- dovecot - dovecot.org

[Dovecot] dbox single instance storage maintenance
by Daniel Miller 03 Jun '11

03 Jun '11

Does the "doveadm purge" operation delete unused attachments? -- Daniel

2 1

[Dovecot] Dovecot 2 + Director: IMAP+GSSAPI and LMTP-proxying
by Jahnke-Zumbusch, Dirk 03 Jun '11

03 Jun '11

Hi all, I have a setup of some nodes running Dovecot 2.0.13 and Postfix 2.5.6 with storage in NFS and authentication happening with PAM. Poking around in the wiki, I tried out different possibilities but now I am stuck -- and are hoping for your (enlightening) comments and tips. My planned setup for IMAP and LMTP is like this: IMAP ==== 1. A loadbalancer directs new sessions to a Dovecot-Director instance; this extra amount of traffic routing is mostly done for automatic handling in case of a failing Director node; this works fine, the same IP will end always on the same backend, but user-based Directory proxying seems more appropriate, especially taking the NFS-setup and the delivery process into account 2. Then the Director decides where to direct the user's session; this works fine when I let the backend servers do the authentication work 3. The backend Dovecot server will happily provide the user's e-mail For now my section for the passdb in the Director instance is passdb { driver = static args = proxy=y nopassword=y } So the backend will do the authentication of the session. But this setup inhibits using Kerberos, as the TGT is not forwarded to the backend server. I would very much like to provide GSSAPI/Kerberos authentication, which already works fine with the backend servers being directly connected by mail clients. The backend servers are using the PAM driver. I could not figure out, how to setup the passdb entry for the director instance to use PAM (this way enabling GSSAPI/Kerberos) and also giving back the necessary "proxy=y" to make director proxying the IMAP session. Is this setup feasible at all? LMTP ==== 1. Postfix accepts incoming e-mail and decides, which e-mail will go the LMTP-way to be deliverd into a Dovecot mailbox with something like virtual_transport = lmtp:inet:DOVECOT-DIRECTORS-IP:LMTP-PORT-ABC 2. the Dovecot Director instance accepts the LMTP connections and will proxy this to the appropriate==user's current backend server For now Postfix delivers e-mails using LMTP to the _backend_ w/o proxying and everything like SIEVE-filtering works fine: virtual_transport = lmtp:inet:DOVECOT-BACKENDS-IP:LMTP-PORT-XYZ But when I use the Director's IP/Port combination for LMTP, I see an error "554 5.4.6 <the(a)addr.ess> Proxying loops to itself". Where may I specify the port number of the backend server's LMTP-port? I suspect that the missing port number implies that the receiving LMTP-backend server uses the same port and so this would lead to a loop, hence the loop error?! I browsed the Wiki and the archives but still cannot find the information I am looking for. Any hints on that? Cheers, Dirk And here's the director.conf (via doveconf -n): # 2.0.13: /dovecot/code/etc/dovecot-director/director.conf # OS: SunOS 5.10 sun4v auth_debug = yes auth_verbose = yes base_dir = /var/run/director/ debug_log_path = /var/adm/dovecot-director.debug director_mail_servers = <LOCAL-BACKEND-IP> director_servers = <PUBLIC-DIRECTOR-IP> lmtp_proxy = yes login_greeting = Dovecot Director ready. mail_debug = yes 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 passdb { args = proxy=y nopassword=y driver = static } protocols = imap lmtp service auth-worker { user = root } service auth { client_limit = 6000 unix_listener auth-userdb { mode = 0644 user = vmail } } service director { fifo_listener login/proxy-notify { mode = 0666 } inet_listener { port = 9090 } unix_listener director-userdb { mode = 0600 } unix_listener login/director { mode = 0666 } } service imap-login { executable = imap-login director inet_listener imap { port = 10143 } inet_listener imaps { port = 10993 ssl = yes } process_min_avail = 1 } service imap { process_limit = 1024 } service lmtp { inet_listener lmtp { address = <LOCAL-DIRECTOR-IP> port = 24242 } } ssl_cert = </dovecot/code/dovecot-homedir/dovecot.crt ssl_key = </dovecot/code/dovecot-homedir/dovecot.key syslog_facility = local0 protocol lmtp { auth_socket_path = director-userdb } -- Dirk Jahnke-Zumbusch Deutsches Elektronen-Synchrotron DESY IT Information Fabrics Member of the Helmholtz Association D-22603 Hamburg Notkestrasse 85 / 22607 Hamburg T: +49-40-899.81760 F: +49-40-899.41760 dirk.jahnke-zumbusch(a)desy.de

3 2

[Dovecot] doveadm "-S <socket_path>" option enabled in 2.0.8?
by David Warden 03 Jun '11

03 Jun '11

I apologize if this is a stupid question but I just compiled 2.0.8 and am trying to use doveadm to get the quota for a user using the -S <socket_path> option because I have different options for the various local IPs that dovecot listens on. For example, my default quota is filesystem (NFS rquota) and I can get that by doing: doveadm quota get -u warden which tells me: Quota name Type Value Limit % User quota STORAGE 1313476 1766400 74 But when I try to get my quota on a different local IP using the -S flag to doveadm: doveadm quota get -u warden -S 137.238.2.244:143 I am told: get: invalid option -- S The strange thing is doveadm's usage says that the -S option is valid for quite a few commands and in the man pages for those commands it looks like they should work for the way I'm calling doveadm. In the 2.0.8 source I also see code that looks like it should be parsing the -S flag. I can telnet to 127.0.0.1:143 and do ". getquotaroot inbox" which gives me the same information as "doveadm quota get -u warden" and I can telnet to 137.238.2.244:143 and ". getquotaroot inbox" also tells me my quota, but I'm hoping to use doveadm to periodically recalc my Maildir++ quota and update the squat FTS indexes and for that to happen doveadm needs to be able to be told what local IP address/port to connect to (I think). Here's my doveconf -n: # 2.0.8: /etc/dovecot/dovecot.conf # OS: Linux 2.6.18-194.26.1.0.1.el5 x86_64 Red Hat Enterprise Linux Server release 5.5 (Tikanga) nfs auth_cache_size = 10 M auth_master_user_separator = * auth_mechanisms = plain login auth_username_format = %Ln deliver_log_format = msgid="%m" subject="%s" from="%f" size=%p result="%$" first_valid_uid = 0 lda_mailbox_autocreate = yes lda_mailbox_autosubscribe = yes login_log_format_elements = user=%u method=%m rip=%r lip=%l lport=%a mpid=%e encryption=%c mail_location = maildir:/home/%Lu/Maildir:CONTROL=/Mail/mailhome/%Lu/.dovecot:INDEX=/Mail/mailhome/%Lu/.dovecot mail_log_prefix = "service=%s user=%u rip=%r lip=%l " mail_nfs_storage = yes mail_plugins = zlib quota mail_log notify fts fts_squat 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 = INBOX. separator = . type = private } passdb { args = /etc/dovecot/passwd.masterusers driver = passwd-file master = yes } passdb { args = cache_key=%u dovecot driver = pam } plugin { autocreate = Trash autocreate2 = Drafts autocreate3 = Sent autosubscribe = Trash autosubscribe2 = Drafts autosubscribe3 = Sent fts = squat mail_log_events = delete undelete expunge copy mailbox_delete mailbox_rename flag_change append mail_log_fields = uid box msgid size from subject flags mail_log_group_events = yes quota = fs:User quota quota_exceeded_message = Quota exceeded (mailbox for user is full). Please see http://go.geneseo.edu/emailoverquota for help deleting messages while over quota. sieve = /Mail/mailhome/%Lu/.filter.sieve sieve_dir = /Mail/mailhome/%Lu/.sievedir } postmaster_address = postmaster(a)geneseo.edu protocols = imap pop3 sieve quota_full_tempfail = yes service managesieve-login { inet_listener sieve { port = 4190 } } ssl_cert = </etc/pki/dovecot/certs/dovecot.pem ssl_key = </etc/pki/dovecot/private/dovecot.pem userdb { driver = passwd } verbose_proctitle = yes protocol lmtp { mail_plugins = zlib quota mail_log notify fts fts_squat sieve } protocol lda { mail_plugins = zlib quota mail_log notify fts fts_squat sieve zlib } protocol imap { imap_logout_format = bytes(in/out)=%i/%o mail_max_userip_connections = 50 mail_plugins = zlib quota mail_log notify fts fts_squat quota imap_quota autocreate imap_zlib fts fts_squat } protocol sieve { managesieve_logout_format = bytes(in/out)=%i/%o } protocol pop3 { pop3_client_workarounds = outlook-no-nuls oe-ns-eoh pop3_logout_format = bytes(in/out)=%i/%o, top=%t/%p, retr=%r/%b, del=%d/%m, mailbox-size=%s pop3_uidl_format = UID%u-%v } local 137.238.2.244 { mail_location = maildir:/Mail/mailhome/$Lu/mailtest/Maildir:CONTROL=/Mail/mailhome/%Lu/mailtest/.dovecot2-control:INDEX=/Mail/mailhome/%Lu/mailtest/.dovecot2-index namespace { location = prefix = } plugin { quota = maildir:User quota quota_rule = *:storage=200M quota_rule2 = Trash:storage=+50M zlib_save = gz zlib_save_level = 6 } } local 137.238.1.1 { protocol imap { ssl_cert = </etc/pki/tls/certs/imaps.geneseo.edu.crt ssl_key = </etc/pki/tls/private/imaps.geneseo.edu.key } } local 137.238.2.244 { protocol imap { ssl_cert = </etc/pki/tls/certs/mailtest.geneseo.edu.crt ssl_key = </etc/pki/tls/private/mailtest.geneseo.edu.key } } local 137.238.1.1 { protocol pop3 { ssl_cert = </etc/pki/tls/certs/imaps.geneseo.edu.crt ssl_key = </etc/pki/tls/private/imaps.geneseo.edu.key } } local 137.238.2.244 { protocol pop3 { ssl_cert = </etc/pki/tls/certs/mailtest.geneseo.edu.crt ssl_key = </etc/pki/tls/private/mailtest.geneseo.edu.key } }

3 3

[Dovecot] [dovecot] notifications when delivery fails
by Héctor Moreno Blanco 03 Jun '11

03 Jun '11

Hello Timo, We have dovecot 1.2.13. Is there any way to send notifications to the postmaster when a delivery fails? For example, when the dovecot index are corrupted and the mail is not sent to the user (the user who sent the message receives an error informing that the mail has not sent). Thank you very much in advanced. Héctor Moreno Blanco. ______________________ This message including any attachments may contain confidential information, according to our Information Security Management System, and intended solely for a specific individual to whom they are addressed. Any unauthorised copy, disclosure or distribution of this message is strictly forbidden. If you have received this transmission in error, please notify the sender immediately and delete it. ______________________ Este mensaje, y en su caso, cualquier fichero anexo al mismo, puede contener informacion clasificada por su emisor como confidencial en el marco de su Sistema de Gestion de Seguridad de la Informacion siendo para uso exclusivo del destinatario, quedando prohibida su divulgacion copia o distribucion a terceros sin la autorizacion expresa del remitente. Si Vd. ha recibido este mensaje erroneamente, se ruega lo notifique al remitente y proceda a su borrado. Gracias por su colaboracion. ______________________

2 1

[Dovecot] 2.0: deliver run from multiple uids and configuration files
by Arkadiusz Miskiewicz 03 Jun '11

03 Jun '11

I'm trying to run deliver from exim transport in a way that it doesn't need to query userdb AND doesn't need to read configuration files. The problem is that config files are readable for root only and if I run deliver with multiple UIDs then I would have to allow reading config files for everyone. Of course that's not a option because configs contain database passwords and such stuff. I wonder if it is possible to avoid reading config files by deliver? The only solution I see it to run deliver via sudo which doesn't look nice. -- Arkadiusz Miśkiewicz PLD/Linux Team arekm / maven.pl http://ftp.pld-linux.org/

2 1

[Dovecot] LMTP/Sieve/sendmail path
by Andreas Schulze 03 Jun '11

03 Jun '11

Hello, I have a system using postfix multi instance feature. The first, main postfix-instance is a nullclient as suggested by http://www.postfix.org/MULTI_INSTANCE_README.html#split The second instance is for real use. Postfix receive mail and handover them to the dovecot lmtp server. Thats working. The lmtp server supports sieve. One account has activated a vacation rule. So while delivering a mail to this account, dovecot creates a new mail based on the vacation rules. *This one* fail. It is injected in the main postfix instance, which is wrong in this case. I tried to start dovecot with an additional environment MAIL_CONFIG=/etc/postfix-instance2 and enabled alternate_config_directories in main.cf of the first insance. ( http://www.postfix.org/postconf.5.html#alternate_config_directories ) I also configured the userdb to return an extrafield userdb_mail_config=/etc/postfix-instance2. But all produce strance postfix errors I never saw in my last 6 years as postfix-admin. Would it be possible to configure dovecot lmtp server to use an external smtp-server to submit new mails ? Any other Ideas ? Thanks Andreas -- Andreas Schulze Internetdienste | P252 DATEV eG 90329 Nürnberg | Telefon +49 911 319-0 | Telefax +49 911 319-3196 E-Mail info @datev.de | Internet www.datev.de Sitz: 90429 Nürnberg, Paumgartnerstr. 6-14 | Registergericht Nürnberg, GenReg Nr.70 Vorstand Prof. Dieter Kempf (Vorsitzender) Dipl.-Kfm. Wolfgang Stegmann (stellvertretender Vorsitzender) Dipl.-Kfm. Michael Leistenschneider Jörg Rabe v. Pappenheim Dipl.-Vw. Eckhard Schwarzer Vorsitzender des Aufsichtsrates: Reinhard Verholen

4 3

[Dovecot] Baffled after upgrade
by Bert Todger 02 Jun '11

02 Jun '11

Hello all, I have just upgraded my Fedora 13 system to F15. With it came a shiny new dovecot 2.0.13 (Details of dovecot -n below). The configuration is different from the dovecot 1.something that I had been used to with F13, but I managed to get it working and all was well... Without changing anything related to dovecot (as far as I know), after a reboot, dovecot stopped working. I did a bit of digging, upped the verbosity of logging and found this in the logs: =======8<================================================================= Jun 02 09:58:21 auth: Debug: passwd-file /etc/passwd.dovecot: Read 3 users Jun 02 09:58:21 auth: Debug: auth client connected (pid=16699) Jun 02 09:58:21 auth: Debug: client in: AUTH 1 PLAIN service=imap secured lip=192.168.2.2 rip=192.168.2.2 lport=993 rport=33960 Jun 02 09:58:21 auth: Debug: client out: CONT 1 Jun 02 09:58:21 auth: Debug: client in: CONT 1 AG1hcmsAaWxIYWFkSFA= Jun 02 09:58:21 auth: Debug: Loading modules from directory: /usr/lib/dovecot/auth Jun 02 09:58:21 auth: Debug: Module loaded: /usr/lib/dovecot/auth/libauthdb_ldap.so Jun 02 09:58:21 auth: Debug: Module loaded: /usr/lib/dovecot/auth/libdriver_sqlite.so Jun 02 09:58:21 auth: Debug: Module loaded: /usr/lib/dovecot/auth/libmech_gssapi.so Jun 02 09:58:21 auth: Debug: passwd-file /etc/passwd.dovecot: Read 3 users Jun 02 09:58:21 auth: Debug: pam(mark,192.168.2.2): lookup service=dovecot Jun 02 09:58:21 auth: Debug: pam(mark,192.168.2.2): #1/1 style=1 msg=Password: Jun 02 09:58:23 auth: Info: pam(mark,192.168.2.2): pam_authenticate() failed: Authentication failure (password mismatch?) (given password: mypass) Jun 02 09:58:23 auth: Debug: passwd-file(mark,192.168.2.2): lookup: user=mark file=/etc/passwd.dovecot Jun 02 09:58:23 auth: Debug: client out: OK 1 user=mark Jun 02 09:58:23 auth: Debug: master in: REQUEST 1125646337 16699 1 69152b70b5461508ad2a3eb02942b883 Jun 02 09:58:23 auth: Debug: passwd(mark,192.168.2.2): lookup Jun 02 09:58:23 auth: Debug: master out: USER 1125646337 mark system_groups_user=mark uid=500 gid=500 home=/home/mark Jun 02 09:58:23 imap-login: Info: Login: user=<mark>, method=PLAIN, rip=192.168.2.2, lip=192.168.2.2, mpid=16705, TLS =========8<=============================================================== Now it's entirely possible that I simply don't properly understand the password authentication process. I have PLAIN passwords set in /etc/passwd.dovecot which it seems to be reading OK, my /etc/dovecot/conf.d/auth-passwdfile.conf.ext file contains just =========8<============================================= passdb { driver = passwd-file args = scheme=PLAIN username_format=%u /etc/passwd.dovecot } =========8<============================================= and nothing else. ...and as I said - it *was* working fine. I am puzzled by the reference to PAM authentication in the logs and in the dovecot -n output. Should I comment out all references to PAM? (in auth-system.conf.ext?) I and my family access mail through a variety of means, using Evolution on the home network, mobile devices (eg iPhone), webmail (using squirrelmail configured to access dovecot) and over ssh (using mutt). I don't necessarily want my (or their) email passwords to be the same as our PAM logon passwords (and at the moment they're not). So for me PLAIN has been the best solution up until now. What is my best (and simplest) solution? Thanks in advance Mark # dovecot -n # 2.0.13: /etc/dovecot/dovecot.conf # OS: Linux 2.6.38.6-27.fc15.i686.PAE i686 Fedora release 15 (Lovelock) auth_debug = yes auth_debug_passwords = yes auth_verbose = yes auth_verbose_passwords = plain log_path = /var/log/dovecot mbox_write_locks = fcntl passdb { driver = pam } passdb { args = scheme=PLAIN username_format=%u /etc/passwd.dovecot driver = passwd-file } ssl_cert = </etc/pki/dovecot/certs/dovecot.pem ssl_key = </etc/pki/dovecot/private/dovecot.pem userdb { driver = passwd }

2 1

[Dovecot] DOVECOT v2.0.11 using SIEVE not working
by Matt Mc Namara 02 Jun '11

02 Jun '11

Hi, I'm trying to get sieve working with dovecot. I seem to have everything enabled but my scripts done seem to work. both sieve-filter (2000) and sieve (4190) are running along with pop3 and imap. I have a simple script which checks the email address and moves it to a folder. I'm at a loss as to what to check next. Thanks in advance, Matt

5 10

[Dovecot] Unknown setting: dbox_attachment_dir
by imap＠tnde.org 01 Jun '11

01 Jun '11

Hi, I am trying to set up Dovecot with sdbox and sis (single instance storage of attachments), however I get the following error when attempting to start dovecot: Starting Dovecot Imap: doveconf: Fatal: Error in configuration file /etc/dovecot/dovecot.conf line 89: Unknown setting: dbox_attachment_dir I'd appreciate help on getting this up and running. This is a virtual (for the time being) server, running Centos, and Dovecot 2. [root@localhost ~]# dovecot --build-options Build options: ioloop=epoll notify=inotify ipv6 openssl io_block_size=8192 Mail storages: cydir maildir mbox mdbox raw sdbox shared SQL driver plugins: mysql postgresql sqlite Passdb: checkpassword ldap pam passwd passwd-file shadow sql Userdb: checkpassword ldap(plugin) nss passwd prefetch passwd-file sql [root@localhost ~]# dovecot -n # 2.0.13: /etc/dovecot/dovecot.conf # OS: Linux 2.6.18-238.9.1.el5 x86_64 CentOS release 5.6 (Final) auth_debug = yes auth_debug_passwords = yes auth_verbose = yes disable_plaintext_auth = no mail_location = sdbox:~/dbox mbox_write_locks = fcntl passdb { driver = pam } ssl_cert = </etc/pki/dovecot/certs/dovecot.pem ssl_key = </etc/pki/dovecot/private/dovecot.pem userdb { driver = passwd }

2 2

[Dovecot] Client IP log variable for proxy/director setups?
by Kelsey Cummings 01 Jun '11

01 Jun '11

Timo - it would be very convenient if we could get a variable that held the real client ip for use in logging with director/proxy setups. It's clearly already passed around in client->user->remote_ip but at this point, I don't see anyway to get it into the logs and it would take me ages to figure out the Right place to add it since client->user doesn't seem to be available in client-common.c: get_var_expand_table(..). Perhaps I'm missing something obvious? -- Kelsey Cummings - kgc(a)corp.sonic.net sonic.net, inc. System Architect 2260 Apollo Way 707.522.1000 Santa Rosa, CA 95407

1 1