[Dovecot] authenticate plain and utf-8 with special chars

2012-07-31 Thread Matthias Lay
Hi,

I experienced some problems with authenticate. 

seems like the mail clients like thunderbird send their base64 string in
ISO- encoding, which doesnt work on my dovecot setup if there are some
Umlauts

I am not able to login as 
 ömer for example


If I authenticate from telnet with an base64 encoded utf-8 string like
 'echo -en \0ömer\0Start12 |base64'
everything works fine


-dovecot 2.1.7
-userdb is an openldap which queries an windows AD so results are in
utf8, I think.

any clues at which point to fix this are welcome ;)





[Dovecot] openbsd + dovecot not downloading emails

2012-07-31 Thread GVG GVG
Dear list members,

first my doveconf -n output:

--
# 2.0.17 (684381041dc4+): /etc/dovecot/dovecot.conf
# OS: OpenBSD 5.1 i386  ffs
mail_location = mbox:/var/mail:INBOX=/var/mail/%u
mbox_write_locks = fcntl
mmap_disable = yes
passdb {
  driver = bsdauth
}
ssl_cert = /etc/ssl/dovecotcert.pem
ssl_key = /etc/ssl/private/dovecot.pem
userdb {
  driver = passwd
}
protocol imap {
  imap_client_workarounds = delay-newmail tb-extra-mailbox-sep tb-lsub-flags
}
protocol pop3 {
  pop3_client_workarounds = outlook-no-nuls oe-ns-eoh
}
---

when using evolution from a remote linux machine, and try to download/fetch
the e-mails from my e-mail server, I get a pop-up window indicationg that
is trying to do so (after I did sign-in with my user-id and password!). The
problem is taht this process stays there forever and no messages get
downloaded!

Thanks for your help

Tim


Re: [Dovecot] [GoLugTech] thunderbird

2012-07-31 Thread Steve Litt
On Mon, 30 Jul 2012 23:43:49 -0400, Chris Clifton said:
 Hi all,
 
 Do I read it correctly that thunderbird funding is being cut back by
 mozilla?
 
 I'm not on a linux mail client yet, wondering what others are using in
 ubuntu for mail client, I need multiple pop3 profiles, good html
 support, fast indexing and searching, solid gpg support and all the
 usual bells and whistles. Evolution seems to be in pretty widespread
 use. Wondering what others like and use heavily.

Hi Chris,

I tried Evolution once about 10 years ago, and it was too complex and
comprehensive for me to understand.

The centerpiece of my emailing activity is a Dovecot IMAP server on my
daily driver desktop computer. The way it works is that fetchmail grabs
all my email from my various POP accounts, pushes them through
procmail, whose filters distribute the emails into the proper Dovecot
IMAP directories (and also /dev/null stuff from PITA people). As far as
I know, my Dovecot has never malfunctioned.

Currently, I access my local IMAP with Claws-Mail. Claws is capable,
robust, and does what I need without being an ecological disaster like
Kmail or intermittently pig slow like Thunderbird. Claws' configuration
is difficult to wrap your head around, and very much in need of better
documentation, but so far, thanks to a knowledgeable and helpful user
community on the mailing list, I've been able to make almost every
config change to speed my interaction with Claws.

My one complaint about Claws that can't be cured by documentation is
its propensity to do everything with one process and one thread, so
while you're sending an email or scanning your IMAP, you cannot look at
other emails. I also wanted their powerful search to be able to recurse
folders, and when nobody volunteered to do it, I tried to do it myself.
Making something recursive sounds like simply a matter of writing a
simple loop around the existing functionality, but what I found out is,
at least this part of Claws was written in a manner resembling MVC,
with process and UI completely commingled. For that reason I was unable
to make it recursive.

I have Claws on my daily driver and all my laptops. Thanks to Kevin
Korb's instructions, on the laptops I was able to tell Claws its IMAP
server was at 127.0.0.1. I pinholed my OpenBSD/pf firewall appliance to
port forward incoming ssh to my daily driver. Then, on my laptop, if
I'm at home and on the LAN, I run the following inhouse.sh

sudo ssh -NTL 993:127.0.0.1:993 slitt@192.168.1.88

While on the road I run this travelling.sh:

sudo ssh -NTL 993:127.0.0.1:993 slitt@99.99.99.99

In the preceding, 99.99.99.99 is the IP address my ISP gave me, while
192.168.1.88 is the LAN IP address of my daily driver. This makes
127.0.0.1:993 get pushed through ssh to the ISP supplied internet
address or the LAN address.

So I can use Claws to interact with my daily driver desktop anywhere I
go.

I'm extremely pleased with this setup. My IMAP tree is trivial to back
up, and gives me the assurance that if Claws-Mail ever commits suicide
the way Kmail did when they went to Kmail2, I can simply plug in a
different IMAP enabled email client. Meanwhile, Claws gives me an
extremely productive and easy to use user interface. With my
client/server setup, I no longer need to copy all my email too and from
the laptop when travelling.

I'm going to present on my remote use of my daily driver desktop
Dovecot, using Claws and ssh, at tomorrow night's GoLUG meeting.

HTH,

SteveT

Steve Litt*  http://www.troubleshooters.com/
  *  http://twitter.com/stevelitt
Troubleshooting Training  *  Human Performance



Re: [Dovecot] openbsd + dovecot not downloading emails

2012-07-31 Thread GVG GVG
here is my new doveconf-n:

--
# 2.0.17 (684381041dc4+): /etc/dovecot/dovecot.conf
# OS: OpenBSD 5.1 i386
auth_debug_passwords = yes
auth_verbose = yes
auth_verbose_passwords = sha1
mail_debug = yes
mail_location = mbox:~/mail:INBOX=/var/mail/%u
mbox_write_locks = fcntl
mmap_disable = yes
passdb {
  driver = bsdauth
}
ssl_cert = /etc/ssl/dovecotcert.pem
ssl_key = /etc/ssl/private/dovecot.pem
userdb {
  driver = passwd
}
verbose_ssl = yes
protocol imap {
  imap_client_workarounds = delay-newmail tb-extra-mailbox-sep tb-lsub-flags
}
protocol pop3 {
  pop3_client_workarounds = outlook-no-nuls oe-ns-eoh
}


now the retrieving pop-up window gets closed almost immediately but no
messages get downloaded! Also the detailed debigging doesn't show any extra
info in the log! The only info that I still get is:

--
Jul 31 15:16:23 [myserver] dovecot: imap-login: Login: user=user,
method=PLAIN, rip=x.x.x.x, lip=x.x.x.x, mpid=26976, TLS
---

Thanks

On Tue, Jul 31, 2012 at 7:53 PM, Pascal Volk 
user+dove...@localhost.localdomain.org wrote:

 On 07/31/2012 05:35 PM GVG GVG wrote:
  Dear list members,
 
  first my doveconf -n output:
 
  --
  # 2.0.17 (684381041dc4+): /etc/dovecot/dovecot.conf
  # OS: OpenBSD 5.1 i386  ffs
  mail_location = mbox:/var/mail:INBOX=/var/mail/%u
  mbox_write_locks = fcntl
  mmap_disable = yes
  passdb {
driver = bsdauth
  }
  ssl_cert = /etc/ssl/dovecotcert.pem
  ssl_key = /etc/ssl/private/dovecot.pem
  userdb {
driver = passwd
  }
  protocol imap {
imap_client_workarounds = delay-newmail tb-extra-mailbox-sep
 tb-lsub-flags
  }
  protocol pop3 {
pop3_client_workarounds = outlook-no-nuls oe-ns-eoh
  }
  ---
 
  when using evolution from a remote linux machine, and try to
 download/fetch
  the e-mails from my e-mail server, I get a pop-up window indicationg that
  is trying to do so (after I did sign-in with my user-id and password!).
 The
  problem is taht this process stays there forever and no messages get
  downloaded!

 Your mail_location setting looks a little bit strange to me.
 I'd suggest to use: mail_location = mbox:~/mail:INBOX=/var/mail/%u

 If that shouldn't solve the issue, make Dovecot more verbose:
 mail_debug = yes


 Regards,
 Pascal
 --
 The trapper recommends today: c01dcafe.1221...@localdomain.org



Re: [Dovecot] [GoLugTech] thunderbird

2012-07-31 Thread Timo Sirainen
On Tue, 2012-07-31 at 14:29 -0400, Steve Litt wrote:

 My one complaint about Claws that can't be cured by documentation is
 its propensity to do everything with one process and one thread, so
 while you're sending an email or scanning your IMAP, you cannot look at
 other emails. I also wanted their powerful search to be able to recurse
 folders, and when nobody volunteered to do it, I tried to do it myself. 

You could also do this on Dovecot side by creating a virtual folder out
of your all folders and do the search there:
http://wiki2.dovecot.org/Plugins/Virtual




Re: [Dovecot] maildir_copy_with_hardlinks on v.2.0.19

2012-07-31 Thread Timo Sirainen
On 30.7.2012, at 13.36, mailing list subscriber wrote:

 userdb {
 driver = passwd
 }
 
 Looks like you're using system users. Each mail then needs to be written 
 using different permissions, so hard linking can't work.
 
 
 I am afraid this is incorrect:

The code that determines if Dovecot even attempts to use hard links is in:

http://hg.dovecot.org/dovecot-2.1/file/f74557336910/src/lib-storage/mail-copy.c#l101

So you could make it work if you chmod 0770 all of your maildir directories and 
chgrp them to the same group.



Re: [Dovecot] Problem with running two dovecot instances on the same server

2012-07-31 Thread Timo Sirainen
On 30.7.2012, at 10.24, Alexandr Sabitov wrote:

 Jul 26 11:51:20 dovecot-test-1 dovecot: director: Error: Empty server list
 Jul 26 11:51:20 dovecot-test-1 dovecot: director: Fatal: Invalid value for 
 director_mail_servers setting
 Jul 26 11:51:20 dovecot-test-1 dovecot: master: Error: service(director): 
 command startup failed, throttling for 60 secs
 Director process gets started for the backend installation where it of 
 course isn't configured. My guess: doveadm (from some cronjob?) is 
 attempting to connect to /var/run/dovecot/director-admin instead of 
 /var/run/dovecot-proxy/director-admin.
 
 Hi Timo,
 
 Thanks for reply.
 
 That's right:  strace shows that director process is a child of non-proxy 
 dovecot (pid 7089) which is should not be happening. And I don't have any 
 cronjob running.
 
 Might we need just update dovecot wiki that if someone runs dovecot and 
 dovecot-proxy then they have to be installed in different folders as
 separate binaries otherwise one instance cross over another somehow 
 magically. :)

I have several such setups and they don't have this problem, so you've managed 
to create something that tries to connect there. One easy way to make sure this 
never happens is to just disable the socket:

service director {
  unix_listener director-admin {
mode = 0
  }
}



Re: [Dovecot] Help needed with configuration giving dict error No such file

2012-07-31 Thread Timo Sirainen
On 29.7.2012, at 22.09, Ajax wrote:

 file_dotlock_open(/var/vmail/XXX.com/jax/home/maildir/dovecot-quota)
 failed: No such file or directory

The /var/vmail/XXX.com/jax/home/maildir/ directory doesn't exist.

   quota = dict:user::file:%h/maildir/dovecot-quota

This should be %h/dovecot-quota. I'm not entirely sure if Dovecot autocreates 
the home directory.



Re: [Dovecot] BUG: LDAP extension failed to authenticate if 'base' DN configuration is empty ''

2012-07-31 Thread Timo Sirainen
On 29.7.2012, at 15.00, Hendy Irawan wrote:

 We have several context entries and want to authenticate a user against the
 whole LDAP server.
 
 However when setting base =  Dovecot LDAP fails with following message:
 
 Jul 29 10:50:59 nitik1 dovecot: auth: Error:
 ldap(he...@staging.member.berbatik.com,127.0.0.1):
 ldap_search((virtualMail=he...@staging.member.berbatik.com)) failed: Other
 (e.g., implementation specific) error

Does the attached patch help? If yes, I'll add it to next version.



diff
Description: Binary data


Re: [Dovecot] Dovecot Director behaviour when two or more fail

2012-07-31 Thread Timo Sirainen
On 29.7.2012, at 7.30, Dominic Malolepszy wrote:

 On the Dovecot wiki here http://wiki2.dovecot.org/Director it says the 
 following:
 
 The directors are going to connect to each others in a ring. For example if 
 you have servers called A, B and C, director will create connections A-B, 
 B-C and C-A
 
 I am curious what happens in a scenario when two or more directors fail and 
 potentially break a ring in half, do the active directors detect this and 
 reform a new ring?

Directors always attempt to create the ring as fully as they can, doesn't 
matter how many servers die.



Re: [Dovecot] fts solr : out of memory

2012-07-31 Thread Timo Sirainen
On 29.7.2012, at 2.20, to...@starbridge.org wrote:

 it seems ok now with memory problem, but an other error shows up:
 
 doveadm -v index -u c...@spamguard.fr INBOX
 doveadm(c...@spamguard.fr): Info: INBOX: Caching mails seq=1..59363
 21500/59363doveadm(c...@spamguard.fr): Error: fts_solr: Invalid XML
 input at line 1: mismatched tag

No idea. You can reproduce this? What does it log with this patch? 
http://hg.dovecot.org/dovecot-2.1/rev/817b69b2b21f

 doveadm(c...@spamguard.fr): Panic: file solr-connection.c: line 545
 (solr_connection_post_more): assertion failed: (maxfd = 0)

I guess related to the previous error, but I'm not really sure why it happens 
either.



Re: [Dovecot] ntlm auth / usernames with white spaces

2012-07-31 Thread Timo Sirainen
On 27.7.2012, at 17.32, Matthias Lay wrote:

 we experienced some problems with white spaces in usernames (this was
 NOT our idea) authenticated bei dovecots ntlmhelper
 
 I wrote a little patch fixing this, if you would like.
 
 patch applies on 2.1.7 and 2.1.8 (tested those)

Does this simpler patch also work?


diff
Description: Binary data


Re: [Dovecot] Empty argument in executable = script ... service

2012-07-31 Thread Timo Sirainen
On 26.7.2012, at 14.26, Florian Wagner wrote:

 what is the point of adding an empty argument to the command called by
 the executable = script ... setting in a service? I guess that would
 be the call to array_append_space at line 57 in util/script.c.

Fixed: http://hg.dovecot.org/dovecot-2.1/rev/1d9d799a2efc




Re: [Dovecot] mdbox with separate index location results in many emptydirectories

2012-07-31 Thread Timo Sirainen
On 20.7.2012, at 15.46, Jürgen Obermann wrote:

 mail_location = mdbox:/dovecot/storage%h:INDEX=/dovecot/index%h
..
 As expected the index files now reside under /dovecot/index%h/mailboxes/, but 
 the path /dovecot/storage%h/mailboxes/ also contains for every user and every 
 mailbox name a directory which in turn contains only an empty subdirectory 
 'dbox-Mails'.  These directories seem to me to sum up to a lot of unneded 
 infrstructure objects. I tried to delete these empty directories, but dovecot 
 did not like that and stopped to work. So somehow they are necessary, in 
 spite of setting a separate index file location.
 
 So my question is, could these empty directories be avoided, perhaps in a 
 future version?

No plans to do it. The empty directories under /dovecot/storage are used to 
find out what mailboxes exist. And in some setups those directories aren't 
always empty (e.g. dovecot-acl files go there).



Re: [Dovecot] FETCH Error

2012-07-31 Thread Timo Sirainen
On 18.7.2012, at 15.49, Micha Krause wrote:

 Jul 18 14:36:09 mailint1 dovecot: imap(micha): Error: FETCH [] for mailbox 
 INBOX.Sent UID 166 got too little data: 120860 vs 120867
 Jul 18 14:36:09 mailint1 dovecot: imap(micha): Error: Maildir filename has 
 wrong W value: 
 /var/mail/mailboxes/users/micha/.Sent/cur/1240390202.Rfd9752b8Q14135.mail-alt,S=280003,W=284007:2,S
 Jul 18 14:36:09 mailint1 dovecot: imap(micha): Error: Corrupted index cache 
 file /var/mail/mailboxes/users/micha/.Sent/dovecot.index.cache: Broken 
 virtual size for mail UID 166

All of those numbers seem wrong..

 so far i tried deleting:
 
 /var/mail/mailboxes/users/micha/.Sent/dovecot.index.cache
 /var/mail/mailboxes/users/micha/.Sent/dovecot.index
 /var/mail/mailboxes/users/micha/.Sent/dovecot.index.log
 
 I tried to renaming:
 /var/mail/mailboxes/users/micha/.Sent/cur/1240390202.Rfd9752b8Q14135.mail-alt,S=280003,W=284007:2,S
  to
 /var/mail/mailboxes/users/micha/.Sent/cur/1240390202.Rfd9752b8Q14135.mail-alt,S=280003:2,S
  

How large is that file really? I think you need to both rename it and delete 
dovecot.index.cache file. If the S=value is also wrong you'd need to delete it 
too.

Re: [Dovecot] fts solr : out of memory

2012-07-31 Thread to...@starbridge.org
Le 31/07/2012 21:04, Timo Sirainen a écrit :
 On 29.7.2012, at 2.20, to...@starbridge.org wrote:

 it seems ok now with memory problem, but an other error shows up:

 doveadm -v index -u c...@spamguard.fr INBOX
 doveadm(c...@spamguard.fr): Info: INBOX: Caching mails seq=1..59363
 21500/59363doveadm(c...@spamguard.fr): Error: fts_solr: Invalid XML
 input at line 1: mismatched tag
 No idea. You can reproduce this? What does it log with this patch? 
 http://hg.dovecot.org/dovecot-2.1/rev/817b69b2b21f

It happens every time on the same mailboxes (very few) around the same
uid number (I think I can find the exact uid with strace and send the
email message to you if it helps)

catalina.out show this at this time:

INFO: {} 0 1
31 juil. 2012 21:19:56 org.apache.solr.common.SolrException log
GRAVE: org.apache.solr.common.SolrException: Illegal character
((CTRL-CHAR, code 4))
 at [row,col {unknown-source}]: [533,500]
at org.apache.solr.handler.XMLLoader.load(XMLLoader.java:81)
at
org.apache.solr.handler.ContentStreamHandlerBase.handleRequestBody(ContentStreamHandlerBase.java:58)
at
org.apache.solr.handler.RequestHandlerBase.handleRequest(RequestHandlerBase.java:129)
at org.apache.solr.core.SolrCore.execute(SolrCore.java:1376)
at
org.apache.solr.servlet.SolrDispatchFilter.execute(SolrDispatchFilter.java:365)
at
org.apache.solr.servlet.SolrDispatchFilter.doFilter(SolrDispatchFilter.java:260)
at
org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(ApplicationFilterChain.java:235)
at
org.apache.catalina.core.ApplicationFilterChain.doFilter(ApplicationFilterChain.java:206)
at
org.apache.catalina.core.StandardWrapperValve.invoke(StandardWrapperValve.java:233)
at
org.apache.catalina.core.StandardContextValve.invoke(StandardContextValve.java:191)
at
org.apache.catalina.core.StandardHostValve.invoke(StandardHostValve.java:127)
at
org.apache.catalina.valves.ErrorReportValve.invoke(ErrorReportValve.java:102)
at
org.apache.catalina.core.StandardEngineValve.invoke(StandardEngineValve.java:109)
at
org.apache.catalina.connector.CoyoteAdapter.service(CoyoteAdapter.java:293)
at
org.apache.coyote.http11.Http11Processor.process(Http11Processor.java:859)
at
org.apache.coyote.http11.Http11Protocol$Http11ConnectionHandler.process(Http11Protocol.java:602)
at
org.apache.tomcat.util.net.JIoEndpoint$Worker.run(JIoEndpoint.java:489)
at java.lang.Thread.run(Unknown Source)
Caused by: com.ctc.wstx.exc.WstxUnexpectedCharException: Illegal
character ((CTRL-CHAR, code 4))
 at [row,col {unknown-source}]: [533,500]
at
com.ctc.wstx.sr.StreamScanner.throwInvalidSpace(StreamScanner.java:675)
at
com.ctc.wstx.sr.BasicStreamReader.readTextSecondary(BasicStreamReader.java:4668)
at
com.ctc.wstx.sr.BasicStreamReader.readCoalescedText(BasicStreamReader.java:4126)
at
com.ctc.wstx.sr.BasicStreamReader.finishToken(BasicStreamReader.java:3701)
at
com.ctc.wstx.sr.BasicStreamReader.safeFinishToken(BasicStreamReader.java:3649)
at
com.ctc.wstx.sr.BasicStreamReader.getText(BasicStreamReader.java:809)
at org.apache.solr.handler.XMLLoader.readDoc(XMLLoader.java:315)
at
org.apache.solr.handler.XMLLoader.processUpdate(XMLLoader.java:156)
at org.apache.solr.handler.XMLLoader.load(XMLLoader.java:79)
... 17 more

31 juil. 2012 21:19:56 org.apache.solr.core.SolrCore execute
INFO: [] webapp=/solr path=/update params={} status=400 QTime=1

After a quick google search , it seems related to invalid Control
Character sent to SOLR.



I've applied your last patch and the message is now:
Error: fts_solr: Invalid XML input at 4:113: mismatched tag (near:
htmlheadtitleApache Tomcat/6.0.35 - Rapport
d'erreur/titlestyle!--H1
{font-family:Tahoma,Arial,sans-serif;color:white)
doveadm(xx@xx): Panic: file solr-connection.c: line 547
(solr_connection_post_more): assertion failed: (maxfd = 0)
doveadm(xx@xx): Error: Raw backtrace:
/usr/lib/dovecot/libdovecot.so.0(+0x43d9a) [0x7fadd7e1dd9a] -
/usr/lib/dovecot/libdovecot.so.0(default_fatal_handler+0x2a)
[0x7fadd7e1de5a] - /usr/lib/dovecot/libdovecot.so.0(i_fatal+0)
[0x7fadd7df562a] -
/usr/lib/dovecot/modules/lib21_fts_solr_plugin.so(solr_connection_post_more+0x249)
[0x7fadd5f84929] -
/usr/lib/dovecot/modules/lib21_fts_solr_plugin.so(+0x4af7)
[0x7fadd5f81af7] -
/usr/lib/dovecot/modules/lib20_fts_plugin.so(fts_build_mail+0x3bd)
[0x7fadd65a6e5d] -
/usr/lib/dovecot/modules/lib20_fts_plugin.so(+0xc570) [0x7fadd65ab570]
- doveadm() [0x412ce1] - doveadm() [0x40fbd2] - doveadm() [0x4105db]
- doveadm(doveadm_mail_try_run+0x161) [0x410a41] - doveadm(main+0x375)
[0x40f7e5] - /lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xfd)
[0x7fadd7a71ead] - doveadm() [0x40f9d1]
Abandon


and catalina.out:

GRAVE: org.apache.solr.common.SolrException: Illegal character
((CTRL-CHAR, code 4))
 at [row,col {unknown-source}]: [533,500]
 

Re: [Dovecot] dovecot-lda setting quota rules

2012-07-31 Thread Timo Sirainen
On 18.7.2012, at 4.33, Lazy wrote:

 I wan't it to stay that way. The problem is that it dosn't update the
 quota rules in Maildir when quota is changed in userdb.
 I have the rules avaiable in qmail-lda's environment, but I need an
 idea on how to feed the quota rule string to quota plugin without
 doing userdb lookup.
 
 What whould be the best way to do this, maybe in
 plugins/quota/quota.c:quota_root_add_rules() ?

dovecot-lda -o plugin/quota_rule=*:storage=1G



Re: [Dovecot] doveadm director move behaviour

2012-07-31 Thread Timo Sirainen
On 12.7.2012, at 14.14, Angel L. Mateo wrote:

   So I have used doveadm director move user backend ip to move them 
 to its hashed server. Although a doveadm director status user now shows 
 me that this user is directed to the new server, doveadm who in the temporal 
 server shows me that the user is actually connected to it, although messages 
 are delivered to it through is correct server. So I have had to manually kick 
 that users in the temporal servers.
 
   Is this the expected behaviour of the move command for director?
 
   Is there any way to also force the kick of the imap/pop user's 
 connection in the temporal server?

Do you see any errors in logs? Especially:

director: Error: connect(ipc) failed: Permission denied



[Dovecot] zlib plugin maildir broken ?/ Cached message size smaller than expected

2012-07-31 Thread Robert Schetterer
Hi Timo,
by having errors like

Cached message size smaller than expected
and
Maildir filename has wrong S value
and
Corrupted index cache file

in dove 2.1.8

i can reproduce the problem with tb 14
with 2 nat machines connected to the same server

the problem relates with create new folder then copy
some uncompressed and compressed mail to it, the i.e
open it by other client

sometime this happens at once , sometime after a few copy actions

however thunderbird ,then errors above apear, sometimes no mail is
shown, sometimes only uncompressed or only compressed

so funktion read both formats gets lost somehow
any idea, you need a better debug ?
-- 
Best Regards
MfG Robert Schetterer


Re: [Dovecot] zlib plugin maildir broken ?/ Cached message size smaller than expected

2012-07-31 Thread Timo Sirainen
On 1.8.2012, at 0.29, Robert Schetterer wrote:

 Cached message size smaller than expected
 and
 Maildir filename has wrong S value
 and
 Corrupted index cache file
 
 in dove 2.1.8
 
 i can reproduce the problem with tb 14
 with 2 nat machines connected to the same server
 
 the problem relates with create new folder then copy
 some uncompressed and compressed mail to it, the i.e
 open it by other client
 
 sometime this happens at once , sometime after a few copy actions

I can't reproduce this. Are you sure the problem isn't that the old mails 
already have a wrong S value and the errors happen when Dovecot tries to read 
them? It would help to know how to reproduce this on an entirely empty mail 
account.



Re: [Dovecot] BUG: LDAP extension failed to authenticate if 'base' DN configuration is empty ''

2012-07-31 Thread Hendy Irawan

Dear Timo,

Thank you very much for your kind work.

I apologize, my previous statement were wrong. At least on ApacheDS 1.5.7,
it doesn't search entries using the '' base DN (it doesn't give error
though, just cannot return results, ApacheDS requires the search base to be
in a partition).

However, your patch would be great for servers who support the '' search
base DN.

If I may ask, would you improve it further to accept multiple bases? I'm not
sure what the proper character separator would be, but I think something
like this would work :

base = dc=prd,dc=berbatik,dc=com | dc=stg,dc=berbatik,dc=com

The proposed configuration above asks to try the two search bases in
succession, whichever first succeeds get logged in, else fails like usual.

I hope it's okay with you. I'll definitely help with the testing.

Hendy



Timo Sirainen wrote:
 
 On 29.7.2012, at 15.00, Hendy Irawan wrote:
 
 We have several context entries and want to authenticate a user against
 the
 whole LDAP server.
 
 However when setting base =  Dovecot LDAP fails with following message:
 
 Jul 29 10:50:59 nitik1 dovecot: auth: Error:
 ldap(he...@staging.member.berbatik.com,127.0.0.1):
 ldap_search((virtualMail=he...@staging.member.berbatik.com)) failed:
 Other
 (e.g., implementation specific) error
 
 Does the attached patch help? If yes, I'll add it to next version.
 
 
  
 


-
http://www.Soluvas.com/ Soluvas - Making eCommerce Work for You 
-- 
View this message in context: 
http://old.nabble.com/BUG%3A-LDAP-extension-failed-to-authenticate-if-%27base%27-DN-configuration-is-empty-%27%27-tp34226738p34238207.html
Sent from the Dovecot mailing list archive at Nabble.com.



Re: [Dovecot] BUG: LDAP extension failed to authenticate if 'base' DN configuration is empty ''

2012-07-31 Thread Timo Sirainen
On 1.8.2012, at 1.12, Hendy Irawan wrote:

 I apologize, my previous statement were wrong. At least on ApacheDS 1.5.7,
 it doesn't search entries using the '' base DN (it doesn't give error
 though, just cannot return results, ApacheDS requires the search base to be
 in a partition).
 
 However, your patch would be great for servers who support the '' search
 base DN.

But does my patch actually make it work in some such servers? I did it only as 
a guess. If it doesn't fix anything I'd rather just drop the patch.

 If I may ask, would you improve it further to accept multiple bases? I'm not
 sure what the proper character separator would be, but I think something
 like this would work :
 
 base = dc=prd,dc=berbatik,dc=com | dc=stg,dc=berbatik,dc=com
 
 The proposed configuration above asks to try the two search bases in
 succession, whichever first succeeds get logged in, else fails like usual.

You can already do this by creating multiple passdb/userdb ldap sections with 
different config files. Searching multiple bases requires doing multiple LDAP 
lookups anyway.

Re: [Dovecot] BUG: LDAP extension failed to authenticate if 'base' DN configuration is empty ''

2012-07-31 Thread Hendy Irawan

Dear Timo,

Thank you for your information about multiple sections! I didn't realize
that was possible.

Yes, your patch will work on at least ApacheDS 2.0.0-M7 (see
https://issues.apache.org/jira/browse/DIRSERVER-1742 ) and I would
appreciate it very much !

Hendy


Timo Sirainen wrote:
 
 On 1.8.2012, at 1.12, Hendy Irawan wrote:
 
 I apologize, my previous statement were wrong. At least on ApacheDS
 1.5.7,
 it doesn't search entries using the '' base DN (it doesn't give error
 though, just cannot return results, ApacheDS requires the search base to
 be
 in a partition).
 
 However, your patch would be great for servers who support the '' search
 base DN.
 
 But does my patch actually make it work in some such servers? I did it
 only as a guess. If it doesn't fix anything I'd rather just drop the
 patch.
 
 If I may ask, would you improve it further to accept multiple bases? I'm
 not
 sure what the proper character separator would be, but I think something
 like this would work :
 
 base = dc=prd,dc=berbatik,dc=com | dc=stg,dc=berbatik,dc=com
 
 The proposed configuration above asks to try the two search bases in
 succession, whichever first succeeds get logged in, else fails like
 usual.
 
 You can already do this by creating multiple passdb/userdb ldap sections
 with different config files. Searching multiple bases requires doing
 multiple LDAP lookups anyway.
 


-
http://www.Soluvas.com/ Soluvas - Making eCommerce Work for You 
-- 
View this message in context: 
http://old.nabble.com/BUG%3A-LDAP-extension-failed-to-authenticate-if-%27base%27-DN-configuration-is-empty-%27%27-tp34226738p34238917.html
Sent from the Dovecot mailing list archive at Nabble.com.



Re: [Dovecot] doveadm director move behaviour

2012-07-31 Thread Daniel Parthey


Timo Sirainen t...@iki.fi schrieb:

On 12.7.2012, at 14.14, Angel L. Mateo wrote:

  So I have used doveadm director move user backend ip to move
them to its hashed server. Although a doveadm director status user
now shows me that this user is directed to the new server, doveadm who
in the temporal server shows me that the user is actually connected to
it, although messages are delivered to it through is correct server. So
I have had to manually kick that users in the temporal servers.

  Is this the expected behaviour of the move command for director?

  Is there any way to also force the kick of the imap/pop user's
connection in the temporal server?

Do you see any errors in logs? Especially:

director: Error: connect(ipc) failed: Permission denied

Looks very familiar to me.

I'm having the same problem as Angel with director move that I have to kick the 
user logged into the proxy. Plus I'm also getting the ipc permission denied 
errors you mentioned.

Regards
Daniel



Re: [Dovecot] zlib plugin maildir broken ?/ Cached message size smaller than expected

2012-07-31 Thread Robert Schetterer
Am 31.07.2012 23:44, schrieb Timo Sirainen:
 On 1.8.2012, at 0.29, Robert Schetterer wrote:
 
 Cached message size smaller than expected
 and
 Maildir filename has wrong S value
 and
 Corrupted index cache file

 in dove 2.1.8

 i can reproduce the problem with tb 14
 with 2 nat machines connected to the same server

 the problem relates with create new folder then copy
 some uncompressed and compressed mail to it, the i.e
 open it by other client

 sometime this happens at once , sometime after a few copy actions
 
 I can't reproduce this. Are you sure the problem isn't that the old mails 
 already have a wrong S value and the errors happen when Dovecot tries to read 
 them? 

at my recent onwledge
mixing  compressed new mails from my sent folder, and older uncompressed
ones ( which are shown nice both in tb ), with some copy action in a new
folder from another client leads to wrong S
and tb does not display them any longer sometime, perhaps a double zip
happens?

It would help to know how to reproduce this on an entirely empty mail
account.
 

i will try to test, but first i have to deal with now broken maildirs
i try disable zlip plugin now ,as workaround hopefully this does not
lead to more problems on old maildirs , and fix the problem with new mail

-- 
Best Regards
MfG Robert Schetterer