- dovecot - dovecot.org

Sieve replication - Debian 9.11 - Dovecot 2.2.7 - script changes not replicated
by Daniel Botting 14 Oct '19

14 Oct '19

Hi, I am trying to setup Sieve replication, it is almost working: 1/ Create a script - sync'd across 2/ Activate/de-activate a script - change sync'd 3/ Delete a script - change sync'd But if I make a change to a script it is not replicated, unless I de-activate and activate again. I am seeing in the logs the below during this process (de-activating/activating): Oct 9 17:01:39 hostname dovecot: dsync-local(a_user): Debug: doveadm-sieve: Iterating Sieve mailbox attribute: vendor/vendor.dovecot/pvt/server/sieve/files/0910-active-1640 Oct 9 17:01:39 hostname dovecot: dsync-local(a_user): Debug: doveadm-sieve: Value missing for key `vendor/vendor.dovecot/pvt/server/sieve/default' (last change: 2019-10-09 17:01:37) # 2.2.27 (c0f36b0): /etc/dovecot/dovecot.conf # Pigeonhole version 0.4.16 (fed8554) # OS: Linux 4.9.0-3-amd64 x86_64 Debian 9.11 auth_debug = yes auth_debug_passwords = yes 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_debug = 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 imap4flags copy include variables body enotify environment mailbox date index ihave duplicate mime foreverypart extracttext editheader imapflags 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:replication_server_url:port_number sieve = ~/.dovecot.sieve sieve_dir = ~/sieve sieve_extensions = +editheader +imapflags } protocols = " imap sieve pop3" replication_max_conns = 4 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 = 12345 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 = 4190 } 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/certs/a_domain/fullchain.pem ssl_client_ca_dir = /etc/ssl/certs ssl_key = # hidden, use -P to show it userdb { args = /etc/dovecot/dovecot-ldap.conf.ext driver = ldap override_fields = home=/var/mail/%u uid=mail gid=mail } verbose_ssl = yes 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 Botting

2 1

Password issue
by ＠lbutlr 13 Oct '19

13 Oct '19

One of my accounts was having login failures when trying to send mail, but was able to check mail. I tried everything I could think of to see what the issue might be, but eventually went in and reset the password in the sql database (I knew the password, so I reset it to the same password). {SHA256-CRYPT}$5$VuS… {SHA256-CRYPT}$5$VI7… So the password was updated properly. Clients can still login to check mail, but are failing to send mail, so the password was not the issue. First, logins in to check mail and succeeds > Oct 09 16:02:16 imap-login: Info: Login: user=<myuser(a)covisp.net>, xx.xx.xx.xx, PLAIN, TLS Sends device MUA info > Oct 09 16:02:16 imap(myuser(a)covisp.net)<84553><bP4acIGUJttJDqGg>: Info: ID sent: name=iPad Mail, version=17A860, os=iOS, os-version=13.1.2 (17A860) Fails to send mail > Oct 09 16:02:50 imap-login: Info: Aborted login (auth failed, 5 attempts in 33 secs): user=<myuser(a)covisp.net>, xx.xx.xx.xx, PLAIN, TLS Postfix logs "Client host rejected: Access denied” but as I said, other accounts can submit and there’s nothing special in the submission service in master.cf. Passwords are all copied/pasted so identical in both IMAP (p993) and SMTP (p587) settings. Settings for the SMTP server are identical to the other accounts using the same server. I don’t know how to login to IMAP via the openssl command line interface, but this is what happens when I login to the submission port.  % openssl s_client -connect mail.covisp.net:587 -starttls smtp … auth login 334 VXNlcm5hbWU6 base64(username)= 334 UGFzc3dvcmQ6 base64(pasword)= DONE (Same results without the terminating ‘=‘) I’m sure I am missing something really obvious? -- Lithium will no longer be available on credit

6 12

Error: SSL_accept() syscall failed
by C. James Ervin 11 Oct '19

11 Oct '19

In setting up my new mail server, I am getting the following in the logs: Oct 11 07:10:59 kumo dovecot[5704]: imap-login: Disconnected (no auth attempts in 0 secs): user=<>, rip=24.53.79.10, lip=172.26.12.90, *TLS handshaking: SSL_accept() syscall failed: Success*, session=<B9OokqCUD+UYNU8K> I have tried various ssl_protocols entries, but for now have defaulted back to ssl_protocols = !SSLv3 (the "out of the box" setting). The certificate (ssl_cert = </etc/ssl/certs/certificate_and_key.crt) is valid and the same one used by postfix. However, I belive the error appeared only after removing the self signed certificate and installing the one I purchased. Warning: though I was a sysadmin a long long time, it has been a while! I have not run a mail server in 15 years, but am moving my mail back "in-house" as I approach retirement! I searched for a forum where I could post this issue, but only found this list. # 2.2.33.2 (d6601f4ec): /etc/dovecot/dovecot.conf # Pigeonhole version 0.4.21 (92477967) # OS: Linux 4.15.0-1051-aws x86_64 Ubuntu 18.04.1 LTS auth_mechanisms = plain login mail_location = maildir:~/Maildir mail_privileged_group = mail 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 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 { driver = pam } plugin { sieve = ~/.dovecot.sieve sieve_dir = ~/sieve } protocols = " imap sieve pop3" service auth { unix_listener /var/spool/postfix/private/dovecot-auth { group = postfix mode = 0660 user = postfix } } ssl_cert = </etc/ssl/certs/certificate_and_key.crt ssl_cipher_list = ALL:!ADH:!LOW:!EXP:!aNULL:+HIGH:+MEDIUM ssl_client_ca_dir = /etc/ssl/certs ssl_key = # hidden, use -P to show it userdb { driver = passwd } protocol lda { deliver_log_format = msgid=%m: %$ mail_plugins = sieve postmaster_address = postmaster quota_full_tempfail = yes rejection_reason = Your message to <%t> was automatically rejected:%n%r } protocol imap { imap_client_workarounds = delay-newmail mail_max_userip_connections = 10 } protocol pop3 { mail_max_userip_connections = 10 pop3_client_workarounds = outlook-no-nuls oe-ns-eoh }

3 3

Re: Broken link on doc.dovecot.org
by Karen Woodman 11 Oct '19

11 Oct '19

Hi again, I wanted to check in and see if you got my note about the broken link on your site. Thanks! Karen On Monday, October 7, 2019 at 5:05 PM, Karen Woodman <karen(a)getmailbird.co> wrote: > Hi there, > > I noticed that you have a broken link to a website called Qmail.org. That > site was first published 23 years ago (back in 1996!) but unfortunately, it > is no longer a working website. > > You link to Qmail.org on this page: > https://doc.dovecot.org/configuration_manual/authentication/checkpassword/ > > We just published an article that explores what happened to the site. I think > it's an interesting story, and it could be useful to your readers. > > In addition, since all of the original articles are gone, we included a > curated list of the best Qmail resources and mirrors. > > Here's the article if you want to take a look: > https://www.getmailbird.com/what-resources-are-available-now-that-qmail-org… > > Would you consider swapping out the broken link for our article? > > Please let me know if you have any questions and thank you for your time. > > Sincerely, > -Karen > > -- > Karen Woodman, Editor > 5 Ross Rd > Durham, NH 03824 > > By the way, if you didn't like getting this email you can unsubscribe > <http://ec2-52-26-194-35.us-west-2.compute.amazonaws.com/x/u?u=df0298ca-a3e4…> and > I won't email you anymore. :) >

2 1

Using maldet
by Ignacio García 10 Oct '19

10 Oct '19

Hi there. Sometimes I notice that some malware gets through amavis+clamav+postfix and lands in the user's maildir files. I wonder if it would make sense, or even if it would be advisable to scan and clean those files using a program as maldet Thanks for your help. Ignacio

1 0

Case sensitivity in :addresses in sieve vacation scripts
by Julian Kippels 10 Oct '19

10 Oct '19

Hi, I have recently updated from Dovecot 2.2 to 2.3. Since I have noticed that vacation responses from sieve are not working the same anymore. For example, my sieve script looks like this: vacation :days 1 :addresses ["Kippels(a)hhu.de","julian.kippels(a)hhu.de"] it used to be that I got a vacation response if I sent a mail to kippels(a)hhu.de and Kippels(a)hhu.de. Now I only get a response for Kippels(a)hhu.de, not for kippels(a)hhu.de. I cant say for sure, but I suppose this behaviour changed with the update. If not, why could this have happened? And in any case, how can I enable case insensitivity? Thanks Julian

2 2

Show messages within 60 days
by jacky HU 10 Oct '19

10 Oct '19

Hello, Can my inbox be set up, for example, only for messages within 60 days? I don't want to create a new virtual mailbox. Can anyone help me?

2 1

Dovecot core debug
by MediumDave 10 Oct '19

10 Oct '19

Hi all, I have a Dovecot core that I am trying to track down, in an attempt to replicate so I can log a bug report, and am hoping others might be able to help me as I’m reaching my limit. My working theory is that two user sessions are active at the same time, one of the sessions is “mid-search” (or perhaps in idle state) on a virtual folder where the number of results is zero (due to a null results array), when another session either performs an append or perhaps a new message arrives in the virtual view and qualifies for the search. It seems that the current list of UID’s in the search results is zero (null), and new messages are being appended to the search results, which results in a divide by zero. But I have not been able to create a scenario to replicate this on my test box, but is occurring regularly (every hour or so) on my mail server. I have been able to replicate on my test box using a single account by performing many random operations over and over again, with multiple sessions, but took over 24 hours and I can’t isolate the action that caused it. Dovecot version is 2.2.36 (1f10bfa63) as shipped with Centos 7 (latest update). If I can reproduce, I can then test the latest 2.3 and see if this is still a problem in that version. Here is the full back trace: #0 0x00007f8d40636e58 in array_count_i (array=0x557be80cc168) at array.h:155 No locals. #1 array_get_i (count_r=<synthetic pointer>, array=0x557be80cc168) at array.h:209 No locals. #2 seq_range_lookup (array=0x557be80cc168, seq=2222, idx_r=0x7ffead4141d4) at seq-range-array.c:14 data = <optimized out> idx = <optimized out> left_idx = <optimized out> right_idx = <optimized out> count = <optimized out> #3 0x00007f8d406379ae in seq_range_exists (array=array@entry=0x557be80cc168, seq=seq@entry=2222) at seq-range-array.c:391 idx = 32766 #4 0x00007f8d408fd3d9 in mailbox_search_result_add (result=result@entry=0x557be80cc150, uid=2222) at mailbox-search-result.c:121 __FUNCTION__ = "mailbox_search_result_add" #5 0x00007f8d40962f30 in index_search_result_update_appends (result=result@entry=0x557be80cc150, old_messages_count=<optimized out>) at index-search-result.c:168 t = 0x557be7fb1820 search_ctx = 0x557be7fef860 mail = 0x557be8100198 search_arg = {next = 0x557be7fab460, type = SEARCH_SEQSET, value = {subargs = 0x0, seqset = {arr = {buffer = 0x557be7f69208, element_size = 8}, v = 0x557be7f69208, v_modifiable = 0x557be7f69208}, str = 0x0, time = 0, size = 0, flags = 0, search_flags = (unknown: 0), date_type = 0, thread_type = MAIL_THREAD_NONE, modseq = 0x0, search_result = 0x0, mime_part = 0x0}, initialized = {search_args = 0x0, keywords = 0x0, mailbox_glob = 0x0}, context = 0x0, hdr_field_name = 0x0, match_not = 0, match_always = 0, nonmatch_always = 0, fuzzy = 0, no_fts = 0, result = -1} message_count = 2008 ret = <optimized out> __FUNCTION__ = "index_search_result_update_appends" #6 0x00007f8d4096dc37 in search_result_update (result=0x557be80cc150, ctx=0x557be7ff8950) at index-sync-search.c:78 No locals. #7 index_sync_search_results_update (ctx=ctx@entry=0x557be7ff8950) at index-sync-search.c:88 results = 0x557be7f9d480 i = <optimized out> count = <optimized out> #8 0x00007f8d4096e4b4 in index_mailbox_sync_deinit (_ctx=0x557be7ff8950, status_r=0x557be7ff0108) at index-sync.c:353 ctx = 0x557be7ff8950 sync_rec = {seq1 = 0, seq2 = 0, type = MAILBOX_SYNC_TYPE_EXPUNGE} delayed_expunges = false ret = <optimized out> #9 0x00007f8d40950400 in mailbox_list_index_sync_deinit (ctx=0x557be7ff8950, status_r=0x557be7ff0108) at mailbox-list-index.c:976 ---Type <return> to continue, or q <return> to quit--- ibox = 0x557be7fd0bf0 box = 0x557be7f9d078 #10 0x00007f8d408ed568 in mailbox_sync_deinit (_ctx=<optimized out>, status_r=<optimized out>) at mail-storage.c:1953 ctx = <optimized out> box = <optimized out> errormsg = <optimized out> error = MAIL_ERROR_NONE ret = <optimized out> #11 0x0000557be6613431 in imap_sync_finish (ctx=ctx@entry=0x557be7ff0090, aborting=aborting@entry=false) at imap-sync.c:279 client = 0x557be7f8a6d8 ret = 0 #12 0x0000557be6613fed in imap_sync_more (ctx=0x557be7ff0090) at imap-sync.c:594 str = <optimized out> ret = 1 __FUNCTION__ = "imap_sync_more" #13 0x0000557be6614676 in cmd_sync_continue (sync_cmd=0x557be7f8be88) at imap-sync.c:629 cmd = 0x557be7f75e40 prev = <optimized out> client = 0x557be7f8a6d8 ctx = 0x557be7ff0090 ret = <optimized out> __FUNCTION__ = "cmd_sync_continue" #14 0x0000557be6614b34 in cmd_sync_client (sync_cmd=0x557be7f8be88) at imap-sync.c:720 ctx = <optimized out> flags = <optimized out> client = 0x557be7f8a6d8 imap_flags = <optimized out> no_newmail = <optimized out> #15 cmd_sync_delayed_real (client=0x557be7f8a6d8) at imap-sync.c:833 cmd = <optimized out> first_expunge = <optimized out> first_nonexpunge = <optimized out> #16 cmd_sync_delayed (client=client@entry=0x557be7f8a6d8) at imap-sync.c:841 _data_stack_cur_id = 3 #17 0x0000557be6607a18 in client_handle_input (client=client@entry=0x557be7f8a6d8) at imap-client.c:1233 remove_io = <optimized out> handled_commands = <optimized out> __FUNCTION__ = "client_handle_input" #18 0x0000557be6607eb5 in client_input (client=0x557be7f8a6d8) at imap-client.c:1264 ---Type <return> to continue, or q <return> to quit--- cmd = 0xcfee4 output = 0x557be7f8bd90 bytes = 2052 __FUNCTION__ = "client_input" #19 0x00007f8d40624be2 in io_loop_call_io (io=0x557be7f84cb0) at ioloop.c:600 ioloop = 0x557be7f719c0 t_id = 2 __FUNCTION__ = "io_loop_call_io" #20 0x00007f8d406262cf in io_loop_handler_run_internal (ioloop=ioloop@entry=0x557be7f719c0) at ioloop-epoll.c:223 ctx = 0x557be7f734e0 events = <optimized out> list = 0x557be7f74f40 io = <optimized out> tv = {tv_sec = 1795, tv_usec = 114881} events_count = <optimized out> msecs = <optimized out> ret = 1 i = 0 call = <optimized out> __FUNCTION__ = "io_loop_handler_run_internal" #21 0x00007f8d40624c7c in io_loop_handler_run (ioloop=ioloop@entry=0x557be7f719c0) at ioloop.c:649 No locals. #22 0x00007f8d40624e38 in io_loop_run (ioloop=0x557be7f719c0) at ioloop.c:624 __FUNCTION__ = "io_loop_run" #23 0x00007f8d405a9c03 in master_service_run (service=0x557be7f71860, callback=callback@entry=0x557be6614e30 <client_connected>) at master-service.c:724 No locals. #24 0x0000557be65fa262 in main (argc=1, argv=0x557be7f715f0) at main.c:484 set_roots = {0x557be6825700 <imap_setting_parser_info>, 0x7f8d40df95a0, 0x0} login_set = {auth_socket_path = 0x557be7f69048 "\001", postlogin_socket_path = 0x0, postlogin_timeout_secs = 60, callback = 0x557be6615600 <login_client_connected>, failure_callback = 0x557be6614f40 <login_client_failed>, request_auth_token = 1} service_flags = <optimized out> storage_service_flags = <optimized out> username = 0x0 auth_socket_path = <optimized out> c = <optimized out> And the output of array in frame #0: (gdb) f 0 #0 0x00007f8d40636e58 in array_count_i (array=0x557be80cc168) at array.h:155 155 return array->buffer->used / array->element_size; (gdb) p *array $2 = {buffer = 0x0, element_size = 0} Here is the virtual folder config in use: # cat /etc/dovecot/virtual/combined/dovecot-virtual !INBOX INBOX/Sent INBOX/Drafts all Any help would be greatly appreciated in diagnosing the cause here. Thanks, Dave

1 0

Panic at "doveadm backup" (mailbox_symlink_exists): assertion failed: (ret > 0)
by Alex La 09 Oct '19

09 Oct '19

Hi. I try to sync from a dovecot server to another server via the following command. doveadm -D -o mail_fsync=never -o imapc_user=info(a)DOMAIN.COM -o imapc_password='PASS' backup -R -u info(a)DOMAIN.COM imapc: After some short runtime I get the following panic. ``` dsync(info(a)DOMAIN.COM): Panic: file mailbox-alias-plugin.c: line 77 (mailbox_symlink_exists): assertion failed: (ret > 0) dsync(info(a)DOMAIN.COM): Error: Raw backtrace: /usr/lib64/dovecot/libdovecot.so.0(+0xa17ce) [0x7ff2f4da77ce] -> /usr/lib64/dovecot/libdovecot.so.0(default_fatal_handler+0x2a) [0x7ff2f4da783a] -> /usr/lib64/dovecot/libdovecot.so.0(i_fatal+0) [0x7ff2f4d36dd7] -> /usr/lib64/dovecot/lib20_mailbox_alias_plugin.so(+0x133e) [0x7ff2f3afa33e] -> /usr/lib64/dovecot/lib20_mailbox_alias_plugin.so(+0x14d4) [0x7ff2f3afa4d4] -> /usr/lib64/dovecot/lib01_acl_plugin.so(+0xf2aa) [0x7ff2f432a2aa] -> /usr/lib64/dovecot/lib01_acl_plugin.so(+0xf908) [0x7ff2f432a908] -> /usr/lib64/dovecot/libdovecot-storage.so.0(mailbox_list_iter_next+0x44) [0x7ff2f50ec5e4] -> doveadm(dsync_mailbox_tree_fill+0x19b) [0x558004586e5b] -> doveadm(dsync_brain_mailbox_trees_init+0x124) [0x55800457b204] -> doveadm(dsync_brain_run+0x3d7) [0x558004578317] -> doveadm(+0x29b71) [0x55800455eb71] -> doveadm(+0x2bd2c) [0x558004560d2c] -> doveadm(+0x2c92a) [0x55800456192a] -> doveadm(doveadm_mail_try_run+0x215) [0x558004562285] -> doveadm(main+0x46a) [0x5580045513fa] -> /lib64/libc.so.6(__libc_start_main+0xf5) [0x7ff2f495a505] -> doveadm(+0x1c565) [0x558004551565] ``` There is this message in the archive wich mention that similar directory names could be a problem. https://dovecot.org/list/dovecot/2016-December/106528.html I have cleanuped the similar directories but I still get the panic. This are the main folders in the Account ``` ... some custom folders =Trash =Sent Messages =Junk =Drafts =Coaching =Archiv =INBOX ``` Dovecot version # 2.2.36.4 (baf9232c1): /etc/dovecot/dovecot.conf # Pigeonhole version 0.4.24.2 (aaba65b7) # OS: Linux 3.10.0-1062.1.2.el7.x86_64 x86_64 CentOS Linux release 7.7.1908 (Core) Thanks for any help Regards Alex

1 0

Recommendations for shared storage
by Dan 09 Oct '19

09 Oct '19

Hello all, This email is by no means an "Educate me step by step" query, But more of an inquiry about which direction to go based on your experience. I've been working with Dovecot for some years now, so the configuration of that piece is no stranger. However, Choosing the right style of shared storage has proven to be somewhat of a challenge. A bit of background of what I need to provide service for: ~80,000 users, that will potentially grow throughout the future ~100 unique domains Ability to scale horizontally. As resources grow, adding a node (or group of nodes) with the exact configuration as the other nodes. Director should be able to send the connection to a node that has the same data as it's neighbors, Allowing high availability. What I've experimented with so far: _Server pairs, using dsync to replicate data between the nodes - Director to send users to a specific pair (director_tag returned from passdb) _ Pros: - Easy way to replicate data, without the need for external tools Cons: - Limited to one pair of nodes - dsync doesn't replicate all sieve files, for example, one file that tracks out of office replies has not been replicated in my experience _NFS storage managed by corosync and pacemaker - Director to send users to any backend _ Pros: - One location for all data. New nodes that get stood up will be easy to configure Cons: - Complex to set up In your experience, what did you choose for mail storage? Thanks.

1 0