On 7.1.2014, at 20.49, mego...@inboxalias.com wrote:
> I'm pretty sure I've found a bug with doveadm when concatenating with
> another search key.
>
> While doveadm HEADER works fine when not using additional
> search keys, it doesn't work properly when concatenating with other
> serarch keys
huh? You are logging a bug for antique version, if you grab latest
source, build and install it, and problem remains with current
version, THEN it is a bug, most devs would ignore you for such
sillyness otherwise
On 2/20/14, Jon Thompson wrote:
> Since I'm obviously using version 2.2.5, which is
Since I’m obviously using version 2.2.5, which is apparently a supported
version by dovecot, I’d still maintain that it should be submitted as a bug.
I’ve also submitted the bug to Apple as well.
Regardless, has anyone compiled a drop-in update that supports Apple Push
Notifications, Sieve, and
I'm using release 2.2.12, during a copy operation i get segfault on imap
process. I have full bt of the issue:
(gdb) bt
#0 0x7fcc521029cd in mail_cache_header_fields_read
(cache=0x15c7210) at mail-cache-fields.c:369
#1 0x7fcc52100664 in mail_cache_open_and_verify (cache=0x15c7210)
at
since you are obviously using the binary shipped with
Mac OSX and packaged by Apple Inc. which is heavily
outdated while current dovecot is version 2.2.12
you may complain to Apple Inc.
Am 18.02.2014 00:19, schrieb Jon Thompson:
> So when I try and create a sub mailbox in shared-folders, I get thi
So when I try and create a sub mailbox in shared-folders, I get this error…
imap(pid 56143 user -): Fatal: master: service(imap): child 56143 killed
with signal 11 (core dumps disabled)
I’ve not been able to figure out how to get dovecot in OS X to core dump, so I
haven’t included that. The
So E.g. When doing
doveadm fetch -u exampleuser hdr SUBJECT Silvester HEADER
Return-Path someu...@example.com
it initially outputs 13 headers (same as when using with SUBJECT Silvester
only) , when piping it to "grep -i 'Return-Path'" it even found only 3 out
of 13 headers containing a Return-P
Hey Joseph!
Thanks for testing.
> I couldn't recreate this with 2.1.16. Also, my version of "doveadm
> search" does not accept missing header patterns. Is this corroborated
> by dumping out the header with this same search query?
>
> doveadm fetch -u exampleuser hdr SUBJECT Silvester \
>
On Wed, 8 Jan 2014, dovecot-requ...@dovecot.org wrote:
doveadm search -u exampleuser SUBJECT Silvester
(outputs correctly 10 matches)
...
doveadm search -u exampleuser SUBJECT Silvester HEADER Return-Path
someu...@example.com
(gives exactly the same output - but this is WRONG, as the given Sub
Hey there,
I'm pretty sure I've found a bug with doveadm when concatenating with
another search key.
While doveadm HEADER works fine when not using additional
search keys, it doesn't work properly when concatenating with other
serarch keys - it then simply seems to ignore the additional HEADE
On 6.12.2013, at 15.33, email-from-bobby wrote:
> A negated subject search on an empty folder causes an assertion failure in
> doveadm. To reproduce:
>
> bash-3.2# doveadm mailbox create -u ephraim "Freshly Empty"
> bash-3.2# doveadm search -u ephraim mailbox "Freshly Empty" NOT subject
> '***
Since reporting this (from the wrong account, sorry), I've tried to
investigate whether the problem exists in 2.2.9. While building 2.2.9
was no problem, I balked at actually installing it for fear of causing
trouble with future system updates. (Or current operation, for that matter.)
Could so
A negated subject search on an empty folder causes an assertion failure
in doveadm. To reproduce:
bash-3.2# doveadm mailbox create -u ephraim "Freshly Empty"
bash-3.2# doveadm search -u ephraim mailbox "Freshly Empty" NOT subject
'***JUNK MAIL***'
doveadm(ephraim): Panic: file mail-index-map.c:
> If doveadm says it's invalid UTF-8, it's invalid UTF-8. I guess your
> terminal isn't actually using UTF-8 then, but something else. ("locale"
> output should say something about UTF-8.) I guess doveadm could also
> automatically translate parameters to UTF-8, but that's a bit annoying
> to i
On 23.9.2013, at 16.52, mego...@inboxalias.com wrote:
>> I suspect your shell is breaking the 8bit chars into invalid UTF-8. I
>> just tried and it works fine with me.
>
> Sorry, we got confusion here... apparently my (lousy) freemail provider
> which I use to post on this mailinglist doesn't se
Hi there Timo,
first of all thank you for taking your time for support!
> Forget about mUTF-7 with Dovecot. It's an IMAP-protocol relic, and
> Dovecot is trying to get rid of it in as many places as possible.
> Although by default the filesystem users mUTF-7, but that can also
> be changed with
On 20.9.2013, at 17.42, mego...@inboxalias.com wrote:
> So when using "doveadm rename" trying to rename a folder with special
> characters e.g. german umlauts like "ä" it seems to create the folder
> wrongly in the filesystem.
>
> Example: When I want to rename a users folder 'geschaftliches'
What the heck, I don't get it... seems like it's bug's day for me today... the
d*mn freenet web-mailclient omitted half of the sentence on exact the same part
*again* - so I split it to several lines and omit some quotes...
Hope this is the last try to get the _whole_ message out:
Naturally re
Sorry, it seems like some part of the sh*tty freenet webmailer omitted half of
my second sentence of the message I just posted before...
. ..so please replace the weird sentence:
"Naturally renaming a users mail folder should be done via "doveadm es
consistent."
with:
"Naturally renaming a use
Hello there,
I'm pretty sure I found a bug in latest Dovecot 2.2.5.5 (EE) when playing with
the doveadm command.
Naturally renaming a users mail folder should be done via "doveadm es
consistent.
So when using "doveadm rename" trying to rename a folder with special
characters e.g. german umlau
On Fri, 2013-01-11 at 00:08 -0800, ForestYang wrote:
> I was testing dovecot server , auth via redis.
> first I follow http://master.wiki2.dovecot.org/AuthDatabase/Dict
> but auth always fail, last , I found
> value { "uid": 123, "gid": 123, "home": "/home/username" } was wrong,
> the uid and gi
hi all,
I was testing dovecot server , auth via redis.
first I follow http://master.wiki2.dovecot.org/AuthDatabase/Dict
but auth always fail, last , I found
value { "uid": 123, "gid": 123, "home": "/home/username" } was wrong,
the uid and gid must as string , not number.
{ "uid": "123", "gid": "
On Fri, 2012-05-25 at 18:52 +0200, Peter Meier wrote:
> I was doing some migration from a 1.2 installation to a 2.1. While
> testing my new installation dovecot crashed at two test-cases constantly
> with with a "Panic: Trying to allocate 0 bytes" message.
..
> I see two problems:
>
> 1. Don't pa
Hi
I was doing some migration from a 1.2 installation to a 2.1. While
testing my new installation dovecot crashed at two test-cases constantly
with with a "Panic: Trying to allocate 0 bytes" message.
Both times it was because I was using a wrong group (or none) in my config.
The output I got by
On Fri, 2011-08-19 at 17:52 +0300, Kostas Zorbadelos wrote:
> The problem is more easily introduced than I imagined.
Well, I still couldn't reproduce it. But I can kind of see the problem.
http://hg.dovecot.org/dovecot-2.0/rev/030394c74f54 should help.
The problem is more easily introduced than I imagined.
Included attached is an example mailbox containing just 3 mails.
Use it to make a couple of POP sessions. Make sure indexes are
generated for it by setting appropriately mbox_min_index_size
e.g. mbox_min_index_size = 2k
Now, in the first
On 08/16/2011 04:42 PM, Kostas Zorbadelos wrote:
On 08/15/2011 11:17 AM, kzo...@otenet.gr wrote:
Quoting Timo Sirainen:
OK,
we managed to reproduce the problem with the following scenario.
Using Thunderbird (3.1.11 if that matters) we set up a mail account
using POP served by dovecot. In t
On 08/15/2011 11:17 AM, kzo...@otenet.gr wrote:
> Quoting Timo Sirainen :
>
>> With a quick test I can't reproduce pop3_lock_session=yes causing a
>> crash. I guess it needs something else besides what I tested. It would
>> be helpful if your Dovecot binaries weren't stripped of debug symbols. I
>
Quoting Timo Sirainen :
With a quick test I can't reproduce pop3_lock_session=yes causing a
crash. I guess it needs something else besides what I tested. It would
be helpful if your Dovecot binaries weren't stripped of debug symbols. I
could then ask for some more information from the core dumps
With a quick test I can't reproduce pop3_lock_session=yes causing a
crash. I guess it needs something else besides what I tested. It would
be helpful if your Dovecot binaries weren't stripped of debug symbols. I
could then ask for some more information from the core dumps with gdb.
On Wed, 2011-08
On 07/22/2011 01:02 PM, Kostas Zorbadelos wrote:
Hello,
since I saw no action on this, here is a newer update we discovered today.
After setting pop3_lock_session = no the core dumps went away.
We will leave it like that and watch it for the next few days. If we set
pop3_lock_session = yes, th
Greetings to all.
It's my first post to the list. We just completed a migration from qpopper to
dovecot
for our IMAP and POP3 services. We have a rather large mail environment
(we are the biggest provider in Greece).
So, here are the details:
- Keep getting errors like these in our production
Hi,
I am getting the following error trying to search a virtual folder.
May 26 11:35:29 xam dovecot: imap(egh): Panic: Trying to sync mailbox archive
with open transactions
May 26 11:35:29 xam dovecot: imap(egh): Error: Raw backtrace:
/usr/local/stow/dovecot-2.0.13/lib/dovecot/libdovecot.so.0(+
I have a set of automated tests for a system that can authenticate a
user by forwarding authentication details to a server. That is, in order
to authenticate a user, this system logs on to, e.g., an IMAP server
using the credentials supplied by the user to verify the correctness of
those credential
Timo Sirainen wrote:
On 7.1.2010, at 11.40, Stephan Bosch wrote:
Timo, are you going to change this in Dovecot or should I implement the
non-ASCII-only check in Sieve?
I don't think my function should do such checks. It might be wanted in some
situations that ASCII-only text gets encoded.
On 7.1.2010, at 11.40, Stephan Bosch wrote:
> Timo, are you going to change this in Dovecot or should I implement the
> non-ASCII-only check in Sieve?
I don't think my function should do such checks. It might be wanted in some
situations that ASCII-only text gets encoded.
Christian Rohmann wrote:
Hey folks,
allow me to bump this notice about a possible plain bug in the sieve
filter conversion.
Oh, sorry, this one got a bit backlogged.
According to RFC5230 section 4.3 this shouldn't happen:
The ":subject" parameter specifies a subject line to attach to any
Hey folks,
allow me to bump this notice about a possible plain bug in the sieve
filter conversion.
Regards
Christian
On 12/30/2009 01:37 PM, Roland Rosenfeld wrote:
> Hi!
>
> I noticed the following problem in dovecot-1.2-sieve-0.1.14:
> The following vacation causes trouble with the Subject
On Wed, 2009-11-11 at 20:11 -0800, Bryan Seitz wrote:
> Nov 11 23:11:20 triton dovecot: IMAP(sam): Panic: file
> istream-raw-mbox.c: line 578 (istream_raw_mbox_get_body_size):
> assertion failed: (rstream->body_offset != (uoff_t)-1)
Are you still seeing these? Wonder if these patches would at le
One of my users is causing dovecot to exit with signal 6 and generate a core
file. Seems to be in the mbox code.
Dovecot configuration:
# 1.2.4: /usr/local/etc/dovecot.conf
# OS: FreeBSD 7.2-STABLE amd64
protocols: imap imaps pop3s
listen(default): localhost:143
listen(imap): localhost:143
li
On Fri, 2008-12-26 at 11:10 +0100, Jörg Leis wrote:
> I tried using a "user_field = user" additionally as is written here:
> http://wiki.dovecot.org/Plugins/Expire. (Incidentally, there must be
> some mismatch in the documentation because the tables for v1.1 do not
> contain a field username). O
I did the same thing on the console and this is what happens:
bash-3.2$ openssl s_client -connect example.com:993
CONNECTED(0003)
[...]
Server certificate
-BEGIN CERTIFICATE-
[...]
-END CERTIFICATE-
[...]
CA
---
No client certificate CA names sent
---
SSL handshake has read 45
If the expire plugin is enabled and a new message (not in trash) is
deleted via IMAP in Mozilla Thunderbird (only tested in this client), it
will indefinately copy the message into the folder. Dovecot, meanwhile,
occupies itself with this:
r...@server:~# cat /var/log/dovecot.log
[...]
dovecot
Tue, 27 Nov 2007 14:15:29 +0200
Temats: Re: [Dovecot] Fwd: dovecot bug report (64 bit file offsets)
I agree that it’s kind strange behavior.
Also I’ve googled that exactly the same errors (unexpected eof
found … on file ‘UNOPENED’) were related to mysql.
But how mysql compile options m
On 2007-11-27 14:15:29 +0200, [EMAIL PROTECTED] wrote:
> I agree that it’s kind strange behavior.
>
> Also I’ve googled that exactly the same errors (unexpected eof
> found … on file ‘UNOPENED’) were related to mysql.
> But how mysql compile options might affect configuration file parsing
> and r
I agree that it’s kind strange behavior.
Also I’ve googled that exactly the same errors (unexpected eof
found … on file ‘UNOPENED’) were related to mysql.
But how mysql compile options might affect configuration file parsing
and read/write operations? (indexes for example) Does any one have
any i
er.ibm.com/iseries/v5r2/ic2924/index.htm?info/apis/pread64.htm
>
>
> Regards, Andris
> -- Forwarded message --
> No: [EMAIL PROTECTED]
> Kam: [EMAIL PROTECTED]
> Reply-To: [EMAIL PROTECTED]
> Date: Tue, 27 Nov 2007 10:45:23 +0200
> Temats: dovecot b
PROTECTED]
Kam: [EMAIL PROTECTED]
Reply-To: [EMAIL PROTECTED]
Date: Tue, 27 Nov 2007 10:45:23 +0200
Temats: dovecot bug report (64 bit file offsets)
Hello. I found 2 issues while using dovecot in the 64-bit system.
Before that I was using it on the Solaris 10 32bit and everything
worked fin
On Sat, 2007-09-08 at 12:36 -0400, Bill Cole wrote:
> The only way a program
> could know of the failure would be to trace the syslog configuration
> and look for whether messages are appearing, then take action
> (probably calling closelog() and openlog()) as needed.
I already sent a patch tha
On Sep 8, 2007, at 12:36 PM, Bill Cole wrote:
If you run them, you would also note that Postfix, BIND, INN, and
MIMEDefang all suffer from the same problem. In fact, I don't know
of any non-Apple daemon that does not have this problem. I believe
it is a bug in Apple's ground-up re-implement
At 9:54 AM -0400 9/8/07, Bruce Bodger wrote:
Timo,
Compiled and installed v1.0.4 this morning on OS X 10.4.10...
./configure --with-ssldir=/System/Library/OpenSSL --with-ssl=openssl
--with-notify=kqueue
...all is well.
I wonder if you remember the conversation that we had back on May
11.
Timo,
Compiled and installed v1.0.4 this morning on OS X 10.4.10...
./configure --with-ssldir=/System/Library/OpenSSL --with-ssl=openssl
--with-notify=kqueue
...all is well.
I wonder if you remember the conversation that we had back on May
11. This was the situation where after the syste
On Fri, 2007-05-11 at 10:57 -0400, Bruce Bodger wrote:
> Applied the patch, recompiled v1.1.0 using same line to config as
> before..
>
> ./configure --with-ssldir=/System/Library/OpenSSL --with-ssl=openssl
>
> Sorry... no change: does not fix it.
In that case there's nothing I can do. You'll
Applied the patch, recompiled v1.1.0 using same line to config as
before..
./configure --with-ssldir=/System/Library/OpenSSL --with-ssl=openssl
Sorry... no change: does not fix it.
Regards,
Bruce
On May 11, 2007, at 10:24 AM, Timo Sirainen wrote:
On Fri, 2007-05-11 at 09:26 -0400, Bruce Bo
On Fri, 2007-05-11 at 09:26 -0400, Bruce Bodger wrote:
> That would be accurate, Bill. I suspect the problem to be within
> dovecot since, as I mentioned earlier, neither uw-imap or other
> running daemons exhibit this problem.
That's probably because UW-IMAP and most other servers keep cre
On May 11, 2007, at 8:48 AM, Bill Cole wrote:
My system.log is rotated nightly followed by a pretty standard
syslogd restart...
if [ -f /var/run/syslog.pid ]; then kill -HUP $(cat /var/run/
syslog.pid | head -1);
That's not a restart, it is a 'hangup' signal that will make a
normal sysl
At 7:04 AM -0400 5/11/07, Bruce Bodger imposed structure on a stream
of electrons, yielding:
It appears as though dovecot doesn't like syslogd being restarted.
That's probably not it.
The whole point of syslog is for individual message sources (e.g.
dovecot) to not have to worry about where
Bruce Bodger wrote:
Thank you very much, Charles. I have resorted to that workaround
(restarting dovecot) which does, indeed, work. I just wanted to
document the situation since dovecot's behavior in this regard is unlike
other imap servers.
This thread mentioned only sending an HUP - or ma
Thank you very much, Charles. I have resorted to that workaround
(restarting dovecot) which does, indeed, work. I just wanted to
document the situation since dovecot's behavior in this regard is
unlike other imap servers.
Best regards,
Bruce
On May 11, 2007, at 8:27 AM, Charles Marcus w
Bruce Bodger wrote:
It appears as though dovecot doesn't like syslogd being restarted.
Hmmm... could this possibly be a fix, or at least a workaround?
www.dovecot.org/list/dovecot/2004-November/005219.html
--
Best regards,
Charles
It appears as though dovecot doesn't like syslogd being restarted.
My system.log is rotated nightly followed by a pretty standard
syslogd restart...
if [ -f /var/run/syslog.pid ]; then kill -HUP $(cat /var/run/
syslog.pid | head -1);
It appears from that point on dovecot no longer writes t
Please disregard the below reported bug report. I experienced the
same problem today and this time "sleep" was not involved.
I rotate mail.log weekly. This morning I noticed that dovecot was no
longer logging to mail.log although, again, it was running fine.
Is it possible that dovecot ha
Version: dovecot --version: 1.0.0
OS: OS X - 10.4.9
Architecture: G5 Dual/1.8Ghz
Had a fairly normal time correction last night:
May 9 20:56:56 G518X2 ntpd[219]: time set -1.212733 s
Which dovecot duly noted:
May 9 20:56:56 G518X2 dovecot: auth(default): Time just moved
backwards by 1 secon
On 28.3.2007, at 23.47, Erik van Oosten wrote:
When I try to open the attached mbox file I get the following
warning in Thunderbird 1.5.0.10:
'The current command failed. The e-mail server responded: Mailbox
isn't a valid mbox file'.
You have an extra empty line at the beginning of the file
64 matches
Mail list logo
