the AppStream now too?

February 01, 2022 15:43

I would look inside the specific repo file to see the path it's using. But yes, that one likely needs to get moved as well.

February 02, 2022 10:09

I moved CentOS-AppStream.repo and now it's updating with no issues. Thanks a lot :)

February 02, 2022 15:53

You're very welcome! It sounds like something happened during the transition to AlmaLinux that kept some of the CentOS 8 data in there, causing confusion for the updates.

cPanel & WHM update failure in upcp script

Comments

Didn't find what you were looking for?