October 2021 - dovecot

What is the state of SMTPUTF8 support?
by Sylvhem 02 Jun '22

02 Jun '22

Hi! It’s been some time since I last saw a mention of SMTPUTF8 here. Last time I heard about it, it was worked on. Does anybody have news about that? -- Sylvhem

5 8

Shared mailboxes, users with dots and a bug in subscriptions
by Tobias Stein 17 May '22

17 May '22

Hi, i'm running Dovecot 2.3.14.alpha0 with shared namespaces and stumbled across some errors messages logged, when the list of subscribed mailboxes is queried by a client. For every distinct account of in the list of subscriptions two corresponding lines are logged: Jan 28 11:42:34 mx1 dovecot: auth: missing passwd file: /etc/dovecot/private/example/users Jan 28 11:42:34 mx1 dovecot: auth: missing passwd file: /etc/dovecot/private/example/users Jan 28 11:42:34 mx1 dovecot: auth: missing passwd file: /etc/dovecot/private/example/users Jan 28 11:42:36 mx1 dovecot: imap(example_user(a)example.com)<3638><xl6li/O5VsVfWxyn>: Error: mkdir(/var/run/dovecot/user-not-found/noc@example) failed: Permission denied (euid=109(vmail) egid=118(vmail) missing +w perm: /var/run/dovecot, dir owned by 0:0 mode=0755) Jan 28 11:42:36 mx1 dovecot: imap(example_user(a)example.com)<3638><xl6li/O5VsVfWxyn>: Error: mkdir(/var/run/dovecot/user-not-found/info@example) failed: Permission denied (euid=109(vmail) egid=118(vmail) missing +w perm: /var/run/dovecot, dir owned by 0:0 mode=0755) Jan 28 11:42:36 mx1 dovecot: imap(example_user(a)example.com)<3638><xl6li/O5VsVfWxyn>: Error: mkdir(/var/run/dovecot/user-not-found/root@example) failed: Permission denied (euid=109(vmail) egid=118(vmail) missing +w perm: /var/run/dovecot, dir owned by 0:0 mode=0755) Similar messages are logged, when invalid entries are listed in '/var/lib/dovecot/db/shared-mailboxes', which i already pruned and haven't received them anymore since. I think these errors are caused by an unintended behaviour when writing "~/Maildir/subscriptions", which looks (shortened) like this. V 2 INBOX/INBOX shared/noc@example com/INBOX shared/info@example com/INBOX shared/root@example com/test shared/root@example com/test test_sub The subscription-file.c explodes the name on every hierarchy separator ('.','\0') and inserts a TAB character. Unfortunately it also explodes on the DNS label delimiter „.‟. This should probably be fixed by passing a structure containing the required information to the formatter to distinguish mailboxes from domain-names. Subscription in combination with multiple domains and shared mailboxes seems broken to me. Actually i can't even explain to me, why it is working in face of the errors. :-) Unfortunately in Maildir++ the separator dot is hard-coded. There is a very old thread on this mailing list, that suggests using „auth_username_translation‟ to replace dots with a different character, but this idea is getting worse the longer i think about it. I absolutely dislike the idea to set LAYOUT=fs, namespace/separator = § to change the separators to split on, because this would mean to restructure the physical layout of all mailboxes (hierarchically) and mess around with lots of files. I attached a dovecot-sysreport to reproduce the behaviour. /etc/dovecot/private/example.com/users looks like this: ### user:password:uid:gid:(gecos):home:(shell):extra_fields noc:{SSHA512}_hash_:::::: info:{SSHA512}_hash_:::::: root:{SSHA512}_hash_:::::: Please correct me if i'm wrong or point me to a workaround, but i think the layout code needs some love. :-) Best regards Tobias

2 4

FTS solr or xapian and Public Folders
by David Williams 13 May '22

13 May '22

Hi all I hope someone can help with this. I'm looking at setting up FTS on a Dovecot server which has some quite big pubic folders (100,000+ emails each) and would appreciate some info about how fts both Xapian and Solr works with public folders. As background of the folders can receive emails and I use a dummy user and an alias file to deliver them for example testfolder: dummy+Public/MYTESTFOLDER(a)example.com delivers email addressed to testfolder(a)example.com to MYTESTFOLDER in the namespace Public. Has anyone successfully setup FTS with public folders using either Xapian or Solr and what happens when one user changes a folder, does that change show up for other users? E.G.: When an email is delivered by dummy and indexed will that email be indexed for other users as well? Similarly what happens that when an email is moved between, deleted from or copied to public folders by one user will that change then show up for the other users or not? I guess the basic question is: Are public folders indexed once and then the indexes are made available to each user with access or are they indexed for each user? Any info or experience that anyone has will be appreciated. Many thanks David

2 1

test-file-cache.c needs #ifdef HAVE_RLIMIT_AS
by Rupert Gallagher 07 Feb '22

07 Feb '22

test-file-cache.c:259:24: error: use of undeclared identifier 'RLIMIT_AS' test_assert(getrlimit(RLIMIT_AS, &rl_cur) == 0); ^ test-file-cache.c:267:24: error: use of undeclared identifier 'RLIMIT_AS' test_assert(setrlimit(RLIMIT_AS, &rl_new) == 0); ^ test-file-cache.c:270:24: error: use of undeclared identifier 'RLIMIT_AS' test_assert(setrlimit(RLIMIT_AS, &rl_cur) == 0); ^ test-file-cache.c:276:24: error: use of undeclared identifier 'RLIMIT_AS' test_assert(setrlimit(RLIMIT_AS, &rl_new) == 0); ^ test-file-cache.c:279:24: error: use of undeclared identifier 'RLIMIT_AS' test_assert(setrlimit(RLIMIT_AS, &rl_cur) == 0); ^

3 9

Dsync replication - delayed replication (Sync lock)
by Daniel Botting 31 Jan '22

31 Jan '22

Hi, *Our setup:* Two Debian 10 machines that are setup to replicate mail between them, we have round robin DNS setup so a user can connect to either server. *What should happen:* Mail is delivered to either server and replicated across straight away to their mailbox on the other server so it does not matter which one they are connected to they will receive it fairly soon after delivery. *What actually happens:* In some instances the user will experience a delayed receipt of messages if they are not connected to the server that the message is initially delivered to, sometimes the delay is 5/10 minutes, we had a recent support ticket submitted where it was over an hour. Error message seen in mail.err: Sep 1 10:16:15 <host name> dovecot: dsync-local(<username>)<WELSGEERTl8FEAAAZJ7A9A>: Error: Couldn't lock /path/to/mailbox/.dovecot-sync.lock: fcntl(/path/to/mailbox/.dovecot-sync.lock, write-lock, F_SETLKW) locking failed: Timed out after 30 seconds (WRITE lock held by pid 3697) Process 3697 is dovecot/doveadm-server. *Doveconf -n output:* # 2.3.10.1 (a3d0e1171): /etc/dovecot/dovecot.conf # Pigeonhole version 0.5.10 (67bf5bd7) # OS: Linux 4.19.0-10-amd64 x86_64 Debian 10.5 # Hostname: <server hostname> auth_verbose = yes default_vsz_limit = 0 doveadm_password = # hidden, use -P to show it first_valid_gid = 8 first_valid_uid = 8 last_valid_gid = 8 last_valid_uid = 8 lda_mailbox_autocreate = yes lda_mailbox_autosubscribe = yes mail_gid = 8 mail_location = maildir:~/Maildir mail_plugins = " notify replication" mail_uid = 8 managesieve_notify_capability = mailto managesieve_sieve_capability = fileinto reject envelope encoded-character vacation subaddress comparator-i;ascii-numeric relational regex im ap4flags copy include variables body enotify environment mailbox date index ihave duplicate mime foreverypart extracttext editheader imapfla gs namespace inbox { inbox = yes location = mailbox Drafts { special_use = \Drafts } mailbox Junk { special_use = \Junk } mailbox Sent { special_use = \Sent } mailbox "Sent Messages" { special_use = \Sent } mailbox Trash { special_use = \Trash } prefix = } passdb { args = /etc/dovecot/dovecot-ldap.conf.ext driver = ldap } plugin { mail_replica = tcps:<server being replicated to>:<port being used> sieve = ~/.dovecot.sieve sieve_dir = ~/sieve sieve_extensions = +editheader +imapflags } postmaster_address = postmaster@<domain address> protocols = " imap sieve pop3" replication_max_conns = 12 service aggregator { fifo_listener replication-notify-fifo { user = mail } unix_listener replication-notify { user = mail } } service auth { unix_listener /var/run/dovecot-exim-bridge { mode = 0660 user = Debian-exim } } service doveadm { inet_listener { port = <port being used for replication> ssl = yes } } service imap-login { inet_listener imap { port = 143 } } inet_listener imaps { port = 993 ssl = yes } process_limit = 512 process_min_avail = 4 service_count = 1 } service imap { process_limit = 1024 } service managesieve-login { inet_listener sieve { port = <sieve port> } process_min_avail = 1 service_count = 8 vsz_limit = 256 M } service managesieve { process_limit = 1024 } service replicator { process_min_avail = 1 unix_listener replicator-doveadm { mode = 0666 } } ssl = required ssl_cert = </path/to/ssl/cert/fullchain.pem ssl_cipher_list = ALL:!LOW:!SSLv2:!EXP:!aNULL ssl_client_ca_dir = /etc/ssl/certs ssl_key = # hidden, use -P to show it ssl_min_protocol = TLSv1.2 userdb { args = /etc/dovecot/dovecot-ldap.conf.ext driver = ldap override_fields = home=/var/mail/%u uid=mail gid=mail } protocol lda { mail_plugins = " notify replication sieve" } protocol imap { mail_max_userip_connections = 32 mail_plugins = " notify replication" } protocol sieve { mail_max_userip_connections = 10 managesieve_logout_format = bytes=%i/%o managesieve_max_line_length = 65536 } Kind regards Daniel

3 3

public folders, ACLs
by Stefan G. Weichinger 06 Dec '21

06 Dec '21

greetings ... I got the following request from a customer where we run dovecot-2.3.15 on a debian server: They have a public folder with all their project-related subfolders inside: namespace { location = maildir:/home/vmail/oeffentlich prefix = Oeffentlich/ separator = / subscriptions = yes type = public } So far we don't have the mail_plugin ACL enabled. They want to be able to create projectfolders that aren't move/delete/rename-able by the users, only by some admins. But the users should be allowed to move mails between different projectfolders etc As far as I understand we can do that with Dovecot ACLs (although I am still figuring out how to properly design them ...). My questions: * if I enable the mail_plugin for ACLs, does that in any way change the behavior of existing mailboxes? afai understand: no. Only if ACL-lists exist in the various mailboxes, these get applied, right? I don't want to break things, and I want to avoid additional complexity for the non-public user mailboxes etc Maybe I can enable the plugin only for that public namespace? * maybe someone has a link to a similar setup? thanks in advance, Stefan

2 8

sieve - segmentation fault
by kjonca＠o2.pl 01 Dec '21

01 Dec '21

I tried to experiment with "index" extension and I found that, in case "0" sieve-filter segfaults. example code was: --8<---------------cut here---------------start------------->8--- if header :index 0 :last :regex "received" "some regex here" { fileinto "dedicated mailbox" ; } --8<---------------cut here---------------end--------------->8--- KJ -- http://stopstopnop.pl/stop_stopnop.pl_o_nas.html I dote on his very absence. -- William Shakespeare, "The Merchant of Venice"

2 2

Can't unsubscribe from public mailbox when dsync multi-master is enabled
by Salatiel Filho 25 Nov '21

25 Nov '21

Hello, I am trying to build a dovecot environment with 2 servers with dsync master<->master replication. Everything works just fine, one can login in any of the servers and see and manage their mails. The problem I see is that one can not unsubscribe from public mailboxes. Let's call the servers server1 and server2, both of them have an imap webui pointing to localhost, just to make the test easier. One subscribes to a public mailbox connected through the webui on server1. A few seconds later the same user can see the subscribed mailbox on the webui on server2 just as expected, but if he tries to unsubscribe it will not work. The public mailbox folder will still be subscribed to and visible. There is absolutely nothing on the logs, but I know it is related to the dsync replication because if I shutdown any of the servers, I will be able to unsubscribe just fine, UNTIL the shutdown server is powered back on and replication starts again, and, in this case, the previously unsubscribed public mailbox will just reappear. Is this a bug or am I missing something in my configuration? Thanks! # dovecot -n # 2.3.16 (7e2e900c1a): /etc/dovecot/dovecot.conf # Pigeonhole version 0.5.16 (09c29328) # OS: Linux 4.18.0-305.19.1.el8_4.x86_64 x86_64 auth_username_format = %Ln doveadm_password = # hidden, use -P to show it first_valid_uid = 1000 mail_debug = yes mail_home = /var/spool/imap/%n mail_location = maildir:/var/spool/imap/%n/mail:VOLATILEDIR=/var/spool/imap/%n mail_plugins = " notify replication" 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 index ihave duplicate mime foreverypart extracttext namespace { list = children location = maildir:/var/spool/imap/salas/mail/:INDEXPVT=/var/spool/imap/%n/pub-seen-idx/ mailbox * { autoexpunge = 60 days } prefix = salas/ separator = / subscriptions = no type = public } namespace inbox { inbox = yes location = mailbox Drafts { auto = subscribe special_use = \Drafts } mailbox Junk { special_use = \Junk } mailbox Sent { auto = subscribe special_use = \Sent } mailbox "Sent Messages" { special_use = \Sent } mailbox Trash { auto = subscribe special_use = \Trash } prefix = separator = / } passdb { args = scheme=CRYPT username_format=%n /etc/dovecot/users driver = passwd-file } passdb { args = failure_show_msg=yes dovecot driver = pam } plugin { mail_replica = tcps:10.10.10.10:26 sieve = file:/var/spool/imap/%n/sieve;active=/var/spool/imap/%n/.dovecot.sieve sieve_before = /var/lib/dovecot/sieve.d/ sieve_global = /var/lib/dovecot/sieve.d/ } protocols = imap lmtp sieve replication_dsync_parameters = -d -N -l 10 -U service aggregator { fifo_listener replication-notify-fifo { mode = 0666 user = dovecot } unix_listener replication-notify { mode = 0666 user = dovecot } } service doveadm { inet_listener { port = 26 ssl = yes } } service replicator { process_min_avail = 1 unix_listener replicator-doveadm { mode = 0666 user = dovecot } } ssl = required ssl_ca = </etc/ssl/certs/mail-cluster-communication_ca.pem ssl_cert = </etc/ssl/certs/mail-cluster-communication.crt ssl_key = # hidden, use -P to show it userdb { args = username_format=%n /etc/dovecot/users default_fields = uid=vmail gid=mail home=/var/spool/imap driver = passwd-file } userdb { driver = passwd override_fields = uid=vmail gid=mail home=/var/spool/imap/%n } protocol lmtp { mail_plugins = " notify replication sieve" } protocol imap { ssl_cert = </etc/ssl/certs/salas.com.crt ssl_key = # hidden, use -P to show it }

1 1

sieve discard option debug
by Jose Manuel 24 Nov '21

24 Nov '21

Hi, i have a question if someone can help me with this. My sieve file configuration is this: cat enabled_script.sieve require ["fileinto"]; if header :contains "subject" "@ggmaail.com" { fileinto "INBOX.Spam"; } elsif header :contains "subject" "TW" { discard; elsif header :contains :comparator "i;octet" "X-Spam-Status" "autolearn=spam" { discard; } When the condition is true and a mail is discard i can not discriminate in the logs what was the reason. 1) "subject" "@ggmail.com" or 2) "subject" "TW" I can only see this message over the logs: February 15th 2021, 17:39:33.000 lmtp(mail(a)mail.domain)<27188><kJhTEMWjKmA0agAAaUS+oQ>: sieve: msgid=<001701d703b9$234a8b80$69dfa280$(a)correo.es> <001701d703b9$234a8b80$69dfa280$(a)correo.es>: marked message to be discarded if not explicitly delivered (discard action) (Note: i changed the domain names) My question is: Can i setup or add any option in my sieve filter for more info about the discard reason ??? Any rules together with the discard option ? I am out of idea about it. Thanks in advanced!!!

2 1

BUG: imapsieve with virtual mailboxes
by Claudemir Todo Bom 23 Nov '21

23 Nov '21

Hi, I've configured virtual mailboxes and it is working well, including the setting of the real mailbox that will receive messages moved to it. Problem starts when I try to use imapsieve plugin with this setup, I receive a panic on the log and the action isn't executed. To make the problem appears, all that its needed is to have a virtual folder defined with the real mailbox configured (! prefix on the virtual configuration file), having the imapsieve options enabled on dovecot configuration and a sieve script defined. Error occurs even if the sieve script is empty. This error was already reported back in 2017: https://www.dovecot.org/list/dovecot/2017-September/109445.html This is the relevant configuration for imapsieve: plugin { imapsieve_mailbox1_before = file:/etc/wwmail/antispamsieve/learn.sieve imapsieve_mailbox1_causes = COPY APPEND imapsieve_mailbox1_name = * sieve_global_path = /var/lib/dovecot/sieve/default.sieve sieve_pipe_bin_dir = /etc/wwmail/antispamsieve/ sieve_plugins = sieve_imapsieve sieve_extprograms } This is the error received: Oct 27 16:25:06 lab dovecot: imap(monitor(a)lab.wwmail.app)<28726><ug8CjVrPIqoAAAAAAAAAAAAAAAAAAAAB>: Panic: file mail-index-map.c: line 558 (mail_index_map_lookup_seq_range): assertion failed: (first_uid > 0) Oct 27 16:25:06 lab dovecot: imap(monitor(a)lab.wwmail.app)<28726><ug8CjVrPIqoAAAAAAAAAAAAAAAAAAAAB>: Error: Raw backtrace: /usr/lib/dovecot/libdovecot.so.0(backtrace_append+0x42) [0x7f8f68fe1b52] -> /usr/lib/dovecot/libdovecot.so.0(backtrace_get+0x1e) [0x7f8f68fe1c6e] -> /usr/lib/dovecot/libdovecot.so.0(+0xff41b) [0x7f8f68ff041b] -> /usr/lib/dovecot/libdovecot.so.0(+0xff4b1) [0x7f8f68ff04b1] -> /usr/lib/dovecot/libdovecot.so.0(+0x5427c) [0x7f8f68f4527c] -> /usr/lib/dovecot/libdovecot-storage.so.0(+0x49e07) [0x7f8f69104e07] -> /usr/lib/dovecot/libdovecot-storage.so.0(+0xf00a9) [0x7f8f691ab0a9] -> /usr/lib/dovecot/libdovecot-storage.so.0(mail_index_lookup_seq+0xf) [0x7f8f691b3e0f] -> /usr/lib/dovecot/modules/lib20_virtual_plugin.so(+0x829c) [0x7f8f68b1029c] -> /usr/lib/dovecot/libdovecot-storage.so.0(mail_set_uid+0x35) [0x7f8f6910a9a5] -> /usr/lib/dovecot/modules/lib95_imap_sieve_plugin.so(+0x8cf2) [0x7f8f68b01cf2] -> /usr/lib/dovecot/libdovecot-storage.so.0(mailbox_transaction_commit_get_changes+0x56) [0x7f8f69117806] -> dovecot/imap(+0x1584d) [0x56322e61784d] -> dovecot/imap(command_exec+0xa4) [0x56322e621cf4] -> dovecot/imap(+0x25bbf) [0x56322e627bbf] -> dovecot/imap(+0x25c6b) [0x56322e627c6b] -> dovecot/imap(+0x25f04) [0x56322e627f04] -> dovecot/imap(client_handle_input+0x1b5) [0x56322e6280d5] -> dovecot/imap(client_input+0x70) [0x56322e628490] -> /usr/lib/dovecot/libdovecot.so.0(io_loop_call_io+0x69) [0x7f8f690067d9] -> /usr/lib/dovecot/libdovecot.so.0(io_loop_handler_run_internal+0x132) [0x7f8f69008642] -> /usr/lib/dovecot/libdovecot.so.0(io_loop_handler_run+0x50) [0x7f8f690086f0] -> /usr/lib/dovecot/libdovecot.so.0(io_loop_run+0x40) [0x7f8f690088b0] -> /usr/lib/dovecot/libdovecot.so.0(master_service_run+0x13) [0x7f8f68f78dd3] -> dovecot/imap(main+0x500) [0x56322e613fc0] -> /lib/x86_64-linux-gnu/libc.so.6(__libc_start_main+0xea) [0x7f8f68d4cd0a] -> dovecot/imap(_start+0x2a) [0x56322e6140ba] I am using packages of Debian Bookworm (testing) but the problem already occured on Debian Buster. Packages installed # dpkg --list | grep dovecot | awk '{print $2 " " $3}' dovecot-antispam 2.0+20171229-1+b8 dovecot-core 1:2.3.16+dfsg1-3 dovecot-imapd 1:2.3.16+dfsg1-3 dovecot-lmtpd 1:2.3.16+dfsg1-3 dovecot-managesieved 1:2.3.16+dfsg1-3 dovecot-mysql 1:2.3.16+dfsg1-3 dovecot-pop3d 1:2.3.16+dfsg1-3 dovecot-sieve 1:2.3.16+dfsg1-3 Please take a look at it. I can help provide more information if needed, but I think all relevant information to reproduce the bug is here. Best regards, Claudemir

2 4