cPanel FATAL Digest error

It would appear that the lovely people at cPanel have had an issue with mirror replication between their servers – or so goes the explanation / joining the dots / best case scenario. Updates are failing with “***** FATAL: Digest for cpanel-” in the logs – suggesting that the checksum either didnt match – or was not there. Either something was broken, or something have ‘been broken’.

So I arrive to the office this morning to find a number of emails from cPanel boxes – all of which have failed to patch automatically overnight. Curious.

So – running the usual command line goodness – I arrive at the the FATAL: Digest error and the wheels come off, and it all grinds to a halt – thus:

# /usr/local/cpanel/scripts/upcp

[2016-01-05 09:17:08 +0000] Downloading http://httpupdate.cpanel.net/RPM/11.46/centos/6/x86_64/cpanel-perl-514-PathTools-3.47-1.cp1146.x86_64.rpm
[2016-01-05 09:17:10 +0000] Downloading http://httpupdate.cpanel.net/RPM/11.54/centos/6/x86_64/cpanel-clamav-virusdefs-0.98.7-1.cp1154.x86_64.rpm
[2016-01-05 09:18:05 +0000] ***** FATAL: Digest for cpanel-clamav-virusdefs-0.98.7-1.cp1154.x86_64.rpm is different from expected
[2016-01-05 09:18:05 +0000] The Administrator will be notified to review this output when this script completes
[2016-01-05 09:18:05 +0000] ***** FATAL: Failed to stage “new RPMS” because of an error: Digest for cpanel-clamav-virusdefs-0.98.7-1.cp1154.x86_64.rpm is different from expected at /usr/local/cpanel/scripts/updatenow.static-cpanelsync line 31388.
[2016-01-05 09:18:05 +0000] : see http://go.cpanel.net/rpmcheckfailed for more information
[2016-01-05 09:18:05 +0000] The Administrator will be notified to review this output when this script completes

Now as I start to ponder this, with different boxes failing on different files (that is the bit that bugs me the most I guess), as other boxes start to report in – it is clear the issue is wider.

Tickets raised with the mothership – and they confirm.

Solution is to not use mirrors and use the main server.

Sure, the main server has a bad day in the office, but the issue goes away – there are no more stinky emails for the rest of the team to read and give “that look” or comment regarding, and the confidence in knowing you are on the current considered safe point release – go team.

SO – should this continue to be an issue, or raise its head again with that error message (the suggested link in the logging there – while great – is in no way relevant to digest issues incidentally)… here is the work around that worked this time:

echo ‘HTTPUPDATE=75.126.236.226’ >> /etc/cpsources.conf

Then run the update script / GUI again.

On completion – revert back:

mv /etc/cpsources.conf /etc/cpsources.conf.OFF

and for whatever reason – re enable the alerting / restart daemon as advised.

Job done.

One Response to “cPanel FATAL Digest error

  • **ADVISORY**

    Make sure you do turn this off again afterwards.
    The reason why I say this bit me today on one host where I had omitted to do so.
    Vulnerabilities discovered in older releases.
    Machine had attempted to update to new version.
    As far as it was concerned it was up to speed.
    It was not.
    Local version was 11.54.0.1 (insecure) … when other servers we have were on version 11.54.0.7 (secure).
    As such carry out the move to remove that file – and you have a winner when you re-run the script to update cPanel.
    Not so smart now… however resolved.

Leave a Reply

Your email address will not be published. Required fields are marked *

%d bloggers like this:
Skip to toolbar