Hi,
one week ago I upgraded my Plesk from 12.5.30 to 17.0.17. There was no specific reason, except I assumed it would be best to keep it up-to-date.
Since then, I have received some reports of missing emails from my clients. They claim that there was even no reject mail sent back. Looking a the logs, I can not find a lot of issues and most mail seems to go through, but for some claimed ones I can't find any logs.
There is one particular part in the logging which came back regularly though around the suspected time of issue:
Dec 6 07:21:32 mail /usr/lib64/plesk-9.0/psa-pc-remote[22659]: handlers_stderr: SKIP
Dec 6 07:21:32 mail /usr/lib64/plesk-9.0/psa-pc-remote[22659]: SKIP during call 'grey' handler
Dec 6 07:21:32 mail journal: greylisting filter[13940]: Unable to execute settings selection query: database is locked
Dec 6 07:21:32 mail journal: greylisting filter[13940]: Unable to get data from GL database
Dec 6 07:21:32 mail journal: greylisting filter[13940]: Unable to check message
Dec 6 07:21:32 mail /usr/lib64/plesk-9.0/psa-pc-remote[22659]: Error during 'grey' handler
I am not sure this is the actual cause, but it gave me some questions:
- What is causing this db lock?
- Should Postfix not sent back a reject mail in case psa-pc-remote fails/crashes, or is that not guaranteed?
- Is it normal that there is still a binary from plesk-9.0 invoked when using Onyx?
Is there any other place this milter is logging?
I had a quite stable mail server up till now. What worries me most is mails getting lost without error, I can not explain this to the clients.
Thanks for any input.
Kr
Sam
one week ago I upgraded my Plesk from 12.5.30 to 17.0.17. There was no specific reason, except I assumed it would be best to keep it up-to-date.
Since then, I have received some reports of missing emails from my clients. They claim that there was even no reject mail sent back. Looking a the logs, I can not find a lot of issues and most mail seems to go through, but for some claimed ones I can't find any logs.
There is one particular part in the logging which came back regularly though around the suspected time of issue:
Dec 6 07:21:32 mail /usr/lib64/plesk-9.0/psa-pc-remote[22659]: handlers_stderr: SKIP
Dec 6 07:21:32 mail /usr/lib64/plesk-9.0/psa-pc-remote[22659]: SKIP during call 'grey' handler
Dec 6 07:21:32 mail journal: greylisting filter[13940]: Unable to execute settings selection query: database is locked
Dec 6 07:21:32 mail journal: greylisting filter[13940]: Unable to get data from GL database
Dec 6 07:21:32 mail journal: greylisting filter[13940]: Unable to check message
Dec 6 07:21:32 mail /usr/lib64/plesk-9.0/psa-pc-remote[22659]: Error during 'grey' handler
I am not sure this is the actual cause, but it gave me some questions:
- What is causing this db lock?
- Should Postfix not sent back a reject mail in case psa-pc-remote fails/crashes, or is that not guaranteed?
- Is it normal that there is still a binary from plesk-9.0 invoked when using Onyx?
Is there any other place this milter is logging?
I had a quite stable mail server up till now. What worries me most is mails getting lost without error, I can not explain this to the clients.
Thanks for any input.
Kr
Sam