imapsieve suddenly not working anymore

Thomas Stein himbeere at meine-oma.de
Wed Aug 21 14:49:32 EEST 2019


Found the solution. Mail was copied/moved to mailbox INBOX.Spam instead 
of SPAM.

cheers, t.

On 2019-08-21 13:01, Thomas Stein via dovecot wrote:
> On 2019-08-21 12:30, Thomas Stein via dovecot wrote:
>> Setting logging to debug reveals something is happening but the actual
>> scripts do not run i suppose.
>> 
>> Aug 21 11:54:23 imap(himbeere)<31571><ndLQjp2Q2K1/AAAB>: Debug:
>> Mailbox INBOX: Mailbox opened because: SELECT
>> Aug 21 11:54:23 imap(himbeere)<31569><kqPPjp2Q1q1/AAAB>: Debug:
>> imapsieve: mailbox INBOX.Spam: MOVE event
> 
> Maybe that's the problem? The "MOVE" event instead of a "COPY" event?
> 
> 
>> Aug 21 11:54:23 imap(himbeere)<31569><kqPPjp2Q1q1/AAAB>: Debug: sieve:
>> Pigeonhole version 0.5.7.1 (db5c74be) initializing
>> Aug 21 11:54:23 imap(himbeere)<31569><kqPPjp2Q1q1/AAAB>: Debug: sieve:
>> include: sieve_global is not set; it is currently not possible to
>> include `:global' scripts.
>> Aug 21 11:54:23 imap(himbeere)<31569><kqPPjp2Q1q1/AAAB>: Debug: sieve:
>> Sieve imapsieve plugin for Pigeonhole version 0.5.7.1 (db5c74be)
>> loaded
>> Aug 21 11:54:23 imap(himbeere)<31569><kqPPjp2Q1q1/AAAB>: Debug: sieve:
>> Sieve Extprograms plugin for Pigeonhole version 0.5.7.1 (db5c74be)
>> loaded
>> Aug 21 11:54:23 imap(himbeere)<31569><kqPPjp2Q1q1/AAAB>: Debug:
>> imapsieve: Static mailbox rule [1]: mailbox=`Spam' from=`*'
>> causes=(COPY FLAG) =>
>> before=`file:/usr/share/dovecot/sieve/report-spam.sieve' after=(none)
>> Aug 21 11:54:23 imap(himbeere)<31569><kqPPjp2Q1q1/AAAB>: Debug:
>> imapsieve: Static mailbox rule [2]: mailbox=`*' from=`Spam'
>> causes=(COPY) =>
>> before=`file:/usr/share/dovecot/sieve/report-ham.sieve' after=(none)
>> Aug 21 11:54:24 imap(himbeere)<31571><ndLQjp2Q2K1/AAAB>: Debug:
>> Mailbox INBOX: UID 132668: Opened mail because: prefetch
>> Aug 21 11:54:24 imap(himbeere)<31571><ndLQjp2Q2K1/AAAB>: Debug:
>> Mailbox INBOX: UID 132668: Opened mail because: access
>> Aug 21 11:54:24 imap(himbeere)<31571><ndLQjp2Q2K1/AAAB>: Debug:
>> Mailbox INBOX: UID 132668: Opened mail because: MIME part
>> Aug 21 11:54:24 imap(himbeere)<31571><ndLQjp2Q2K1/AAAB>: Info: Logged
>> out in=427 out=4207 deleted=0 expunged=0 trashed=0 hdr_count=1
>> hdr_bytes=507 body_count=1 body_bytes
>> 
>> On 2019-08-20 17:33, Thomas Stein via dovecot wrote:
>>> Hello one and all.
>>> 
>>> Dovecot version 2.3.7.1
>>> 
>>> I've configured imapsieve like
>>> https://wiki.dovecot.org/HowTo/AntispamWithSieve a while a go and it
>>> worked
>>> for years now. Suddenly i noticed moving mails to the spamfolder does
>>> not trigger the report-spam.sieve
>>> script anymore.
>>> 
>>> sieve-test gives:
>>> 
>>>  ~/.maildir/.Spam/cur $ sieve-test
>>> /usr/share/dovecot/sieve/report-spam.sieve
>>> 1542388745.M99384P16720.meine-oma.de\,S\=8173\,W\=8373\:2\,S -D
>>> sieve-test(himbeere): Debug: sieve: Pigeonhole version 0.5.7.1
>>> (db5c74be) initializing
>>> sieve-test(himbeere): Debug: sieve: include: sieve_global is not set;
>>> it is currently not possible to include `:global' scripts.
>>> sieve-test(himbeere): Debug: sieve: Sieve imapsieve plugin for
>>> Pigeonhole version 0.5.7.1 (db5c74be) loaded
>>> sieve-test(himbeere): Debug: sieve: Sieve Extprograms plugin for
>>> Pigeonhole version 0.5.7.1 (db5c74be) loaded
>>> debug: file storage: Using Sieve script path:
>>> /usr/share/dovecot/sieve/report-spam.sieve.
>>> debug: file script: Opened script `report-spam' from
>>> `/usr/share/dovecot/sieve/report-spam.sieve'.
>>> debug: Script binary /usr/share/dovecot/sieve/report-spam.svbin
>>> successfully loaded.
>>> debug: binary save: not saving binary
>>> /usr/share/dovecot/sieve/report-spam.svbin, because it is already
>>> stored.
>>> report-spam: error: the imapsieve extension cannot be used outside 
>>> IMAP.
>>> sieve-test(himbeere): Info: final result: failed; resolved with
>>> successful implicit keep
>>>  ~/.maildir/.Spam/cur $
>>> 
>>> I'm not sure the "the imapsieve extension cannot be used outside 
>>> IMAP"
>>> is the error already or thats only because
>>> the sieve-test script.
>>> 
>>> Any ideas on that?
>>> cheers, t.


More information about the dovecot mailing list