Articles in this section

Unable to syncronize a mailbox or send an email on Plesk server: Mail handler 'check-quota' said: 452:4.2.2 Mailbox full

Anton Kuznetsov
Updated
Plesk for Linux kb: technical ABT: Group B

Applicable to:

  • Plesk for Linux

Symptoms

  • After sending message to mail box john_doe@alias.com the following warning is shown in webmail:

    Roundcube:

    CONFIG_TEXT: An error occurred while saving sent message. Unable to perform operation. No free disk space.

    Horde:

    CONFIG_TEXT: You exceeded mail quota

    Note: The same can be true for mail forwarding.

  • The mailbox quota is set to Unlimited or its limit is not reached at Plesk > Domains > example.com > Email Addresses.

  • Subscription disk space is set to unlimited or its limit is not reached, which can be seen at Plesk > Domains > example.com > View More Statistic.

  • Unable to synchronize a mailbox with a mail client or send an email to john_doe@example.com. The following error appears in /var/log/maillog file:

    CONFIG_TEXT: postfix/pipe[17519]: 7AEB011C1531: to=<john_doe@example.com>, relay=plesk_virtual, delay=25950, delays=25950/0/0/0.12, dsn=4.3.0, status=deferred (temporary failure. Command output: Mail handler 'check-quota' said: REPLY:452:4.2.2 Mailbox full 4.2.1 Message can not be delivered at this time )

    CONFIG_TEXT: postfix/cleanup[16783]: B22C11DC00BE: milter-reject: END-OF-MESSAGE from smtp.gmail.com[127.0.0.1]: 4.2.2 Mailbox full; from=user@gmail.com to=<john_doe@example.com> proto=ESMTP helo=<smtp.gmail.com>

Cause

The file /var/qmail/mailnames/example.com/john_doe/Maildir/maildirsize is corrupted.

Resolution

  1. Connect to the server via SSH.

    Note: if direct SSH access to the server is not possible, contact server administrator for further assistance.

  2. Remove the file maildirsize of the affected domain:

    # rm -rf /var/qmail/mailnames/example.com/john_doe/Maildir/maildirsize

    • example.com - a name of an affected domain

    • john_doe - a name of an affected mailbox

  3. Recreate the file maildirsize by re-saving the same settings defined in Plesk:

  • Log into Plesk.

  • Go to Domains > example.com > Email Addresses > john_doe@example.com > click OK without making any changes

Was this article helpful?

Comments

6 comments
Date Votes
  • Avatar
    Unknown User

    We really need a better way to resolve this. For example an automated way to determine which maildirsize files are corrupted.

    Should we sechedule a run of "plesk repair mail' every week?

    1
  • Avatar
    Tim Schenk

    Yes, indeed! By the way, this problem also appears if you move a domain to a different subscription. I just reorganised my subscription managing system and thus had to repair about 20 mailboxes like this…

    1
  • Avatar
    Adrien Foulon

    plesk repair mail does not solve this issue

    0
  • Avatar
    FutureX

    Just experienced the same thing! from moving a domain to a different subscription.  It seems you either have to check immediately after moving the domain or wait for the client to complain they aren't receiving emails!

    Very sketchy.

    0
  • Avatar
    FutureX

    More data:

    Background / Known Issue
    Mailboxes showing 452 4.2.2 Mailbox full even when quota is unlimited or well under limits, triggered by a corrupted maildirsize file Plesk Support+1.

    Why it matters

    • When moving domains or mailboxes between subscriptions, Plesk does not automatically clean up these outdated quota files.
    • Administrators must manually rm those files to restore correct mailbox behavior.
    • This is not intuitive and leads to client confusion, undelivered mail, and extra support overhead.

    Suggested Improvement

    • Plesk should automatically delete or refresh maildirsize (and potentially other Dovecot index/quota files) whenever mailboxes are moved, subscription limits change, or mail plans are updated.
    • Adding this step would prevent mailbox sync/send errors post-movement and reduce reliance on manual fixes.
    0
  • Avatar
    Anton Kuznetsov

    Hello FutureX 

    We have reproduced this behavior in our test environment and submitted a bug with ID PPPM-15068.

    Thank you for providing the feedback.

    1

Please sign in to leave a comment.