Author: Jeremy Harris Date: To: exim-users Subject: Re: [exim] Connection timed out errors
On 01/02/2023 22:53, MRob via Exim-users wrote: > Sorry, maybe I wrote it wrong: question is more to inquire if Exim checking any internal flags or status that make it different from use "telnet [host] 25" on command line. I dont understand why I could telnet-by-hand with immediate successful connection/no slow connect as soon after I saw the error in the log tail.
If there has been an error for a specific destination host in the
past, it is remembered so as to avoid trying to use that host
again. Most mail destinations run multiple MX's so an alternate
will get used.
That memory does expire eventually.
Look up "hints database" in the Concept Index, if you want
more details.
> Is any tip for how to take other debug steps or a way to "coax" exim to see what I see? Thank you for response, I do not mean to bother but this problem is very hard to understand.
If you have a queued message which needs to be sent to the host
in question, you can run a deliver attempt on it manually, with
debug enabled. See the manual section on commandline options.
> Also, maybe simple problem is the "timer" length was inadvertantly change. Do you mind to say if that timeout comes from a certain exim configuration setting? Thank you!
The value is an option for the transport, in the configuration.
