Articles in this section

See more

DMARC segfault errors in server logs: _mh_fork(): The child process with pid XXXXX killed by signal 11

Avatar
Marc Vidal
Updated
Follow
Plesk for Linux kb: fixed kb: bug ABT: Group A

Applicable to:

  • Plesk for Linux

Symptoms

  • DMARC feature is enabled in Plesk at Tools & Settings > Mail Server Settings > Enable DMARC to check incoming mail.

  • The temporary mail files from /usr/local/psa/handlers/spool aren't removed and the disk space usage increases.

  • In /var/log/mail.log persist following errors for emails from particular domains:

    CONFIG_TEXT: postfix-local[14661]: _mh_fork(): The child process with pid 14665 killed by signal 11
    postfix-local[14661]: Error during 'dmarc' handler

  • In /var/log/messages:

    CONFIG_TEXT: kernel: [251579.818063] dmarc[14665]: segfault at 0 ip 00007f60f12ac0b0 sp 00007ffeeb867998 error 4 in libc-2.24.so[7f60f118b000+195000]

  • The following errors can be found in dmesg output:

    # dmesg | grep 'segfault at'
    [ 36.586810] dmarc[2446]: segfault at 0 ip 00007fe5465070b0 sp 00007ffd84f303e8 error 4 in libc-2.24.so[7fe5463e6000+195000]

Cause

Product issue:

  • #PPPM-10547 "DMARC checks no longer result in a segmentation fault under specific circumstances, which caused errors to be written to the logs and temporary files not being cleaned up."
    Fixed in:

Resolution

Update your Plesk server:

Comments

7 comments

Sort by Date Votes
  • Avatar
    Adam Dobruczky

    Thank you for creating a bug ticket for this error what we have found.

    I'm waiting for the solution. If you have any questions, or need more info, please pm me.

    This can be helpful too: https://talk.plesk.com/threads/dmarc-changes-in-mu-51-introduce-new-segfault-error.352084/#post-863262

    2
    Comment actions Permalink
  • Avatar
    Alisa Kasyanova

    @Adam Dobruczky
    I have forwarded this thread to the developers. Thank you for your cooperation!

    0
    Comment actions Permalink
  • Avatar
    TomBob

    That bug is still not solved mid October 2019 and has carried over to Obsidian as well.

    1
    Comment actions Permalink
  • Avatar
    Ivan Postnikov

    Hello @TomBob,

    You're right. There're plans to fix this bug.

    However, the ETA will be available later.

    Thank you for understanding.

    0
    Comment actions Permalink
  • Avatar
    b_p

    Ivan Postnikov so is there a workaround for 17.8 as well given that you seem to know the issue? Any instructions on how we can fix them?

    0
    Comment actions Permalink
  • Avatar
    Julian Bonpland Mignaquy

    Hi b_p, We plan to publish a microupdate for Plesk 17.8 within a week from next Monday 16th.

    1
    Comment actions Permalink
  • Avatar
    b_p

    Update: the latest log files do not seem to contain the error message any more, so it seems to be fixed. Cool!

    0
    Comment actions Permalink

Please sign in to leave a comment.

Have more questions? Submit a request

Related articles