Re: [Dovecot] Problem with ldap error logging

2009-08-26 Thread Bernie Innocenti
El Tue, 25-08-2009 a las 09:25 -0400, Timo Sirainen escribió:
 That's what it does..:
 
 % ./deliver 
 % echo $?
 75
 % tail -1 /var/log/dovecot.log
 Aug 25 09:24:01 deliver(tss): Fatal: Plugin asdf not found from directory 
 /usr/local/lib/dovecot/lda
 % grep define.*EX_TEMPFAIL /usr/include/sysexits.h 
 #define EX_TEMPFAIL 75  /* temp failure; user is invited to retry */


I checked, qmail-lspawn handles 75 as a temporary failure.

Then I dunno what happened in my case, but some mail definitely went
into a black hole.

-- 
   // Bernie Innocenti - http://codewiz.org/
 \X/  Sugar Labs   - http://sugarlabs.org/




[Dovecot] Disconnected in APPEND sometimes when with attachment

2009-08-26 Thread Igor Bogomazov
Hello!

When I send a message with an attachment bigger than some undefined
size, the folllowing error appears in logs: Disconnected in APPEND

How to increase the allowed size or anything?

# dovecot -n
# 1.0.15: /etc/dovecot/dovecot.conf
log_timestamp: %Y-%m-%d %H:%M:%S
protocols: pop3 pop3s imap imaps
ssl_cert_file: /usr/local/etc/ssl/certs/highlink.ru.pem
ssl_key_file: /usr/local/etc/ssl/private/highlink.ru.key
disable_plaintext_auth: no
login_dir: /var/run/dovecot/login
login_executable(default): /usr/lib/dovecot/imap-login
login_executable(imap): /usr/lib/dovecot/imap-login
login_executable(pop3): /usr/lib/dovecot/pop3-login
mail_privileged_group: mail
mail_location: maildir:~/.maildir
mail_executable(default): /usr/lib/dovecot/imap
mail_executable(imap): /usr/lib/dovecot/imap
mail_executable(pop3): /usr/lib/dovecot/pop3
mail_plugin_dir(default): /usr/lib/dovecot/modules/imap
mail_plugin_dir(imap): /usr/lib/dovecot/modules/imap
mail_plugin_dir(pop3): /usr/lib/dovecot/modules/pop3
pop3_uidl_format(default):
pop3_uidl_format(imap):
pop3_uidl_format(pop3): %08Xu%08Xv
auth default:
  mechanisms: plain login
  passdb:
driver: passwd-file
args: /etc/dovecot/dovecot.passwd
  userdb:
driver: passwd-file
args: /etc/dovecot/dovecot.passwd
  socket:
type: listen
master:
  path: /var/run/dovecot/auth-master
  mode: 384
  user: dovecot
  group: dovecot


signature.asc
Description: PGP signature


Re: [Dovecot] fchown() failed

2009-08-26 Thread Arthur Dent
On Mon, 2009-08-24 at 15:05 -0400, Timo Sirainen wrote:
 On Wed, 2009-08-19 at 13:18 +0100, Arthur Dent wrote:
  dovecot: Dovecot v1.2.0 starting up (core dumps disabled): 1 Time(s)
  dovecot: IMAP(mark): fchown() failed with file 
  /home/mark/Mail/.imap/INBOX/dovecot.index.log.newlock: Operation not 
  permitted: 1 Time(s)
 
 v1.2.3+ would have given a bit better error message here.

Ironically, since I posted my original message last week Fedora issued
an updated Dovecot package, so here is yesterday's message in syslog:

 **Unmatched Entries**
dovecot: Dovecot v1.2.3 starting up (core dumps disabled): 1 Time(s)
dovecot: IMAP(helena): 
fchown(/home/wife/mail/.imap/INBOX/dovecot.index.cache.lock, -1, 12(mail)) 
failed: Operation not permitted (egid=501(helena), group based on 
/var/mail/helena): 1 Time(s)
dovecot: IMAP(mark): chown(/home/mark/Mail/.imap/INBOX, -1, 12(mail)) 
failed: Operation not permitted (egid=500(mark), group based on 
/var/mail/mark): 1 Time(s)
dovecot: IMAP(mark): 
fchown(/home/mark/Mail/.imap/INBOX/dovecot.index.cache.lock, -1, 12(mail)) 
failed: Operation not permitted (egid=500(mark), group based on 
/var/mail/mark): 1 Time(s)
dovecot: IMAP(mark): 
fchown(/home/mark/Mail/.imap/INBOX/dovecot.index.log.newlock, -1, 12(mail)) 
failed: Operation not permitted (egid=500(mark), group based on 
/var/mail/mark): 1 Time(s)
dovecot: IMAP(mark): fchown(/home/mark/Mail/.imap/INBOX/dovecot.index.tmp, 
-1, 12(mail)) failed: Operation not permitted (egid=500(mark), group based on 
/var/mail/mark): 1 Time(s)
dovecot: IMAP(mark): mkdir(/home/mark/Mail/.imap/INBOX) failed: Operation 
not permitted: 1 Time(s)

 Anyway the issue is that nowadays Dovecot tries to preserve mailbox's 
 permissions
 so that shared mailboxes work properly. What permissions do you have in
 the actual INBOX file?

ls -la /var/mail/
total 202024
drwxrwxr-x.  2 root   mail  4096 2009-08-26 10:30 .
drwxr-xr-x. 15 root   root  4096 2009-08-13 12:35 ..
-rw-rw.  1 clamav mail 0 2009-08-12 12:40 clamav
-rw-rw.  1 wife   mail 190669010 2009-08-26 10:00 wife
-rw-rw.  1 mark   mail   8421931 2009-08-26 09:46 mark
-rw---.  1 root   root787636 2009-08-26 10:30 root
-rw-rw.  1 rpcmail 0 2009-08-11 16:26 rpc
-rw-rw.  1 sonmail   6750285 2009-08-26 00:39 son

 
 My guess is that you have 0660 permissions of /var/mail/$USER so Dovecot
 tries to preserve the group. Easiest fix that also makes your system
 more secure is to chmod 0600 it.

Hmmm - OK, seems you're right about that. Will changing it break
anything else? Why does group mail exist?

I will try chmod 0600 and see what happens. 

 
  As I access my mailbox several times a day (and certainly did yesterday)
  from several different mail clients I am surprised that the error count
  is only 1 for most the above entries. Is this being caused by some sort
  of cron job (logrotate perhaps)?
 
 It happens only when index files are being created/rotated, which
 happens automatically every once in a while.

Thanks very much. Your help is greatly appreciated...

Mark



signature.asc
Description: This is a digitally signed message part


Re: [Dovecot] Disconnected in APPEND sometimes when with attachment

2009-08-26 Thread Charles Marcus
On 8/26/2009, Igor Bogomazov (b...@hl.ru) wrote:
 When I send a message with an attachment bigger than some undefined
 size, the folllowing error appears in logs: Disconnected in APPEND

1. Dovecot is not an smtp server - it does not 'send' messages.

2. While 1.0.15 is old and you should upgrade (1.2.4 is current stable),
upgrading won't fix your problem.

3. You need to fix this with your smtp server config.

Also, please don't paraphrase logs... post the entire set of relevant
log lines - that would have probably allowed us to see what smtp server
you were using and give more specific recommendation on next steps..

-- 

Best regards,

Charles


Re: [Dovecot] Two server certificates for two common names

2009-08-26 Thread Δημήτριος Καραπιπέρης

So ,
on one dovecot instance, it is impossible to have two ssl certificates 
for two distinct common names.

right?

thanks
Dimitrios





O/H Ed W έγραψε:

Patrick Domack wrote:
That is an ssl imposibility, and the current tls clients can't really 
do that either. The best way to do it is to use seperate ip's for 
mail1 and mail2.


The only other option is to use a new certificate with 
subject-alt-names, but lots of email clients don't support that. 
Webbrowsers have for a long time now, but email is completely different.


I would be interested to hear which mail clients don't support this?

My experience is that the main culprits are ok (including apple and 
many handhelds).


I use a cheapo S-A-N from GoDaddy and it seems to work ok (but I 
hardly have a wide range of clients using it)


Hope this helps?

Ed W



--
ΔΗΜΗΤΡΙΟΣ ΚΑΡΑΠΙΠΕΡΗΣ 
ΤΕΧΝ. ΥΠ. ΣΥΖΕΥΞΙΣ


ΕΛΛΗΝΙΚΗ ΔΗΜΟΚΡΑΤΙΑ - Ν. ΘΕΣΣΑΛΟΝΙΚΗΣ
ΔΗΜΟΣ ΘΕΣΣΑΛΟΝΙΚΗΣ -  Δ/ΝΣΗ ΟΡΓΑΝΩΣΕΩΣ  ΜΕΘΟΔΩΝ
2310 - 257844 fax 2310 - 244965



Re: [Dovecot] Quotas ignored on INBOX only

2009-08-26 Thread Random Monkey
Thanks

The way I am testing is as follows.

Setup two users on the system. Send email from user A to user B.

Quota counts increase for user A as the email is copied to the Sent folder. 
User B quota count does not increase (size or messages)

Reply from user B to user A to confirm problem effects both accounts.

Remove email from sent items on one account and expunge and note quota record 
decreases.

Remove email from INBOX and expunge, note quota record does not change.

Repeat sending emails from B to A and notice quota limit is reached on sent 
items - error is received copying email to sent items. With quota exceeded 
switch to user account A and continue to send emails to B, note that they 
appear in the INBOX of the user despite the full quota with no bounce back.

This behaviour is identical regardless of quota backend used (have tried 
maildir++ and dict so far).

This is using the default aptitude package of dovecot-postifx for ubuntu 9.0.4 
- Is this a bug?



  

Re: [Dovecot] Two server certificates for two common names

2009-08-26 Thread Eduardo M KALINOWSKI

On Qua, 26 Ago 2009, Δημήτριος Καραπιπέρης wrote:

So ,
on one dovecot instance, it is impossible to have two ssl  
certificates for two distinct common names.

right?


At the moment, yes. In a future version this will be possible, but I  
suppose you will still need two IPs.



--
Eduardo M KALINOWSKI
edua...@kalinowski.com.br


Re: [Dovecot] Dovecot and LDAP-Quotas

2009-08-26 Thread south africa
 It says nothing about LDAP here, which means that unless you filtered
 out some lines it's not using LDAP at all for anything. What does your
 dovecot -n output show now?


dovecot -n
# 1.2.3: /etc/dovecot.conf
# OS: Linux 2.6.18-128.4.1.el5 x86_64 CentOS release 5.3 (Final) ext3
log_path: /var/log/dovecot.log
info_log_path: /var/log/dovecot.log
protocols: pop3 pop3s imap imaps
listen(default): 123.456.789.71
listen(imap): 123.456.789.71
listen(pop3): 123.456.789.72
ssl_listen(default): 
ssl_listen(imap): 
ssl_listen(pop3): 123.456.789.72:995
ssl_ca_file: /etc/pki/dovecot/certs/pop_core_uk/trustcenter_intermediate.crt
ssl_cert_file: /etc/pki/dovecot/certs/pop_core_uk/server.crt
ssl_key_file: /etc/pki/dovecot/private/pop_core_uk/server.key
login_dir: /var/run/dovecot/login
login_executable(default): /usr/libexec/dovecot/imap-login
login_executable(imap): /usr/libexec/dovecot/imap-login
login_executable(pop3): /usr/libexec/dovecot/pop3-login
mail_location: maildir:/home/vmail/%Lu/Maildir/
mail_debug: yes
mail_executable(default): /usr/libexec/dovecot/imap
mail_executable(imap): /usr/libexec/dovecot/imap
mail_executable(pop3): /usr/libexec/dovecot/pop3
mail_plugins(default): quota imap_quota
mail_plugins(imap): quota imap_quota
mail_plugins(pop3): quota
mail_plugin_dir(default): /usr/lib64/dovecot/imap
mail_plugin_dir(imap): /usr/lib64/dovecot/imap
mail_plugin_dir(pop3): /usr/lib64/dovecot/pop3
imap_client_workarounds(default): delay-newmail outlook-idle netscape-eoh
imap_client_workarounds(imap): delay-newmail outlook-idle netscape-eoh
imap_client_workarounds(pop3): 
pop3_client_workarounds(default): 
pop3_client_workarounds(imap): 
pop3_client_workarounds(pop3): outlook-no-nuls oe-ns-eoh
lda:
  debug: yes
  postmaster_address: postms...@core.uk
  hostname: smtp.core.uk
  mail_plugins: quota
  quota_full_tempfail: yes
  log_path: /var/log/dovecot-deliver.log
  log_timestamp: %b %d %H:%M:%S
  rejection_reason: Ihre Nachricht an %t wurde automatisiert abgewiesen:%n%r
auth default:
  mechanisms: plain login
  username_translation: @_._
  verbose: yes
  debug: yes
  debug_passwords: yes
  passdb:
driver: ldap
args: /etc/dovecot-ldap.conf
  userdb:
driver: static
args: uid=500 gid=500 home=/home/vmail/%Lu allow_all_users=yes
  userdb:
driver: ldap
args: /etc/dovecot-ldap-userdb.conf
  socket:
type: listen
client:
  path: /var/spool/postfix/private/auth
  mode: 432
  user: postfix
  group: postfix
master:
  path: /var/run/dovecot/auth-master
  mode: 432
  user: vmail
  group: postfix
plugin:
  quota: maildir:User quota
  quota_rule: *:storage=8M
  quota_warning: storage=90%% /usr/local/bin/quota-warning.sh 90
  quota_warning2: storage=75%% /usr/local/bin/quota-warning.sh 75

--
dovecot-ldap.conf

hosts = 123.456.789.123 123.456.789.124
base = OU=one,OU=two,OU=London,OU=GB,OU=User 
Accounts,DC=three,DC=core,DC=uk
ldap_version = 3
auth_bind = yes
auth_bind_userdn = %...@three.core.uk

user_filter = ((objectClass=person)(cn=%u))
user_attrs = description=quota_rule=*:storage=%$M

--
dovecot-ldap-userdb.conf is a symbolic link on dovecot-ldap.conf






_
Get your FREE TheDoghouseMail email address at http://www.thedoghousemail.com


[Dovecot] mbox to maildir migration question and problems, dovecot-1.0.7-7.el5

2009-08-26 Thread Eric Sammons
Hi,

I am trying to assist a customer with a mbox to maildir migration.  I have
them using the latest and greatest mb2md.pl; the one that support -K,
-U|-u.  They have seen overwhelming success with this; however, a recent
mbox introduced a new scenario and the result was that all messages were
replayed in the client UI.

The mbox in question had _no_ X-UID tags.  To work-a-round the issue the
customer performed the following steps.

[snip]
We found that the fix was to clear /var/indexes/$USER for the account and
log in and log back out.  After doing this X-UID's were populated for each
respective message.
[/snip]

The customer also noted that after the migration all messages went to
~/MailDir/new.

I would appreciate any thoughts on why this scenario played out the way it
did?  Is it expected that these messages should reside in ~/MailDir/new?

[config]
protocol pop3 {
  pop3_uidl_format =  %08Xv%08Xu
  pop3_reuse_xuidl = yes
}
[/config]

Thank you!

-- 
Eric L. Sammons, RHCE


[Dovecot] Weird mirgration problem

2009-08-26 Thread Stewart Dean
Most likely non-Dovecot, but I'd appreciate any comments on WTF might be 
going on here.
Preparatory to migration from mbox to maildir format on our AIX mail 
server, I am migrating from the JFS filesystem (being deprecated by IBM) 
to JFS2 to better handling the jump in the number of files.
Two nights ago, I took the first step and migrated the IT homedir 
filesystem: copied it to a scratch disk, melted down the FS and 
recreated it as a JFS2 FS.  In order to keep changes from happening to 
it during this process, I killed off DC (dovecot -n ouput is appended to 
bottom of this note) and dropped the export of the FS.  The bringup was 
a little ragged due to time constraints and the fact that I had trouble 
re-establishing the export.  FWIW, the homedirs and Inobx FSes are 
mounted by a majordomo mailing list server.
Everything went fine except for an anomaly with old pre-existing mailing 
list emails in my INBOX and that of a few others.  Note that the inbox 
FS wasn't touched by this process, just the homedir FS for the IT 
department.  What happened there was that the mbox headers went from 
looking like this:


 From owner-hcrcstaff  Mon Aug 24 14:53:29 2009

From majordom  Mon Aug 24 14:53:29 2009
 the rest of the normal header here

To looking like this:

 From owner-hcrcstaff  Mon Aug 24 14:53:29 2009
 Status: RO
 X-UID: 871476 
 Content-Length: 1915


 From majordom  Mon Aug 24 14:53:29 2009
 normal header lines here

It appears that the Status, X-UID and Content-Length lines were inserted 
just under the top line of the header in every old mailing list mailing 
email header...and not in any other mailing

As a result of this,
= these emails appear in the email client's TOC listing with blank 
Sender and Subject fields, even though the data for that is properly in 
the header further down

= This happens both in TBird and Horde webmail.
= All these emails (some 300, going back to 2008) are listed in one 
block, in proper order for their actual date, as having the datestamp of 
5AM, when I was bringing everything back up.  Unfortunately, because of 
time pressures, I can say exactly what I was doing then.
= I never took down my TBird session while doing this process...and I 
imagine that others, who haven't said anything and presumably didn't 
have this problem, left their TBird or IMAP email client up...


Does this anomaly bring to mind anything to you?

FWIW, I plan on doing the students' homedir FS early tomorrow morning.  
This time I'll be more scrupulous about tracking things AND I will take 
down the mailing list and login servers to remove the possibility that 
they had something to do with this...


dovecot -n output

# 1.1.15: /usr/local/etc/dovecot.conf
# OS: AIX 3 0001378F4C00  
listen: *:143

ssl_listen: *:993
disable_plaintext_auth: no
verbose_ssl: yes
login_dir: /var/run/dovecot/login
login_executable: /usr/local/libexec/dovecot/imap-login
login_processes_count: 12
login_max_processes_count: 774
max_mail_processes: 1024
verbose_proctitle: yes
first_valid_uid: 200
mail_location: mbox:~/mail:INBOX=/var/spool/mail/%u:INDEX=/var/dcindx/%u
mbox_write_locks: fcntl
mbox_dirty_syncs: no
auth default:
 passdb:
   driver: pam
 userdb:
   driver: passwd



--
 Once upon a time, the Internet was a friendly, 
neighbors-helping-neighbors small town, and no one locked their doors. 
Now it's like an apartment in Bed-Stuy: you need three heavy duty 
pick-proof locks, one of those braces that goes from the lock to the 
floor, and bars on the windows  Stewart Dean, Unix System Admin, 
Bard College, New York 12504 sd...@bard.edu voice: 845-758-7475, fax: 
845-758-7035




Re: [Dovecot] Two server certificates for two common names

2009-08-26 Thread Ed W

Δημήτριος Καραπιπέρης wrote:

So ,
on one dovecot instance, it is impossible to have two ssl certificates 
for two distinct common names.

right?



You are kind of asking two questions here:

1) SSL as it stands maps one IP address to one certificate.  The basic 
issue is that, bar a few exceptions, there is no clear way to connect to 
an IP address and say what domain you are expecting to see on the 
other end, hence allowing the other end to present the domain specific 
cert.  This is currently not fixable, but you can work around it by 
getting one cert with all your CNs on it (see Subject Alt Name)


2) Does Dovecot support running on 2 ips with different certs on each 
IP?  I think the answer is currently no?  You could run two dovecot 
instances though...  I believe this is on the todo list for a later 
version, but as yet not that high up the priority list? (Timo?)  So this 
bit is fixable in various ways


Does that help?

Ed W



Re: [Dovecot] Two server certificates for two common names

2009-08-26 Thread Δημήτριος Καραπιπέρης
Basically, server is not expecting any kind of domain on ssl handshake, 
but what if the server can serve more than one cert, so that clients
using  mail1.dom.gr and mail2.dom.gr , which resolve to the same dovecot 
instance but from different network segments

could be certified.

mail1.dom.gr - 10.65.0.45  (private one)
mail2.dom.gr - 84.205.252.78
(random numbers)

In essence, it is the same dovecot instance.

Dimitrios

O/H Ed W έγραψε:

Δημήτριος Καραπιπέρης wrote:

So ,
on one dovecot instance, it is impossible to have two ssl 
certificates for two distinct common names.

right?



You are kind of asking two questions here:

1) SSL as it stands maps one IP address to one certificate.  The basic 
issue is that, bar a few exceptions, there is no clear way to connect 
to an IP address and say what domain you are expecting to see on the 
other end, hence allowing the other end to present the domain specific 
cert.  This is currently not fixable, but you can work around it by 
getting one cert with all your CNs on it (see Subject Alt Name)


2) Does Dovecot support running on 2 ips with different certs on each 
IP?  I think the answer is currently no?  You could run two dovecot 
instances though...  I believe this is on the todo list for a later 
version, but as yet not that high up the priority list? (Timo?)  So 
this bit is fixable in various ways


Does that help?

Ed W





Re: [Dovecot] Two server certificates for two common names

2009-08-26 Thread Ed W

Δημήτριος Καραπιπέρης wrote:
Basically, server is not expecting any kind of domain on ssl 
handshake, but what if the server can serve more than one cert, so 
that clients
using  mail1.dom.gr and mail2.dom.gr , which resolve to the same 
dovecot instance but from different network segments

could be certified.

mail1.dom.gr - 10.65.0.45  (private one)
mail2.dom.gr - 84.205.252.78
(random numbers)

In essence, it is the same dovecot instance.


I should imagine that you can achieve this using an external SSL wrapper 
such as stunnel?


OR

You could use firewall rules to redirect incoming connections to 
different local ports depending on where the connection originates.  
Then setup appropriate config on each port to serve a different cert


This setup does sound workable

Ed W



Re: [Dovecot] Two server certificates for two common names

2009-08-26 Thread Timo Sirainen

On Aug 26, 2009, at 2:17 PM, Ed W wrote:

2) Does Dovecot support running on 2 ips with different certs on  
each IP?  I think the answer is currently no?  You could run two  
dovecot instances though...  I believe this is on the todo list for  
a later version, but as yet not that high up the priority list?  
(Timo?)  So this bit is fixable in various ways


Dovecot v2.0 supports different certs for different IPs. Until then  
you'll need to run multiple Dovecot instances with different config  
files.




Re: [Dovecot] Weird mirgration problem

2009-08-26 Thread Timo Sirainen

On Aug 26, 2009, at 12:36 PM, Stewart Dean wrote:


From owner-hcrcstaff  Mon Aug 24 14:53:29 2009

From majordom  Mon Aug 24 14:53:29 2009
the rest of the normal header here


Does the above look correct? Did you originally also have that   
there? And that empty line? If so, I think those mails were written  
with a buggy Dovecot version and it's more of a miracle that you  
hadn't got this problem earlier. :) Probably because using  
mbox_very_dirty_syncs and the files' mtimes had never changed without  
there being some other changes in the mbox.


If you have a lot of such mails, I think it's a good idea to get rid  
of that extra new line (and maybe the extra From-line) and delete  
dovecot.index* files.




[Dovecot] TLS / SSL mixed w/ plaintext auth and virtual hosting

2009-08-26 Thread Nathan M
Traditionally this server has only accepted plaintext authentications;
however, we want to change that and enable TLS/SSL.  The challenge is
the server has hundreds of IP addresses it binds to to listen on ports
110/143.

Enabling TLS/SSL is not an option because as this is a virtual hosting
environment, if a connection comes in on any other hostname other than
the specific one tied to the crt all mail clients will throw a
mis-matched certificate error if TLS is initiated by the client, and a
surprisingly large number of customers have use TLS if available
selected in their clients.

According to most of the suggestions on the list, I've setup 2 dovecot
instances. The first listening on *:110 and *:143, and the second
listening on 10.0.0.2:993 and 10.0.0.2:995.  This works great for SSL
support; however, I would also like to offer TLS to connections coming
in on a single IP address.

Because the server has hundreds of IPs, with new IPs adding all the
time, seemingly the only way would be to configure every one of these
IPs (as they occur) into the primary dovecot.conf file, and then only
setup the single IP that's handling SSL/TLS in the dovecot-ssl.conf
(the conf file the SSL/TLS instance loads).  This can be time
consuming and has no way to automate.  It would be terrific if one of
the following exists, or potentially could exist:

1. Ideal scenario.  A config option which tells TLS to only respond on
certain IPs.  In our case if a connection attempts to initiate TLS on
any IP address except 10.0.0.2, it would respond with no TLS support.
This would be ideal as we could continue running just a single dovecot
instance.

2. Secondary scenario.  A way to exclude an IP from being bound to.
Something like the following to bind to all except 10.0.0.2
  listen = *:110, -10.0.0.2

As is, based on my understanding of the config neither of these are
options.  Any support for adding either of these options, or alternate
ideas anyone might have?

- N