Resolved "INFO: pum is called with arguments: ['--list', '--repo-info', '--json']" - But Faultless Updates via CLI

[learning_curve]

Server operating system version

Ubuntu 22.04.2 LTS

Plesk version and microupdate number

Obsidian 18.0.53 MU#2

As per the title really

This warning, is only received (daily) via the panel generated (warning) e-mail. Any updates made via CLI run perfectly with no errors or issues at all. There's no trace of the warnings / errors within any of the logs here: /var/log/plesk or here: /var/log/sw-cp-server and, if searched for within the panel https://***My Domain***:8443/modules/log-browser/ there's nothing within the System, Mail or Plesk tabs although within the Overview tab, the systemd-update-done.service is inactive which is expected, if it's a one point-in -time check, albeit the report for this systemd-update-done.service, is as follows: Execution status has failed with exit code 4, stdout: , stderr: Unit systemd-update-done.service could not be found.

We've had the same warning e-mails previously, but in all previous cases, errors / warnings were provided when running updates via CLI as well and thus, were traceable / fixable and / or were common and so a quick-fix was provided by Plesk very soon after (e.g. Grafana). Has anybody had the same these warnings with the exact same criteria / results that we have? Or, is there another check method / log location that we've maybe missed @Peter Debik ? Thanks!

 

[Peter Debik]

I do not know this one. Previously I have seen that Tools & Settings > System Updates has not been updated with the latest state. But "Recheck now" button helps in that case. I could also imagine that Tools & Settings > Server Components > button "Refresh" could make a difference, because maybe the panel is not yet aware of a new version of some file that has long been installed from the console. Other from that, probably a support staff investigation on the server will be needed to find the cause.

 

[learning_curve]

~~I have seen that Tools & Settings > System Updates has not been updated with the latest state. But "Recheck now" button helps in that case ~~

We've done that just in case. No warning e-mail, but to be fair, thinking that the warning e-mail is generated, only from a cron invoked update process (might be wrong) so we'll need to wait until tomorrow to see.

~~I could also imagine that Tools & Settings > Server Components > button "Refresh" could make a difference, because maybe the panel is not yet aware of a new version of some file that has long been installed from the console ~~

We've also done this and did receive a confirmatory components updated message etc, So again, we'll know more, after tomorrow's cron invoked updates

Other from that, probably a support staff investigation on the server will be needed to find the cause.

Hopefully not ^ if one of your above checks bears fruit, so thanks for your quick response @Peter Debik we'll update tomorrow.

 

[learning_curve]

~~~ Hopefully not ^ if one of your above checks bears fruit, so thanks for your quick response @Peter Debik we'll update tomorrow.

Left it a bit longer (just in case...) but no change unfortunately. Had another long look at where we might find the cause, but still can't see the warnings / error anywhere else, other than within the cron generated e-mail. So, we will proceed with a ticket (so you can then mark this ticket as closed, but we will still post the resolution in due course) FWIW @Peter Debik just before we do that, do you know, if, iptables-persistent and/or netfilter-persistent are required packages, in order to make this function correctly? We don't use them / have them installed, but we don't think they are (Just looking to save any wasted ticket time etc).

 

[Peter Debik]

I do not know whether iptables-persistent, netfilter-persistent have anything to do with the mismatch between the actual update status and the GUI notifications.

 

[Richard18]

As per the title really

This warning, is only received (daily) via the panel generated (warning) e-mail. Any updates made via CLI run perfectly with no errors or issues at all. There's no trace of the warnings / errors within any of the logs here: /var/log/plesk or here: /var/log/sw-cp-server and, if searched for within the panel https://***My Domain***:8443/modules/log-browser/ there's nothing within the System, Mail or Plesk tabs although within the Overview tab, the systemd-update-done.service is inactive which is expected, if it's a one point-in -time check, albeit the report for this systemd-update-done.service, is as follows: Execution status has failed with exit code 4, stdout: , stderr: Unit systemd-update-done.service could not be found.

We've had the same warning e-mails previously, but in all previous cases, errors / warnings were provided when running updates via CLI as well and thus, were traceable / fixable and / or were common and so a quick-fix was provided by Plesk very soon after (e.g. Grafana). Has anybody had the same these warnings with the exact same criteria / results that we have? Or, is there another check method / log location that we've maybe missed @Peter Debik ? Thanks!

Interesting, I've been getting the same emails since the 29/06.

Same OS and Plesk version, no changes to the server recently.

apt-get dist-upgrade runs without errors, System Updates in the GUI runs fine without any errors. Nothing obvious in the logs (but I am currently on holiday sitting by the pool so I have limited access to check more )

 

[learning_curve]

Update (so far). Support Ticket raised. Upon request, we, ourselves, conducted several tests, which produced no specific identification of the cause found. Again upon request, provided temporary SSH access for Plesk Support, but the result of that, i.e. their own tests and investigations was, the same: No specific identification of the cause found. Remember that these warning e-mails are only generated via the cron invoked system update process. They are are not generated if / when system updates are run via CLI or when the update process is run manually within the Plesk Panel. The latter two methods can and will complete the process without any issues, so there's no actual, serious production server issue here. In our opion, there's a small Plesk 'gltch' somewhere...

Our decision was to close the support ticket, accept the daily 'false positive' warning e-mails and to wait and see if the next Plesk Update fixes this issue. IF it does, we're all good to go. IF it doesn't, then this will need an official bug report and some more detailed investigation by the Plesk's Dev Team themselves.

 

[MarkM]

# plesk sbin pum --check

any specific error there?

 

[learning_curve]

# plesk sbin pum --check

any specific error there?

Unfortunately no. Zero errors.
That was checked as part of the support ticket, but we just checked again (after today's false positive' warning e-mail) just in case...

 

[Peter Debik]

Did you verify that the mail is actually sent from the server in question? Maybe it comes from another, forgotten server?

 

[learning_curve]

Did you verify that the mail is actually sent from the server in question? Maybe it comes from another, forgotten server?

It's the same e-mail etc that it's always been, albeit we would normally only receive the 'correct' version, if, we had not run updates via CLI, this ran before we did & there was some actual updates. If, had not run updates via CLI, this ran before we did and there was no updates, no emails were ever sent (correct )

Edit: Sorry, to specifically answer that ^ question, yes we did double check that and yes it is.

 

[Richard18]

My error emails have stopped!

I got a success email listing installed system updates and since then I've not had the daily error emails.

 

[learning_curve]

My error emails have stopped!

I got a success email listing installed system updates and since then I've not had the daily error emails.

Ha! Lucky you!
Having said that... As of yesterday (plus today...) we haven't had any either - yet.
There's been some Plesk updates e.g. PHP / Extension updates, plus some Ubuntu OS updates, all done via CLI in our case, so it's not impossible, that the well hidden cause, has now been overwritten luckily and this suffered collateral elimination!
We'll leave it a couple more days and if it's truly gone for good, we'll get this thread changed to 'Resolved' (by happy accident )

 

[learning_curve]

To conclude, we didn't make any changes ourselves (other than standard updates, as mentioned in the previous post) before this issue, self-corrected itself and now, we've moved onto the next Plesk Release anyway. So this thread can be changed to a resolved status when convenient @Peter Debik Thanks!

 

[LRob]

Hey,

I'll allow myself to put an update on it as I fell on it from Google and found a close solution.

Thank you for this topic and insight, it had me found the issue.

Following command showed and error in "Sury" repo: the key was obsolete since quite a few time:

plesk sbin pum --check

Fix was easy:

apt-key del B188E2B695BD4743
wget -O /etc/apt/trusted.gpg.d/php.gpg https://packages.sury.org/php/apt.gpg

Cheers!