- dovecot - dovecot.org

Trouble getting public folders working
by Bjarne Mark Kondrup 20 Aug '14

20 Aug '14

Hi i have trouble getting my public folders working properly. I have a few issues atm with them. First off i'm using version 2.0.9 as it came with CentOS and would greatly prefer to keep this version. Here is my dovecot.conf file # dovecot -n # 2.0.9: /etc/dovecot/dovecot.conf # OS: Linux 2.6.32-431.23.3.el6.centos.plus.x86_64 x86_64 CentOS release 6.5 (Final) ext4 auth_debug = yes auth_verbose = yes disable_plaintext_auth = no first_valid_uid = 101 last_valid_uid = 101 login_greeting = Dovecot ready. CEGO webmail server mail_debug = yes mail_gid = vmail mail_location = maildir:/home/vmail/domains/%d/%n mail_uid = vmail managesieve_notify_capability = mailto managesieve_sieve_capability = fileinto reject envelope encoded-character vacation subaddress comparator-i;ascii-numeric relational regex imap4flags copy include variables body enotify environment mailbox date mbox_write_locks = fcntl namespace { inbox = yes location = prefix = separator = / type = private } namespace { location = maildir:/home/vmail/shared:LAYOUT=fs:INDEX=/home/vmail/shared/ prefix = shared/ separator = / subscriptions = no type = public } passdb { args = /etc/dovecot/dovecot-ldap.conf driver = ldap } plugin { sieve = /home/vmail/domains/%d/%n/dovecot.sieve sieve_dir = /home/vmail/domains/%d/%n/sieve_scripts sieve_global_dir = /etc/dovecot/sieve/global/ } protocols = imap lmtp sieve service auth { unix_listener auth-client { group = postfix mode = 0660 user = postfix } unix_listener auth-master { group = vmail mode = 0660 user = vmail } user = root } service lmtp { unix_listener /var/spool/postfix/private/dovecot-lmtp { group = postfix mode = 0600 user = postfix } } service managesieve-login { inet_listener sieve { port = 4190 } } ssl = no ssl_cert = </etc/pki/dovecot/certs/dovecot.pem ssl_key = </etc/pki/dovecot/private/dovecot.pem userdb { driver = passwd } userdb { args = /etc/dovecot/dovecot-ldap.conf driver = ldap } valid_chroot_dirs = /home/vmail/domains verbose_ssl = yes protocol lmtp { log_path = /var/log/dovecot-deliver.log mail_plugins = quota sieve sieve acl postmaster_address = postmaster(a)cegotest.dk } protocol sieve { info_log_path = /var/log/dovecot-sieve.log log_path = /var/log/dovecot-sieve-errors.log managesieve_implementation_string = dovecot managesieve_max_line_length = 65536 } And my dovecot-ldap file hosts = localhost dn = cn=dovecot,dc=cegotest,dc=dk dnpass = password auth_bind = yes ldap_version = 3 base = o=hosting,dc=cegotest,dc=dk deref = never scope = subtree user_filter = (&(objectClass=JammMailAccount)(mail=%u)(accountActive=TRUE)(delete=FALSE)) pass_attrs = mail,userPassword pass_filter = (&(objectClass=JammMailAccount)(mail=%u)(accountActive=TRUE)(delete=FALSE)) default_pass_scheme = CRYPT Here is my current issues. 1. I cannot get shared folders to work with pr. user settings. What i want is a public folder that all my users can see but each user marks mail as read individually. Atm no matter how i set it up if user1 marks a mail as read, then user2 will also see that mail as read :( 2. ACL to control the permissions. Right now all users can move mail and remove mail from the shared folders. I want people (except a few specific users) to be read only on this public folders. 3. Need to know if i'm creating the folders correctly. Right now i'm under my shared folder directory creating shared_folder and 3 subfolders beneath it (cur, new and tmp). After this it shows up in eg. roundcube and i can subscribe to it. I have read all the documentation and tried all the solutions i can. Mostly moving the index and/or config files to seperate dirs and from my understanding that should give each user their own index file marking what mails have been read ect. But this doesn't happen. Hope you guys are able to help me out a little here. -Bjarne

2 1

Huge difference between the lucene index size created by v2.1 and v2.2
by Akash 20 Aug '14

20 Aug '14

Hi everyone, While examining dovecot versions v2.1 and v2.2 for their lucene search performances, I have noticed a huge difference in the index sizes created by them. Both versions were compiled on same system, against same libclucene, same configure options and were used with same dovecot.conf during run-time. I deleted the lucene-indexes folder and dovecot* files in the Maildir prior to indexing with both versions. The tests were performed on a untouched mail archive folder containing 300000 mails without any dovecot* files in it: root@server:/home/admin/mails/.Archive# ls -l total 78640 drwx------ 2 2500 2500 14098432 Aug 17 22:16 cur drwx------ 2 2500 2500 12435456 Jul 30 09:46 new drwx------ 2 2500 2500 4096 Aug 2 13:02 tmp The command used was: doveadm -v index -u admin Archive After search indexing on v2.1 resulted in: root@server:/home/admin/mails# ls -lh lucene-indexes total 390M -rw------- 1 2500 2500 390M Aug 18 07:03 _25.cfs -rw------- 1 2500 2500 20 Aug 18 07:03 segments.gen -rw------- 1 2500 2500 46 Aug 18 07:03 segments_4d Whereas dovecot v2.2 resulted in: root@server:/home/admin/mails# ls -lh lucene-indexes total 1.5G -rw------- 1 2500 2500 1.5G Aug 18 06:41 _5g.cfs -rw------- 1 2500 2500 20 Aug 18 06:41 segments.gen -rw------- 1 2500 2500 46 Aug 18 06:41 segments_az 390M vs 1.5G. That is a huge difference in size. Why is that? Thanks in advance. -Regards, Akash

2 5

Problems with dovecot 2.2.13 and monit
by Hanno Böck 20 Aug '14

20 Aug '14

Hello, When I upgraded my servers to dovecot 2.2.13 the monitoring tool monit started to send out warnings that it couldn't reach my imap/pop3 servers through ssl any more. The same problem didn't happen on non-ssl-connections. According to people on the monit list this is likely a dovecot issue: https://lists.gnu.org/archive/html/monit-general/2014-06/msg00031.html Let me quote: > the root cause of the error is, that dovecot 2.2.13 closes the > connection if SSL is used in response to LOGOUT command instead of > sending usual response. When no SSL is enabled, dovecot responses to > LOGOUT command normally. [...] > According to RFC 3501 (http://tools.ietf.org/html/rfc3501) LOGOUT is > any-state command, where the server MUST send response before closing > the connection: http://tools.ietf.org/html/rfc3501#section-3.4 > > => the problem is caused by dovecot 2.2.13 bug ... its behaviour is > inconsistent (LOGOUT in non-authenticated state works per RFC > requirement if no SSL is used and doesn't conform to RFC if SSL is > used). It is possible that the problem is related to their DoS-attack > modification, which has most probably unexpected side-effect. Maybe this is related to the DDoS-protection measures that have been added in dovecot 2.2.13. Would apprechiate if someone could have a look. cu, -- Hanno Böck http://hboeck.de/ mail/jabber: hanno(a)hboeck.de GPG: BBB51E42

3 3

Designing a Dovecot Replicate cluster
by Minhas Listas 19 Aug '14

19 Aug '14

Hi all! Do I need one Dovecot Proxy/Director for each Dovecot Replicate Cluster ? Isn't possible to put a Dovecot Proxy/Director on the border of multiple clusters ? Thanks! -- Thiago Henrique www.adminlinux.com.br

1 0

Health monitoring of backend servers
by Minhas Listas 19 Aug '14

19 Aug '14

Does Dovecot Director health monitoring of backend servers ? Or is poolmon (http://www.dovecot.org/list/dovecot/2010-August/051946.html) the best option for this? Maybe Ldirectord? Thanks! -- Thiago Henrique www.adminlinux.com.br

2 4

[SOLVED] Re: Trying Dovecot Replication with dsync
by Minhas Listas 19 Aug '14

19 Aug '14

> > 1) > Aug 14 15:39:34 ceph1 dovecot: imap(account=domaintest.com): Error: > open(/var/run/dovecot/replication-notify-fifo) failed: Permission denied > > What is the correct perms for /var/run/dovecot/replication-notify-fifo ? > Currently: > # ll /var/run/dovecot/replication-notify-fifo > prw------- 1 root root 0 Aug 14 15:57 > /var/run/dovecot/replication-notify-fifo| > > Solved with: # chown dovemail:dovemail /var/run/dovecot/replication-notify-fifo dovemail is the user for imap processes. > 2) > Aug 14 16:02:34 ceph1 dovecot: dsync-server(account=domaintest.com): > Error: Couldn't create lock */var/lib/imap/account=domaintest.com > <http://domaintest.com>/*.dovecot-sync.lock: No such file or directory > Aug 14 16:02:34 ceph1 dovecot: dsync-local(account=domaintest.com): > Error: Couldn't create lock */var/lib/imap/account=domaintest.com > <http://domaintest.com>/*.dovecot-sync.lock: No such file or directory > > In my production cluster, it is automatically created by dovecot when the > user performs the first login: > */var/lib/imap/*<account> for INDEX > */var/spool/imap/*<account> for Mdbox > This was fixed with correct permission on /var/run/dovecot/replication-notify-fifo Thanks! -- Thiago Henrique www.adminlinux.com.br

2 1

IMAP on 993/SSL or 143/STARTTLS?
by Alex 18 Aug '14

18 Aug '14

Hi, I have a postfix+dovecot-2.2.13 system and have configured it to support IMAPS on 993 with SSL/TLS. I'm noticing with users using Thunderbird, the autodetect defaults to IMAPS on 143 with STARTTLS. Which is preferred? Which is more secure? Which is more common? Why would someone choose one over the other? Can I ask the same question about SMTP and submission? Why would one choose 587 with STARTTLS versus 465 with SSL/TLS? Thanks, Alex

4 3

Segmentation fault while indexing a large mailbox using doveadm..
by Akash 18 Aug '14

18 Aug '14

Hi, I am having the same issue as: http://www.dovecot.org/list/dovecot/2014-April/095681.html although I am using the latest dovecot-core and dovecot-lucene packages from the debian sid repo which already has the patch (http://hg.dovecot.org/dovecot-2.2/rev/5e42a8c18b51) mentioned in the above thread implemented. doveadm(admin): Info: Old.All: Caching mails seq=1..300332 9400/300332 Program received signal SIGSEGV, Segmentation fault. (gdb) bt #0 __wcslen ( s=s@entry=0xb93887a8 L" btnToggle: \"Change size\"}</script>\r\n<link href=\"http://s26 ucoz net/src/ulightbox/ulightbox css\" rel=\"stylesheet\" type=\"text/css\" /><script type=\"text/javascript\" src=\"http://s26 ucoz net/src/ulightb"...) at wcslen.c:36 #1 0xb7c026fd in wcsdup ( s=0xb93887a8 L" btnToggle: \"Change size\"}</script>\r\n<link href=\"http://s26 ucoz net/src/ulightbox/ulightbox css\" rel=\"stylesheet\" type=\"text/css\" /><script type=\"text/javascript\" src=\"http://s26 ucoz net/src/ulightb"...) at wcsdup.c:29 #2 0xb7a65660 in lucene::document::Field::Field(wchar_t const*, wchar_t const*, int, bool) () from /usr/lib/i386-linux-gnu/libclucene-core.so.1 #3 0xb7b5cd48 in lucene_index_build_more () from /usr/lib/dovecot/modules/lib21_fts_lucene_plugin.so #4 0xb7b5b7b9 in ?? () from /usr/lib/dovecot/modules/lib21_fts_lucene_plugin.so #5 0xb7b6ac2a in fts_backend_update_build_more () from /usr/lib/dovecot/modules/lib20_fts_plugin.so #6 0xb7b6b674 in ?? () from /usr/lib/dovecot/modules/lib20_fts_plugin.so #7 0xb7b6bbf4 in fts_build_mail () from /usr/lib/dovecot/modules/lib20_fts_plugin.so #8 0xb7b713cc in ?? () from /usr/lib/dovecot/modules/lib20_fts_plugin.so #9 0xb7e4969b in mail_precache () from /usr/lib/dovecot/libdovecot-storage.so.0 #10 0xb7fa1c7f in ?? () #11 0xb7f9d133 in _start () (gdb) # dovecot --version 2.2.13 Please help.. -Thanks, Akash

2 2

Dovecot failing with exit code 19200 in cron job
by Murray Trainer 18 Aug '14

18 Aug '14

Hi All, I am synchronizing a few hundred dovecot mailboxes with a perl script via a crontab. It works for a while (ie exit status=0) but the rest then start failing with exit code 19200. I tried redirecting STDERR but showed nothing. I suspect the cron job is hitting some sort of resource limit? If I run the perl script from the command line it appears to dsync all the accounts OK. The relevant part of my Perl code is below. Any ideas? Thanks, Murray foreach (@userlist) { my $email = $_; chomp ($email); # get rid of CR logit("Started sync of mailbox $email") if $debug; my $dsync_command="dsync -v -D -u $email mirror ssh -C root@$mailstore dsync -u $email"; logit("dsync_command=$dsync_command"); # Run dsync of mailbox logit("Started secondary sync of $mailbox to $mailstore") if $debug; my $exit_status = system ( $dsync_command ); logit("Dsync exit status: $exit_status") if $debug; logit("Finished secondary sync of $mailbox to $mailstore") if $debug; }

2 1

NFS INTR mount option
by Murray Trainer 18 Aug '14

18 Aug '14

Hi again, We have our mailboxes on several NFS filesystems mounted with the options below. 10.11.0.238:/mailbox_store_01 on /home1 type nfs4 (rw,noatime,vers=4.0,rsize=65536,wsize=65536,namlen=255,hard,nordirplus,proto=tcp,port=0,timeo=600,retrans=2,sec=sys,clientaddr=10.11.0.95,local_lock=none,addr=10.110.238) On the rare occasions that the NFS mounts hang, the system load goes through the roof and I often have to reboot the mailstore server as dovecot process are hung and can't be killed. Will the NFS INTR mount option help in this situation? Thanks Murray Send EMAIL [1] Links: ------ [1] https://webmail.westnet.com.au/index.php/mail#

1 0