h table anti-dupe check would trigger abort. So
some changes in script comparison seem needed anyway. This is what I've
used in my approach. Return early from ext_include_execute_include and
opc_include_dump if sblock is NULL and change sieve_file_script_equals
to compare also
updated patch works
On 21/08/2023 22:29, Aki Tuomi wrote:
Hi!
Thanks for pointing this out, I've updated the patch, and made it bit
differently to ensure it works regardless where the key came from.
Aki
On 21/08/2023 23:04 EEST Michal Hlavinka wrote:
Hi,
this patch does not fi
how some keys are created, but does not affect those that are
loaded directly for example through PEM_read_bio_PUBKEY in:
dcrypt-openssl3.c:
dcrypt_openssl_load_public_key:2460:PEM_read_bio_PUBKEY
Cheers,
Michal Hlavinka
On 21/08/2023 09:08, Aki Tuomi wrote:
I think this should fix it https
EC_GROUP_free(ec_grp);
EC_POINT_free(point);
return ret;
}
and it fixed the issue, but there may be an easier way how to achieve that.
Cheers
Michal Hlavinka
___
dovecot mailing list -- dovecot@dovecot.org
To unsubscribe send an email to dovecot-le...@dovecot.org
Hard to say without reproducer and more information, but I think it was
a downstream bug. There was configuration error in 2.3.13-1 build. I'm
not sure how exactly it could cause this error, but I did not see any
report for the new build so far.
On 07. 01. 21 17:31, Marius Schwarz wrote:
Hi
On 04. 01. 21 13:02, Aki Tuomi wrote:
We are pleased to release v2.3.13. Please find it from locations below:
https://dovecot.org/releases/2.3/dovecot-2.3.13.tar.gz
https://dovecot.org/releases/2.3/dovecot-2.3.13.tar.gz.sig
Binary packages in https://repo.dovecot.org/
Docker images in https://hu
On 4/12/19 12:48 AM, Stephan Bosch wrote:
On 29/03/2019 10:23, Michal Hlavinka via dovecot wrote:
On 3/28/19 6:41 PM, Aki Tuomi via dovecot wrote:
On 28 March 2019 19:40 Michal Hlavinka via dovecot
wrote:
Hi,
when trying to build dovecot 2.3.5.1 pigeonhole testsuite crashes in
On 3/28/19 6:41 PM, Aki Tuomi via dovecot wrote:
On 28 March 2019 19:40 Michal Hlavinka via dovecot wrote:
Hi,
when trying to build dovecot 2.3.5.1 pigeonhole testsuite crashes in
Which version of pigeonhole are you using?
latest available - 0.5.5
Hi,
were you able to reproduce this problem? Do you need more information to
reproduce this?
Cheers,
Michal Hlavinka
On 3/12/19 3:29 PM, Michal Hlavinka wrote:
Hi,
thanks for the answer. I think your environment was not set up correctly
to reproduce this bug. I've retested with 2.3.5
EEDED
make: *** [Makefile:1190:
tests/extensions/editheader/deleteheader.svtest] Segmentation fault
(core dumped)
Let me know if you need more debug information.
Cheers
Michal Hlavinka
#0 0x7f9924faea8a in __strlen_sse2 () from /lib64/libc.so.6
#1 0x7f992514fdd2 in smtp_address_clon
On 3/12/19 10:18 PM, Timo Sirainen via dovecot wrote:
On 18 Sep 2018, at 17.10, Michal Hlavinka <mailto:mhlav...@redhat.com>> wrote:
Seems that for Global ACL directory, namespace prefix is not part of
the path, when looking for acl file.
Is there a reason you're using ACL dir
at condition would have different
result and bug would not trigger:
doveadm(testuser): Debug: acl vfile: reading file
/etc/dovecot/global-acls/pub/subfolder/.DEFAULT
For testing I use this acl configuration:
cat /etc/dovecot/global-acls/pub/.DEFAULT
user=testuser l
but as this acl file location
Hi,
any progress with this issue? Do you need more information to debug and
fix this?
Cheers
Michal Hlavinka
On 9/18/18 4:10 PM, Michal Hlavinka wrote:
Hi
tl;dr:
Seems that for Global ACL directory, namespace prefix is not part of the
path, when looking for acl file.
Long version:
We&#x
Hi
tl;dr:
Seems that for Global ACL directory, namespace prefix is not part of the
path, when looking for acl file.
Long version:
We're planning to update dovecot in next os update to 2.2.36 and while
going through regression testing, we found a problem with ACL
configuration combined with
we've seen SELinux reports from our users that dovecot tried to use something
that needs CAP_NET_ADMIN capability. Before enabling it, we would like to know
where it originated from. I've checked the sources, but was not able to find
anything that would require this capability. Do you know for
ervice (TOS) "IP_TOS"
* clear driver statistics;
* set promiscuous mode;
* enabling multicasting;
* use setsockopt(2) to set the following socket options:
SO_DEBUG, SO_MARK, SO_PRIORITY (for a priority outside the range 0 to
6),SO_RCVBUFFORCE, and SO_SNDBUFFORCE
Cheers,
Michal Hlavinka
patch) and all seems fine.
Cheers,
Michal Hlavinka
diff -up dovecot-2.2.24/src/auth/auth-settings.c.gidlimit dovecot-2.2.24/src/auth/auth-settings.c
--- dovecot-2.2.24/src/auth/auth-settings.c.gidlimit 2016-04-26 15:01:20.0 +0200
+++ dovecot-2.2.24/src/auth/auth-settings.c 2016-06-27 16:5
s in LE
format (0x017f), but it gets address in BE format(0x7f01). Patch
in attachment should fix this issue.
Regards
Michal Hlavinka
--- dovecot-2.2.18/src/lib/test-net.c 2015-05-03 05:12:11.0 -0400
+++ dovecot-2.2.18/src/lib/test-net.c 2015-06-17 03:18:02.571420401
On 05/13/2015 08:36 PM, Timo Sirainen wrote:
http://dovecot.org/releases/2.2/dovecot-2.2.17.tar.gz
http://dovecot.org/releases/2.2/dovecot-2.2.17.tar.gz.sig
A few director improvements since RC2, some fixes and a new plugin:
http://wiki2.dovecot.org/Plugins/QuotaClone
Hi,
there seems to be so
On 11/02/2013 02:31 PM, Timo Sirainen wrote:
On 1.10.2013, at 21.14, Michal Hlavinka wrote:
we found a bug in ssl-params. It calls openssl DH generator for 512 and 1024
bits, but in FIPS mode, openssl won't generate anything for less than 1024, so
it fails with:
error:0506A06E:D
On 11/02/2013 02:31 PM, Timo Sirainen wrote:
On 1.10.2013, at 21.14, Michal Hlavinka wrote:
we found a bug in ssl-params. It calls openssl DH generator for 512 and 1024
bits, but in FIPS mode, openssl won't generate anything for less than 1024, so
it fails with:
error:0506A06E:D
an do the trick. Just change
ssl-params/ssl-params-openssl.c:generate_dh_paramaters:39
- if (dh == NULL) {
+ if (1 || dh == NULL) {
and when you run ssl-params (with no ssl-parameters.dat), it hangs.
Let me know if you need more information.
Regards,
Michal Hlavinka
On 05/04/2012 08:54 PM, Timo Sirainen wrote:
On 16.4.2012, at 12.55, Michal Hlavinka wrote:
I wrote simple patch that close the extra sockets. It's tested and works fine.
You'll maybe want to move that function to different place and/or change
wording of error messages.
I commi
On 03/15/2012 03:05 PM, Timo Sirainen wrote:
On Thu, 2012-03-15 at 14:34 +0100, Michal Hlavinka wrote:
What exactly should happen when
dovecot.conf does not match dovecot.socket configuration?
Dovecot's systemd code was written by one of you Redhat guys. I had some
similar thoughts w
On 03/15/2012 03:05 PM, Timo Sirainen wrote:
On Thu, 2012-03-15 at 14:34 +0100, Michal Hlavinka wrote:
What exactly should happen when
dovecot.conf does not match dovecot.socket configuration?
Dovecot's systemd code was written by one of you Redhat guys. I had some
similar thoughts w
Hi all,
dovecot supports systemd socket activation. Together with standard unit
activation (like old sysv init script), there are two ways how to
configure dovecot(only interface:port, not whole configuration). This
can result in situation where those configurations does not say the
same. Que
On 02/17/2012 06:17 PM, e-frog wrote:
On 17.02.2012 18:09, wrote Marc Perkel:
Just wondering if there's any upgrade docs for 2.1 online?
http://wiki2.dovecot.org/Upgrading/2.1
I know about one Fedora user who had to also change:
namespace {
inbox=yes
...
}
to namespace inbox {
a
On Friday, April 15, 2011 08:31:20 Michal Hlavinka wrote:
> On Tuesday, April 12, 2011 21:27:11 k b wrote:
> > Will check as soon as there is a package for fedora 14.
> >
> Usually there is updated dovecot package in updates-testing
> repository next working day after relea
On Tuesday, April 12, 2011 21:27:11 k b wrote:
> Will check as soon as there is a package for fedora 14.
>
Usually there is updated dovecot package in updates-testing
repository next working day after release, but it seems I've built
new package, but forgot to push it to repository.
I've pushed
On Saturday, March 26, 2011 01:02:04 Timo Sirainen wrote:
> On Fri, 2011-03-25 at 09:57 +0100, Michal Hlavinka wrote:
> > Hi,
> >
> > one Fedora user sees regression in dovecot 2.0.11 (compared to prev.
> > 2.0.9). There was change recently:
> > http:
Hi,
one Fedora user sees regression in dovecot 2.0.11 (compared to prev. 2.0.9).
There was change recently:
http://hg.dovecot.org/dovecot-2.0/diff/bc77e80947c0/src/lib-settings/settings-parser.c
which adds call to settings_find_key_nth(, &parent_def,...). This function
can set parent_def to
> > > 2) Does dovecot.service required to use dovecot lmtp, or it may be
> > > stoppped (i not often check my mails via client, but have many incoming
> > > mails some time)
> >
> > I don't understand. If you want LMTP enabled, you also need to list its
> > socket to dovecot.socket. LMTP is execut
On Thursday 10 of February 2011 02:54:39 Timo Sirainen wrote:
> On Mon, 2011-01-10 at 14:08 +0100, Michal Hlavinka wrote:
> > On Tuesday, January 04, 2011 00:47:16 Timo Sirainen wrote:
> > > On 3.1.2011, at 19.15, Michal Hlavinka wrote:
> > > > Shouldn't mkdir an
On Thursday 13 of January 2011 13:21:26 Timo Sirainen wrote:
> http://dovecot.org/releases/2.0/dovecot-2.0.9.tar.gz
> http://dovecot.org/releases/2.0/dovecot-2.0.9.tar.gz.sig
>
> I'm still lagging behind in my email, but I guess it's time to release
> v2.0.9 anyway.
I've noticed that for some unk
On Tuesday, January 04, 2011 00:47:16 Timo Sirainen wrote:
> On 3.1.2011, at 19.15, Michal Hlavinka wrote:
> > Shouldn't mkdir and chown make together a transaction? When it fails just
> > for first time, it confuses some admins (this is the reason why I'm
> > getti
Hi Timo,
I'm facing problem with mail group privileges (again), because I've heard a
few complains. I was looking at old emails and list archives, but still does
not know the answer. In default (fedora/rhel) setup mail group and 0660
privileges are used. This means that when
mail_location
On Friday, July 16, 2010 16:56:10 Timo Sirainen wrote:
> > #0 mailbox_get_storage (box=0x0) at mail-storage.c:802
> > No locals.
> > #1 0x7ff182bace34 in act_store_start (action=,
> >
> > aenv=0x153c848, tr_context=0x153c988) at sieve-actions.c:405
>
> Looks like the cleanest fix is to
Hi,
my colleague found a crash in dovecot, while he was experimenting with
configuration for using dovecot with fetchmail. Backtrace can be found here:
http://pastebin.com/XQZqYBpH
Let me know if you need any other info.
Michal
# doveconf -n
# 20100716 (26936bdd3801+): /etc/dovecot/dovecot.con
On Friday 09 of July 2010 03:07:00 Timo Sirainen wrote:
> On 8.7.2010, at 19.57, Ray Kirby wrote:
> > Jul 08 12:35:08 master: Error: service(auth): child 5516 killed with
> > signal 11 (core dumped)
>
> Could you get a gdb backtrace? It said core dumped, so there should be one
> in base_dir (dovec
On Wednesday, July 07, 2010 13:27:41 Timo Sirainen wrote:
> On 7.7.2010, at 11.42, Michal Hlavinka wrote:
> >> I was just updating dovecot to nightly snapshot for Fedora rawhide, but
> >> got following issue:
> >> fts-backend-solr.c: In function 'fts_box_name_ge
On Wednesday, July 07, 2010 12:26:54 Michal Hlavinka wrote:
> Hi,
>
> I was just updating dovecot to nightly snapshot for Fedora rawhide, but got
> following issue:
> fts-backend-solr.c: In function 'fts_box_name_get_root':
> fts-backend-solr.c:60: error: 'NAMESP
Hi,
I was just updating dovecot to nightly snapshot for Fedora rawhide, but got
following issue:
fts-backend-solr.c: In function 'fts_box_name_get_root':
fts-backend-solr.c:60: error: 'NAMESPACE_FLAG_INBOX' undeclared (first use in
this function)
fts-backend-solr.c:60: error: (Each undeclared id
On Wednesday, May 26, 2010 18:39:26 Timo Sirainen wrote:
> On Thu, 2010-05-06 at 22:15 +0200, Frode Nordahl wrote:
> > Hello,
> >
> > The following IMAP command crashes the latest version of Dovecot
> > (regardless of format or contents of mailbox): . THREAD REFS US-ASCII
> > ALL
>
> THREAD works
On Friday 14 of May 2010 17:42:25 Timo Sirainen wrote:
> On 14.5.2010, at 14.10, Michal Hlavinka wrote:
> > Looking at strace output, dovecot tries to use socket on
>
> > /etc/dovecot/dovecot.conf which is regular file and no socket:
> Yeah, it was less code to do it th
gular file
dovecot.conf as socket? There is probably some magic in it (from my pov) which
I don't understand.
Regards,
Michal Hlavinka
On Wednesday, May 05, 2010 09:24:13 Michal Hlavinka wrote:
> Hi,
>
> yesterday lib-sql got driver-sqlpool with some changes in Makefile.am, but
> these changes in makefile works only for --with-sql=yes. When
> --with-sql=plugin is used build fails with:
>
> libtool --tag
,-rpath -Wl,/usr/lib64/dovecot
../../src/lib-sql/.libs/libsql.a(sql-api.o): In function `sql_init':
/home/mhlavink/cvsf/dovecot/devel/dovecot-2.0.beta4/src/lib-sql/sql-api.c:70:
undefined reference to `driver_sqlpool_init'
Regards,
Michal Hlavinka
Hi,
there are few documents on the wiki ( http://wiki.dovecot.org/Upgrading )
about migration between old versions. Is there any document for 1.2->2.0 or
list of configuration changes (especially the missing ones) in 2.0?
Regards,
Michal Hlavinka
t;, 14) == 0
message decoder .. :
FAILED
1 / 1 tests failed
Regards,
Michal Hlavinka
> Hi list
> I'm still trying to get managesieve running on centos x64 with the
> packages from atrpm. I used the 1.2.9 rpms (also rebuilt with the
> lastet managesieve patches) but I'm always getting the same error when
> trying to compile managesieve
try to rebuild official src.rpm package for F
On Tuesday 20 October 2009 21:49:45 Timo Sirainen wrote:
> On Tue, 2009-10-20 at 17:10 +0200, Michal Hlavinka wrote:
> > On Tuesday 13 October 2009 19:10:15 Timo Sirainen wrote:
> > > On Tue, 2009-10-13 at 17:26 +0200, Michal Hlavinka wrote:
> > > > would it be possib
On Tuesday 13 October 2009 19:10:15 Timo Sirainen wrote:
> On Tue, 2009-10-13 at 17:26 +0200, Michal Hlavinka wrote:
> > would it be possible to:
> > a) send content of imap_capability in greeting message? It should be
> > user's responsibility if it's set to somethi
On Monday 19 October 2009 23:55:20 Timo Sirainen wrote:
> Can someone find an OS where the attached program doesn't work? It
> should print "success". So far tested for success: Linux 2.6, Solaris
> 10, FreeBSD 7.2, OpenBSD 4.2.
>
on Fedora 11 (Linux 2.6.30.9 64bit) it failed 3 times (of 200 runs)
On Tuesday 06 October 2009 15:35:20 Michal Hlavinka wrote:
> On Thursday 01 October 2009 15:15:09 Timo Sirainen wrote:
> > On Thu, 2009-10-01 at 14:59 +0200, Michal Hlavinka wrote:
> > > Hi,
> > >
> > > one Fedora user complains about not some troubles a
On Thursday 01 October 2009 15:15:09 Timo Sirainen wrote:
> On Thu, 2009-10-01 at 14:59 +0200, Michal Hlavinka wrote:
> > Hi,
> >
> > one Fedora user complains about not some troubles after update to dovecot
> > 1.2. He suspects wrong capability information give
=PLAIN
is this expected? Do you think capability information in hello message really
can befool clients?
Regards,
Michal Hlavinka
On Monday 07 September 2009 17:48:13 Timo Sirainen wrote:
> On Sep 7, 2009, at 6:01 AM, Michal Hlavinka wrote:
> > On Monday 07 September 2009 04:18:06 Timo Sirainen wrote:
> >> On Tue, 2009-09-01 at 16:57 +0200, Michal Hlavinka wrote:
> >>> Would it be possible
On Monday 07 September 2009 04:18:06 Timo Sirainen wrote:
> On Tue, 2009-09-01 at 16:57 +0200, Michal Hlavinka wrote:
> > Would it be possible to use syslog for error output sooner or there are
> > any cons?
>
> http://hg.dovecot.org/dovecot-1.2/rev/eb4ea59e4657
>
Th
onnections. ipv6 support was disabled. After server reboot,
dovecot failed to start, producing
dovecot: Fatal: listen(::, 143) failed: Address already in use
only to stderr, nothing in logs.
Would it be possible to use syslog for error output sooner or there are any
cons?
Regards,
Michal Hlavinka
..@%",
MEMBER(username_translation) "",
MEMBER(username_format) "",
MEMBER(master_user_separator) "",
Would it be possible or is there any reason why this is a wrong idea?
Regards,
Michal Hlavinka
features option is set:
...
make[2]: Entering directory
`/home/mihl/myroot/job/cvsf/dovecot/devel/dovecot-1.2.3/dovecot-1.2-
sieve-0.1.11'
make[2]: *** No rule to make target `doc/man/sieve-filter.1', needed by `all-
am'. Stop.
without this it works.
Regards,
Michal Hlavinka
> > > I guess this fixes it again:
> > > http://hg.dovecot.org/dovecot-1.2/rev/f4ff64dd79a9
> >
> > We've tested dovecot with all four available patches (it means up to date
> > mech-gssapi.c ) and it wokrs.
>
> I've been talking with the main Heimdal guy and he thinks that kind of
> checking is sc
On Wednesday 08 July 2009 03:06:08 Timo Sirainen wrote:
> On Tue, 2009-07-07 at 20:29 -0400, Timo Sirainen wrote:
> > On Tue, 2009-07-07 at 20:20 -0400, Timo Sirainen wrote:
> > > On Wed, 2009-06-24 at 15:38 +0200, Michal Hlavinka wrote:
> > > > dovecot: auth(def
On Wednesday 08 July 2009 03:06:08 Timo Sirainen wrote:
> On Tue, 2009-07-07 at 20:29 -0400, Timo Sirainen wrote:
> > On Tue, 2009-07-07 at 20:20 -0400, Timo Sirainen wrote:
> > > On Wed, 2009-06-24 at 15:38 +0200, Michal Hlavinka wrote:
> > > > dovecot: auth(def
st using new dovecot sieve plugin. It's version 0.1.6 so I
wonder how much stable and feature complete it is comparing to the old sieve
plugin? Which one is better and why?
Regards,
Michal Hlavinka
On Thursday 02 July 2009 07:31:00 Timo Sirainen wrote:
> On Jul 2, 2009, at 7:25 AM, Michal Hlavinka wrote:
> > Is gssapi auth problem fixed in this version or not yet?
>
> If I fix it for you, I break it for someone else. I'd need to find out
> what exacly is that patch
Is gssapi auth problem fixed in this version or not yet?
Regards,
Michal Hlavinka
On Monday 29 June 2009 12:31:53 Edgar Fuß wrote:
> The ManageSieve patch for rc5 applies to rc7, but doesn't build.
>
> The problem is obvoiusly that
> dovecot-managesive/src/managesieve-login/client.c tests:
>
> #if CLIENT_LOGIN_IDLE_TIMEOUT_MSECS >= AUTH_REQUEST_TIMEOUT*1000
> # error client idl
. : static prefetch passwd passwd-file
checkpassword ldap (plugin) sql nss
Building with password lookup modules : passwd passwd-file shadow pam
checkpassword ldap (plugin) sql
Building with SQL drivers : pgsql mysql sqlite (plugins)
There's no other difference in configure output.
Regards,
Michal Hlavinka
On Thursday 25 June 2009 06:54:48 Michal Hlavinka wrote:
> On Wednesday 24 June 2009 17:15:31 Timo Sirainen wrote:
> > On Jun 24, 2009, at 9:38 AM, Michal Hlavinka wrote:
> > > we're facing problem where dovecot 1.2rc5 is not able to
> > > authenticate user v
On Wednesday 24 June 2009 17:15:31 Timo Sirainen wrote:
> On Jun 24, 2009, at 9:38 AM, Michal Hlavinka wrote:
> > we're facing problem where dovecot 1.2rc5 is not able to
> > authenticate user via
> > gssapi. (I'm forwarding information from red hat's bugzilla)
auth_debug=yes can be found here:
https://bugzilla.redhat.com/attachment.cgi?id=348710
Regards,
Michal Hlavinka
On Thursday 18 June 2009 07:50:22 Timo Sirainen wrote:
> On Jun 18, 2009, at 1:43 AM, Michal Hlavinka wrote:
> > On Thursday 18 June 2009 07:29:13 Timo Sirainen wrote:
> >> On Jun 18, 2009, at 1:24 AM, Michal Hlavinka wrote:
> >>> So I'm curious if there is pl
On Thursday 18 June 2009 07:29:13 Timo Sirainen wrote:
> On Jun 18, 2009, at 1:24 AM, Michal Hlavinka wrote:
> > So I'm curious if there is plan to release
> > managesieve for rc5 dovecot?
>
> rc3's managesieve patch works just fine with rc5.
ok, thanks for the info
eed dovecot, sieve and managesieve sources, but the last one
is still missing... For rc4 Stephan Bosch said there will be no managesieve
release because some problems. So I'm curious if there is plan to release
managesieve for rc5 dovecot?
Regards,
Michal Hlavinka
y bug
> and possibly exploitable. MODSEQ code is new in v1.2, so v1.1 and older
> releases are not affected.
>
Hi,
I can't get it compiled when I use --with-ldap=plugin for configure, is this
expected? ("[Dovecot] problem with ldap" thread contains details).
Regards,
Michal Hlavinka
es :
> passwd passwd-file shadow pam checkpassword ldap (plugin)
It seems to me that configure/Makefile is not working or am I doing something
wrong?
Regards,
Michal Hlavinka
Timo Sirainen wrote:
On Tue, 2009-01-27 at 14:20 -0600, Eric Rostetter wrote:
Quoting Timo Sirainen :
I've already written some unit tests in src/tests/. I don't really care
if you continue them the way I started or use some other toolset. And
What branch(es) should I write the
Timo Sirainen wrote:
Good news, everyone!
The most annoying bug in v1.1 series should finally be fixed by these
patches:
http://hg.dovecot.org/dovecot-1.1/rev/f28348c5201e
http://hg.dovecot.org/dovecot-1.1/rev/74c47fcfb420
http://hg.dovecot.org/dovecot-1.1/rev/a28520d26b5a
http://hg.dovecot.org
Hi,
we're trying to solve CVE-2008-4870 = rhbz#436287 = dovecot.conf is
world readable - possible password exposure.
This problem seems to be little more complicated than we thought.
dovecot.conf can contain passphrase for ssl key, which is available for
everyone since dovecot.conf has world
Hi!
I would like to know if there is a plan to fix CVE-2008-4578 bug (ACL
vs. "parent/child/child" mailboxes) in 1.0 branch?
I've tried to backport patch from 1.1.4 but with no success.
Thanks
Michal
81 matches
Mail list logo
