Undeliverable: OTRS Scheduler Daemon Cron: TicketPendingCheck

[MHCDdave]

Hi Folks,

Hopefully I'm doing this right, if not I apologize in advance but OTRS is currently logging tons of calls (spamming them for the most part) related to the error message listed in the subject. The full subject of the ticket is "Undeliverable: OTRS Scheduler Daemon Cron: TicketPendingCheck". In looking at it, it appears this is because some process seems to be trying to e-mail root@localhost which is being denied by our mail server with the following: "Unable to relay recipient in non-accepted domain". I'm assuming the non-accepted domain is "@localhost" so that part I think I understand. The mail it's actually trying to send is an error that I think is related to a Process I have created. That process worked fine for years but between upgrading versions and deprecation of legacy auth for e-mail something's gone south. The mail it's trying to send states the following:

"ERROR: OTRS-otrs.Console.pl-Maint::Ticket::PendingCheck-61 Perl: 5.32.1 OS: linux Time: Tue Jun 20 14:45:09 2023

Message: Need Process config!"

and then further down states:

" Message: Can't process Transition for Process 'Process-465f1a83c2f1d74fd4d258ba02f05e3f', can't get State out of the config!"

I guess my ask is two fold at this point:

1.) How can I resolve the undeliverable issue and,
2.) How can I resolve the underlying process issue. I've marked the process as invalid for now, but it's still triggering that mail message. If I close out any tickets using that process, and delete the process entirely, will that fix the corruption in that transition? Is there are a way to fix that process which doesn't require me to rebuild it from scratch? obviously it was able to get the State at one point, but something changed during Migration/Upgrade that seems to have broken it.

We're currently using version 6.4

[root]

Hi,

Mostly it's the agent root@localhost that causes such undeliverable emails. You have two options: make sure this agent does not receive ticket notifications or disable the agent. The first possibility can be implemented by restricting the permission of the admin root@localhost. Only membership in the admin group is required.

Regarding the process, it's hard to say the reason. I never had issues with processes after migration from 6.0 up to 7.0 and versions in between. What is the last known version where the process did not throw these messages?

- Roy

[skullz]

1. When scheduler error happening, mail is send to root@localhost..
To change it to valid email, check System Configuration > Daemon::SchedulerTaskWorker::NotificationRecipientEmail

2. As root state, im also had no issue with Process after several upgrade..
by the way, check all the Transition back especially their value..perhaps some value misspelling ?

[MHCDdave]

Thank you for the suggestions. I'll look into these and report back. I've removed Root@localhost from all groups except admin, so we'll see if that works. Root@localhost was a member of the users group which does receive some notifications I believe, so we'll see. Follow up question though, if I disable the agent entirely, even in the admin group, is that going to break anything?

Daemon::SchedulerTaskWorker::NotificationRecipientEmail had a valid e-mail address in it already so it wasn't that.

I believe we were on 6.1.8 before moving to 6.4. We were also on a different version of Linux I believe, though I don't know if that would have made a difference. No changes were made to the process so far as I know, so I don't know how a value misspelling would have happened if not as part of the migration.

I only have the one process configured, and while I'd like to solve it without deleting the whole of the process, I'd also like to ask if you folks think that deleting it entirely would solve the issue. It's not a highly sophisticated process, so I could recreate it, assuming the error is at the process level and not some obscure setting in sys config related to the calendar, scheduler Daemon, or ticketPending functions.