mtalog filling up with old message records

[Mark Sapiro]

On 01/06/2018 05:55 AM, Matt . wrote:
> Hi Guys,
> 
> I have a strange issue from a MTA which had a DNS issue on the lookup
> for it's a-record. Because of this a loop existed and all loop
> messages got removed from the postfix spool.
> 
> What is happening at the moment is that the mtalog of Mailscanner is
> filled up wit about 400K of records from time to time about this and I
> cannot find where they come from. I can remove them but then after a
> while they are added again:
> 
> 30 Query     REPLACE INTO mtalog
> (`timestamp`,`host`,`type`,`msg_id`,`relay`,`dsn`,`status`,`delay`)
> VALUES (FROM_UNIXTIME('1515068227'),'backup-01','relay','3787FEEC01D','none','5.4.4','bounced
> (Host or domain name not found. Name service error for
> name=foo.bar.com type=A: Host not found)',SEC_TO_TIME('196'))
> 
> And this multiple times.
> 
> Why is this happening and can I fix it ?


I think you answered your own question. There is some Postfix
misconfiguration that is creating some kind of retry loop when host
lookup fails. This is more of a Postfix question than a MailScanner
question. You need to examine your Postfix config and more than this one
log message to figure it out.

See <http://www.postfix.org/lists.html> for info on Postfix lists.

-- 
Mark Sapiro <mark at msapiro.net>        The highway is for gamblers,
San Francisco Bay Area, California    better use your sense - B. Dylan