Missing email - Mail log says delivered, Postfix says (queue active) but nothing in the que

SYSTEM INFORMATION
OS type and version Ubuntu Linux 22.04.5
Usermin version 2.302
Webmin version 2.402
Virtualmin version 7.30.8 Pro
Theme version 24.02
Apache version 2.4.52
Package updates 41 package updates are available

The issue

I have got an email from a company that does not have the best email setup and I have been working with them to get a signup email from them.

I got an email delivered to my server today but it has not arrived in my inbox

So my system might of classed the email as spam and maybe auto deleted it as this is what I have setup for a certain points threshold.

Useful information

  • Greylisting is disabled
  • the email was sent Oct 02 09:12:32
  • in Search Mail Log (VM Pro) I can see the email and it says it has been forwarded to my email address.

  • i searched in the System Logs as follows and found the email listed

    Note it says (queue active)
  • in the postfix there are no emails
  • When I use postcat -q BCB39380B63 I get
    postcat: fatal: open queue file BCB39380B63: No such file or directory

Questions

  • What can I do now?
  • Where is the email?
  • Are there some Spamassassain logs to read?

Thanks as ever

seems rejected to me. (smtpd : NOQUEUE) what does procmail.log say ?

i don’t know, how would i view this

usually it’s /var/log/procmail.log

1 Like

the webmin syslog view is so restrictive, i am going to see if they will relook at it as it is not user friendly.

I have updated the image above, the one with NOQUEUE is the wrong one

/var/log/procmail.log

From bounces+28910110-f9bb-xxx=lancastrian-it.co.uk@em5527.haldane-fisher.co.uk  Thu Oct  2 09:12:32 2025
 Subject: Reset your Haldane Fisher password
  Folder: /dev/null							  21948
Time:1759392754 From:no-reply@haldane-fisher.co.uk To:xxx@lancastrian-it.co.uk User:xxx@lancastrian-it.co.uk Size:21948 Dest:/dev/null Mode:Spam

I decided to search in all messages for BCB39380B63 and got the following (modified for privacy)

Oct 02 10:24:33 sudo[1237174]: myusername : TTY=pts/0 ; PWD=/home/myusername ; USER=root ; COMMAND=/usr/sbin/postcat -q BCB39380B63
Oct 02 09:12:34 postfix/qmgr[4053358]: BCB39380B63: removed
Oct 02 09:12:34 postfix/local[1211055]: BCB39380B63: to=<"xxx@lancastrian-it.co.uk"@myserver.england.uk>, orig_to=<xxx@lancastrian-it.co.uk>, relay=local, delay=7.6, delays=5.8/0/0/1.8, dsn=2.0.0, status=sent (delivered to command: /usr/bin/procmail-wrapper -o -a $DOMAIN -d $LOGNAME)
Oct 02 09:12:32 postfix/qmgr[4053358]: BCB39380B63: from=<bounces+28910110-f9bb-xxx=lancastrian-it.co.uk@em5527.haldane-fisher.co.uk>, size=18426, nrcpt=1 (queue active)
Oct 02 09:12:32 opendkim[810]: BCB39380B63: key retrieval failed (s=s1, d=haldane-fisher.co.uk): 's1._domainkey.haldane-fisher.co.uk' query timed out
Oct 02 09:12:27 postfix/cleanup[1211034]: BCB39380B63: message-id=<NuUkODA8Sk2IC12UDzMk4Q@geopod-ismtpd-10>
Oct 02 09:12:26 postfix/smtpd[1211021]: BCB39380B63: client=xxxxxxxx.outbound-mail.sendgrid.net[159.183.209.122]

Is this correct

  • postfix delivers the email to procmail
  • the OpenDKIM deletes the email because the DKIM records are faulty
  • Mail logs feature does not show this, and should.
  • The OpenDKIM log only shows the key retrieval failed

My settings

image

somewhat correct. procmail delivers emails to mailbox.
or /dev/null if instructed to reject based on whatever reasons.. (eg. invalid DKIM)

you shouldn’t reject invalid DKIM to receive those, or let your senders know you require valid DKIM records. (which they don’t have in this case).

Where this stuff gets logged is on the developers further up the chain.

But, have you simply tried white listing them?

I was going to suggest that. Old habit of mine after years and years.

A quick edit: It can get confusing for people because when Postfix does its’ job the message leads people to believe the whole chain was successful delivery to the inbox. In this case Postfix did it’s job and handed off to procmail so from the perspective of Postfix, it succeded in its’ part. That’s why searching on the whole string is good. Since journalctl logs everything, searching on the record number is more useful. I think I finally see the need to learn it. Damnit. I guess it is better than grepping the entire log directory. :slight_smile:

1 Like

I am not looking for workarounds becasue there will always be anotehr email address I have issues with.

So the server is doing what I asked of it but the Logs are a bit ambiguous.

I have submitted a couple of Github issues

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.