- dovecot - dovecot.org

submission proxy -- where to config/present client cert?
by PGNet Dev 20 Jul '20

20 Jul '20

I've a dovecot instance setup with submission proxy, protocols = imap lmtp submission sieve hostname = internal.mx.example.com submission_relay_host = internal.mx.example.com submission_relay_port = 465 submission_relay_trusted = yes submission_relay_ssl = smtps submission_relay_ssl_verify = yes service submission-login { inet_listener submission { address = 10.2.2.10, 127.0.0.1 port = 50465 ssl = yes } } protocol submission { ssl_cert = < /sec/vmail/mx.example.com.server.EC.crt.pem ssl_key = < /sec/vmail/mx.example.com.server.EC.key.pem ssl_verify_client_cert = no } It submits to a postfix instance, atm on the same host, [internal.mx.example.com]:465 inet n - n - - smtpd -o smtpd_tls_req_ccert=no with that setup, all works as expected. If I turn ON required client cert verification @ postfix, - -o smtpd_tls_req_ccert=no + -o smtpd_tls_req_ccert=yes with that add'l req't, postfix log reports "no client certificate presented" Jul 18 14:52:38 mx postfix/submit-from-dovecot/smtpd[37133]: connect from internal.mx.example.com[10.2.2.10] Jul 18 14:52:39 mx postfix/submit-from-dovecot/smtpd[37133]: Anonymous TLS connection established from internal.mx.example.com[10.2.2.10]: TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature ECDSA (P-384) server-digest SHA384 Jul 18 14:52:39 mx postfix/submit-from-dovecot/smtpd[37133]: NOQUEUE: abort: TLS from internal.mx.example.com[10.2.2.10]: No client certificate presented Jul 18 14:53:09 mx postfix/submit-from-dovecot/smtpd[37133]: disconnect from internal.mx.example.com[10.2.2.10] commands=0/0 Clearly, I'm missing config. What in my dovecot submission config^^ do I need to change/add to get it to 'present' a client cert to the postfix relay for client cert verification?

2 2

Set mail to only display mail within 30 or 60 days
by lty 20 Jul '20

20 Jul '20

Set mail to only display mail within 30 or 60 days, Here are two examples. 1. normal E.g: A654 FETCH 2:4 (FLAGS BODY[HEADER.FIELDS (DATE FROM)]) * 2 FETCH (FLAGS (\Seen) BODY[HEADER.FIELDS (DATE FROM)] {73} Date: Tue, 07 May 2020 12:26:55 +0800 From: <test(a)www.chengmail.ml> ) * 3 FETCH (FLAGS (\Seen) BODY[HEADER.FIELDS (DATE FROM)] {62} Date: Tue, 07 May 2020 12:35:34 +0800 From: lty(a)cndns.com ) * 4 FETCH (FLAGS (\Seen) BODY[HEADER.FIELDS (DATE FROM)] {62} Date: Tue, 07 Jul 2020 17:41:53 +0800 From: lty(a)cndns.com ) A654 OK Fetch completed (0.002 + 0.000 + 0.001 secs). 2. Set to display mail within 30 days E.g: A654 FETCH 2:4 (FLAGS BODY[HEADER.FIELDS (DATE FROM)]) * 4 FETCH (FLAGS (\Seen) BODY[HEADER.FIELDS (DATE FROM)] {62} Date: Tue, 07 Jul 2020 17:41:53 +0800 From: lty(a)cndns.com ) A654 OK Fetch completed (0.002 + 0.000 + 0.001 secs). Is there any way to display the result of case 2? I found lua. https://doc.dovecot.org/admin_manual/lua/#mail-lua But I don't know how to edit the lua script, Are there more examples? I also watched the help, but it doesn't help much. https://doc.dovecot.org/configuration_manual/authentication/lua_based_authe… https://doc.dovecot.org/configuration_manual/push_notification/#lua-push-no… Can Lua fulfill my needs?

1 0

I need some help with my Dovecot and Postfix configs - I'm unable to log in on my mail server
by Nils 20 Jul '20

20 Jul '20

Hello, currently I'm working on my first mailserver but I can't log in to the account I made. For now my client (Mozlla Thunderbird) detects the server config (STARTTLS, Port, ...) but it is "Unable to log in at server. Probbly wrong configuration, username or passsword". I know that I added username and password correctly so there must something else be wrong. I run Debian 10 x64 on a fresh installation. Installed are: postfix, postfix-pcre, dovecot-core, dovecot-imapd and dovecot-pop3d. Dovecot is version 2.3.4.1, Postfix is 3.4.10. Here are my configurations: Postfix main.cf: https://pastebin.com/S17jWDQd Postfix master.cf: https://pastebin.com/W0nynZG6 dovecot.conf: https://pastebin.com/tMD90Znn This is what my server logs (mail.info, mail.warn) tell me: root@bgrsld-mail0:~# tail /var/log/mail.info Jul 17 18:22:08 bgrsld-mail0 postfix/submission/smtpd[8472]: improper command pipelining after EHLO from unknown[192.168.2.110]: QUIT\r\n Jul 17 18:22:08 bgrsld-mail0 postfix/submission/smtpd[8465]: disconnect from unknown[192.168.2.110] ehlo=1 quit=1 commands=2 Jul 17 18:22:08 bgrsld-mail0 postfix/submission/smtpd[8472]: disconnect from unknown[192.168.2.110] ehlo=1 quit=1 commands=2 Jul 17 18:24:59 bgrsld-mail0 postfix/submission/smtpd[8485]: warning: dict_nis_init: NIS domain name not set - NIS lookups disabled Jul 17 18:24:59 bgrsld-mail0 postfix/submission/smtpd[8485]: connect from unknown[192.168.2.110] Jul 17 18:24:59 bgrsld-mail0 postfix/submission/smtpd[8485]: improper command pipelining after EHLO from unknown[192.168.2.110]: QUIT\r\n Jul 17 18:24:59 bgrsld-mail0 postfix/submission/smtpd[8485]: disconnect from unknown[192.168.2.110] ehlo=1 quit=1 commands=2 Jul 17 18:28:19 bgrsld-mail0 postfix/anvil[8473]: statistics: max connection rate 2/60s for (submission:192.168.2.110) at Jul 17 18:22:08 Jul 17 18:28:19 bgrsld-mail0 postfix/anvil[8473]: statistics: max connection count 2 for (submission:192.168.2.110) at Jul 17 18:22:08 Jul 17 18:28:19 bgrsld-mail0 postfix/anvil[8473]: statistics: max cache size 1 at Jul 17 18:22:08 root@bgrsld-mail0:~# tail /var/log/mail.warn Jul 17 18:21:50 bgrsld-mail0 postfix/postfix-script[8290]: warning: symlink leaves directory: /etc/postfix/./makedefs.out Jul 17 18:22:08 bgrsld-mail0 postfix/submission/smtpd[8465]: warning: dict_nis_init: NIS domain name not set - NIS lookups disabled Jul 17 18:22:08 bgrsld-mail0 postfix/submission/smtpd[8472]: warning: dict_nis_init: NIS domain name not set - NIS lookups disabled Jul 17 18:24:59 bgrsld-mail0 postfix/submission/smtpd[8485]: warning: dict_nis_init: NIS domain name not set - NIS lookups disabled (I removed any information from older issues that are already solved, thats why mail.warn misses a few lines and there is no mail.err, mail.err hadn't any newer messages) Here is my only line from dovecot's userpass file (/maildata/userpass): nils@intern.xxx.xxx:{CRYPT}$2y$05$kkxElxFGsLSyTJwbEaPZzO/QA33vGY3oGyHcgEZa9rpklgBTKQw9O My postfix vhosts file (/maildir/vhosts) intern.xxx.xxx My postfix vmaps file (/maildir/vmaps) nils(a)intern.xxx.xxx intern.xxx.xxx/nils/ I've also created the hased database with 'postmap /maildata/vmaps' This is the script I've used to create my user database: https://pastebin.com/9uViNqjB I haven't worked with postfiy or dovecot yet so there are many things I don't know. Because of this a lot of my configurations is copied from an iRedMail installation. I know that taking these configs may cause some problems but I don't know where to change them or what else is wrong or missing. Thank you for any help!

6 11

Sieve and handling multiple addresses
by ＠lbutlr 19 Jul '20

19 Jul '20

I an trying to write a sieve action that will take mail that is addressed to a user only it is filed in a mailbox (e.g. "Direct") if it is to the email AND to any other email address, then file it in a mailbox named, e.g. "Cc". And if the mail is not addressed to the user at all, sieve does nothing with it. Message 1: to foo(a)example.com => Direct Message 2: to/cc foo(a)example.com & to/cc to anyone(a)anydomain.tlf => CC Optional: If it contains a specific header like "X-bypass: secret string" => no action Message 3: NOT to/cc foo(a)example.com => no action I've gotten exactly nowhere so far Something like if :allof (header :contains ["to"] ["foo#example.com"], header :contains ["cc"] ["foo#example.com"]) Takes care of the first part of the message 1 and 2 criteria. But then what? -- Was it everything you wanted to find? and did you miss me while you were looking for yourself out there?

5 8

problem with client using TLS
by ratatouille 18 Jul '20

18 Jul '20

Hello! On my testserver running CentOS8 I have installed dovecot v2.3.8. I can connect to the server using claws-mail on my PC but can't using K9-mail on android device. Jul 18 12:24:57 imap-login: Debug: SSL: where=0x10, ret=1: before SSL initialization Jul 18 12:24:57 imap-login: Debug: SSL: where=0x2001, ret=1: before SSL initialization Jul 18 12:24:57 imap-login: Debug: SSL: where=0x2002, ret=-1: before SSL initialization Jul 18 12:24:57 imap-login: Debug: SSL: where=0x2002, ret=-1: before SSL initialization Jul 18 12:24:57 imap-login: Debug: SSL: where=0x2001, ret=1: before SSL initialization Jul 18 12:24:57 imap-login: Debug: SSL alert: where=0x4008, ret=582: fatal protocol version Jul 18 12:24:57 imap-login: Debug: SSL: where=0x2002, ret=-1: error Jul 18 12:24:57 imap-login: Debug: SSL error: SSL_accept() failed: error:14209102:SSL routines:tls_early_post_process_client_hello:unsupported protocol Jul 18 12:24:57 imap-login: Info: Disconnected (no auth attempts in 0 secs): user=<>, rip=87.182.44.142, lip=185.84.81.115, TLS handshaking: SSL_accept() failed: error:14209102:SSL routines:tls_early_post_process_client_hello:unsupported protocol, session=<uppwr7SqrrpXtiyO> Jul 18 12:24:57 imap-login: Debug: SSL error: SSL_accept() syscall failed: Invalid argument Tried different settings without luck. grep -v '^#' 10-ssl.conf ssl = yes ssl_cert = </etc/letsencrypt/live/smtp.dualbit.de/fullchain.pem ssl_key = </etc/letsencrypt/live/smtp.dualbit.de/privkey.pem ssl_dh = </etc/dovecot/dh.pem ssl_min_protocol = TLSv1.2 ssl_cipher_list = PROFILE=SYSTEM ssl_prefer_server_ciphers = yes Can somebody help solving this? Kind regards Andreas

3 8

Non-standard socket locations w/ Dovecot + AWS Linux
by Antonio Leding 18 Jul '20

18 Jul '20

# === The post relates to Dovecot v2.2 running on AWS Linux v1 === # Hello all… The Dovecot sockets on this server are installed in two different locations one of which is different from what I have on my other Dovecot installations which run on Centos-7. Here are the sockets paths: — all sockets EXCEPT "auth-master” —> "/usr/local/var/run/dovecot” <<<<< this appears to be the default for this server — "auth-master” —> "/var/run/dovecot” <<<<< without this statement, Dovecot authentication always failed. The config stanza for this is shown below: service auth { unix_listener auth-userdb { path = /var/run/dovecot/auth-master mode = 0600 user = vmail } } ============ The above has been working great for several years but the time eventually came where I needed to upgrade. This of course meant I had to move lots of maildata between the servers. Unfortunately, the “auth-userdb" config mentioned above prevented “dsync" from working properly. This is the error I received when trying “dsync”: Error: userdb lookup: connect(/usr/local/var/run/dovecot/auth-userdb) failed: No such file or directory This error led me to the eventual fix which was adding the following to the “dovecot.conf” file: auth_socket_path = /var/run/dovecot/auth-master ============ So, to my questions: (1) Why would the socket path on this AWS installation be different from what I see on my Centos-based Dovecot servers and also what I’ve read in several Dovecot mailer discussions? (2) Can the default socket path be set at compile time? If so, then I'm thinking maybe the AWS folks — for whatever reason — decided to install all sockets (except ”auth-master”) in the “/usr/local/var/run/dovecot” directory. Are there any benefits or reason to do this? (3) Is there any requirement that the “auth-master” socket be located in the “/var/run/dovecot” directory? Also, there is nothing else in that directory — just the “auth-master” socket file — that’s it. (4) Why is “dsync" looking for "auth-userdb" in the “/usr/local/var/run/dovecot” directory despite the “auth" stanza stating it is located elsewhere in "/var/run/dovecot”? (5) And last, why do I need to add the “auth_socket_path” statement to get “dsync” working even though the socket path is stated in the “auth" stanza? Thanks in advance for your time on this. I know this is mostly to help educate me on what is going on so I really appreciate your feedback…

1 0

Namespace issues & duplicate maiboxes
by Antonio Leding 18 Jul '20

18 Jul '20

Hello, I am seeing duplicate mailboxes when I perform the “doveadm mailbox list” command…the output is as follows: ================= INBOX INBOX.Drafts INBOX.Trash INBOX.Sent INBOX.INBOX INBOX.INBOX.Sent Messages INBOX.INBOX.Drafts INBOX.INBOX.Trash INBOX.INBOX.Deleted Messages ================= Here is namespace config that seems to have caused the above duplicates. Please help me understand what I’m doing wrong here…and also, not sure if this is relevant but the client is Apple Mail… ================= namespace inbox { type = private separator = . inbox = yes prefix = INBOX. mailbox Trash { auto = subscribe special_use = \Trash } mailbox Drafts { auto = subscribe special_use = \Drafts } mailbox Sent { auto = subscribe special_use = \Sent } }

2 2

Stuck here - help please
by Kishore Potnuru 18 Jul '20

18 Jul '20

Hi All, I have 2 test servers with the below configuration. ========================== Linux OS - Red Hat Enterprise Linux Server release 7.7 (Maipo) Dovecot version - 2.2.36 (1f10bfa63) Postfix version - 2.10.1 ========================== Trying to create High Availability. I have added both of the above servers behind a F5 load balancer. I have got a Load Balancer FQDN "intl-dev-imaptest.testorg.com". I have enabled/opened the ports (25/110/143/993/995) on the above " intl-dev-imaptest.testorg.com". When I send 10 emails to "intl-dev-imaptest.testorg.com", then those 10 emails are getting distributed between the above 2 backend servers (5 emails to each server). I see those 5 emails each in both the servers. >From Outlook I have configured the email address using "POP and IMAP", when I gave the IMAP server as "intl-dev-imaptest.testorg.com" ,then it shows only 5 emails from server1 in outlook and after a few seconds/minutes, automatically it shows/refreshes the other 5 emails from server2. But I am not seeing all the 10 emails at the same time. why? So I tried the sync command. When I execute sync command like below from server1, it reflects the same emails in other server2 also. Then I see the same number of emails in both the servers. Is it not possible to access the both servers emails at one time with the "sync" command? Do we need to run this on all the email boxes on both servers? don't we miss/lose any emails during this sync process multiple times? "doveadm sync -f -u kishore(a)test.testorg.com remote:vmail@bal3200dev002.testorg.com" Is "replication" and "sync" are same? Why are we not able to see all the emails at one time without the "sync" command? What is the best and easiest way to create High Availability with just 2 servers, like emails should travel to both servers equally and if one server goes down also, another server should take care of the emails/functionality. This is my requirement. My current real time environment: I have around 10 email domains and each domain is having 10 imap emails. In total around 100 email boxes/addresses. We receive around 50K emails in a day to those email addresses. We are using the "Maildir" format in our environment. Want to move to the High Availability option with 2 servers. Please help me to fix the issue. Thanks & Regards, Kishore Potnuru

3 6

Outlook vs Thunderbird
by The Doctor 16 Jul '20

16 Jul '20

Got a client that usually uses Outlook I think 2010. This person tends to move their e-mails to certain folers. On Thunderbird, the move shows. Not on Outlook. Any explanation? -- Member - Liberal International This is doctor@@nl2k.ab.ca Ici doctor@@nl2k.ab.ca Yahweh, Queen & country!Never Satan President Republic!Beware AntiChrist rising! https://www.empire.kred/ROOTNK?t=94a1f39b A full cup must be carried steadily. -English proverb

12 22

Variable expansion in passwords
by lukn 16 Jul '20

16 Jul '20

Hello list A user reports failure to log in after password change. The client side error message indicates a lost or dropped connection by the server. A peek into the server logs reveales this: dovecot: imap(user(a)example.com)<53267><el0HoZGqHIdb0K2G>: Error: Failed to expand plugin setting plain_pass = 'asdf%?asdf': Unknown variable '%?' This looks a lot like a %-variable in the user provided password is tried to be expanded... Is this a bug or did I misconfigure something? output of dovecot -n is attached. regards lukn

1 0

NFS vs Replication
by Andrea Gabellini 16 Jul '20

16 Jul '20

Hello list, I built an email system using a proxy / director pair (IMAP, POP3, LMTP) and a backend pair. To have an HA system, I would like to understand if it is better to use an NFS export or replication to save emails and index files NFS is provided by a NAS (in HA), while for replication I would use the local backend disks Which of the two systems is more reliable? Are there any drawbacks for one or the other? Thanks, Andrea -- __________________________ Buy a Pentium 586/200 so you can reboot faster. __________________________ TIM San Marino S.p.A. Andrea Gabellini Engineering R&D TIM San Marino S.p.A. - https://www.telecomitalia.sm Via Ventotto Luglio, 212 - Piano -2 47893 - Borgo Maggiore - Republic of San Marino Tel: (+378) 0549 886237 Fax: (+378) 0549 886188 -- Informativa Privacy Questa email ha per destinatari dei contatti presenti negli archivi di TIM San Marino S.p.A.. Tutte le informazioni vengono trattate e tutelate nel rispetto della normativa vigente sulla protezione dei dati personali (Reg. EU 2016/679). Per richiedere informazioni e/o variazioni e/o la cancellazione dei vostri dati presenti nei nostri archivi potete inviare una email a privacy(a)telecomitalia.sm. Avviso di Riservatezza Il contenuto di questa e-mail e degli eventuali allegati e' strettamente confidenziale e destinato alla/e persona/e a cui e' indirizzato. Se avete ricevuto per errore questa e-mail, vi preghiamo di segnalarcelo immediatamente e di cancellarla dal vostro computer. E' fatto divieto di copiare e divulgare il contenuto di questa e-mail. Ogni utilizzo abusivo delle informazioni qui contenute da parte di persone terze o comunque non indicate nella presente e-mail potra' essere perseguito ai sensi di legge.

5 8

Error in namespace config; how to move messages
by Antonio Leding 16 Jul '20

16 Jul '20

Hello, I inadvertently made an error re: the namespace prefix when setting up a few accounts and several messages have been saved in the Sent\Deleted folders under the incorrect namespace. Can I simply copy these messages to the relevant folder in the proper namespace? Or do I need to export\import these? Some other procedure? Thanks in advance for any feedback…

2 1

Massive alias / bulk delivery problem
by gnd 16 Jul '20

16 Jul '20

Hello list, we are experimenting with using Dovecot as part of a simplified mail stack with SQL for a internal messaging / notification system for our customers. Sometimes we need to send a message to thousands of customers at once and we do it by having aliases resolve to 10k and 20k+ mail accounts. When we send such a message the server load (obviously) is very high for a few minutes (in case of an alias containing 10k addresses its around 6-7 minutes for a 2 core machine with load around 20-30). Are there any optimizations or ways how to design the system around this problem ? Eg. is there a way how to deliver messages more slowly, and spread the load over longer time ? The stack is very simple with Postfix receiving the messages and Dovecot acting as a LMTP and IMAP server. Mailboxes are in the Maildir format on disk. Customers get the messages on login, a PHP script connects to the mail account associated with their ID via IMAP and sends unseen messages in the form of a JSON. Best regards, gnd/

5 6

doveadm quota question
by hfh＠cndns.com 15 Jul '20

15 Jul '20

I found that when the mailbox name contains special characters such as \x7f, when the doveadm quota recalc -u xxx command is executed, the messages under the mailbox name with special characters will not be counted. doveadm mailbox list -u xxx Actual mailbox name: doveadm quota recalc -u xxx, Mail under 'INBOX.Metaltex' will not be counted. hfh(a)cndns.com

1 0

dovecot 2.2.13
by Santhosh 14 Jul '20

14 Jul '20

all, below command not reporting the actual usage report on dovecot 2.2.13. please share your suggestion to how to fix this issue doveadm -f flow mailbox status -A -t 'messages vsize' '*' is just showing rkreddy messages=0 vsize=0 jcplho messages=0 vsize=0 planning messages=0 vsize=0 accounts messages=0 vsize=0 spam.notify messages=0 vsize=0 jsraju messages=0 vsize=0 amarnath messages=0 vsize=0 kamesh messages=0 vsize=0 moulivarma messages=0 vsize=0 test messages=0 vsize=0 electrical messages=0 vsize=0 subbarao messages=0 vsize=0 contracts messages=0 vsize=0 rrsrinivas messages=0 vsize=0 sralluri messages=0 vsize=0 jvgkraju messages=0 vsize=0 harikrishna messages=0 vsize=0 kvvnrao messages=0 vsize=0 chandrasekhar messages=0 vsize=0 admin messages=0 vsize=0 gsnraju messages=0 vsize=0 Thanks Kumar -- Sent from: http://dovecot.2317879.n4.nabble.com/

2 1

IMAP SEARCH over ALL mailboxes
by Filip Hanes 14 Jul '20

14 Jul '20

Hi, is it possible to SEARCH in dovecot IMAP over all messages in all mailboxes without selecting each mailbox?

2 2

Static users and doveadm
by Antonio Leding 14 Jul '20

14 Jul '20

Hello Dovecot users… I wanted to ask is using a static userdb will cause issues with some of the account and user related doveadm commands? If so, any suggestions on how to use doveadm in such a configuration? I have pasted both the cmd + error and userdb config below — Thanks in advance… sudo doveadm mailbox list -A Error: auth-master: userdb list: User listing returned failure doveadm: Error: Failed to iterate through some users userdb { args = uid=vmail gid=vmail home=/var/vmail allow_all_users=yes userdb_warning_disable=yes driver = static }

2 1

Apple Mail and subfolders
by Filip Hajný 14 Jul '20

14 Jul '20

Hi all, does anybody here use MacOS (Catalina) Mail, who can confirm that subfolders (i.e. folders inside folders) work for them? It seems that I cannot create/access Maildir subfolders at all. Subfolders created in Mail show up as '.folder\2esubfolder' on disk, suggesting that Apple Mail is URL-coding the dot separator before it passes it onto Dovecot. No such translation takes place when interpreting the IMAP tree though, resulting in subfolders rendered as plain entries with the dot separator preserved. Subfolders can be created from elsewhere (SoGO, Roundube, Thunderbird) just fine and work here just fine, suggesting that all is well in terms of Dovecot configuration. Subfolders created here though show up as a proper hierarchy in Apple Mail, but their contents is not displayed, so the user thinks the subfolder is empty. This definitely used to work fine in the past. Running 2.3.10.1 as part of the Mailcow suite. -F

5 7

Syncing emails from external server like Gmail
by Vitalii 14 Jul '20

14 Jul '20

Greeting I've managed to backup my emails from external email server via dsync and imapc: protocol like this: doveadm -Dv -o imapc_user='user(a)gmail.com' -o imapc_password="pass" -o imapc_host=imap.gmail.com backup -R -u duke(a)ejoom.com imapc: taken from https://wiki.dovecot.org/Migration/Dsync I wish to do regular syncing of 5min interval to substitute 'getmail' program. When I repeat the command It doesn't download all the attachments(!good) but It seems that at retrieve all the the GUIDs and UIDs(!bad) again that take quite long considering 30K emails. The question is how to speed it up and retrieve only new emails? Does anyone have a solution for regular synchronization? What is the best way? Logs look like this: https://gist.githubusercontent.com/vnagara/87e0da1b21b51f68fd74270b3ec653b3… It wrights that "Mailboxes are equal" but still verifies all the UIDs again. Thanks.

4 11

Fwd: Dsync mail migration from v2.2 --> v2.3
by Antonio Leding 14 Jul '20

14 Jul '20

Two items I forgot to mention in my initial post: (A) I am running the sync command on the new v2.3 server. (B) To get sync to work, I had to add the following to the old v2.2 server: auth_socket_path = /var/run/dovecot/auth-master I think this relates to what I said in my initial post about AWS linux installing all of the sockets into a non-standard location. I say this because when I was trying get this working but before I had added the above setting, I noticed the following message: Error: userdb lookup: connect(/usr/local/var/run/dovecot/auth-userdb) failed: No such file or directory This seemed to indicate that the old server was looking for the auth socket in the incorrect location. So, I decided to correct this with the setting I noted above. Please let me know if I had this correct or perhaps there is something else going on here… Thanks... > Begin forwarded message: > > From: Antonio Leding <tech(a)leding.net> > Subject: Dsync mail migration from v2.2 --> v2.3 > Date: July 13, 2020 at 8:45:18 PM PDT > To: dovecot(a)dovecot.org > > Hello all, > > I’ve been battling how to migrate my imap mail following a new server install that has Dovecot and Postfix upgrades. > > The research seems pretty clear that the preferred method is to use dsync to get the mail from Server-A to B. After several hours and several different combos, I think I might finally have this working. However, before committing to my procedure and running it against all mailboxes, I wanted to bounce this off of the community and get any feedback that might catch possible issues and\or offer some advice. > > Environment > > Old server: Dovecot 2.2.2 running on AWS linux 1 (I believe that is based on RHEL 4 or 5). > New server: Dovecot 2.3.10.1 running on CentOS 7.6. > All authentication is done by Dovecot against a mySQL (and Maria) database — nothing is in files. > All mail data and mailboxes are owned by a dedicated user “vmail” and I’m using a static userdb. > SSH is via public key file only — no PW based SSH is allowed. > SSH is on a non-standard port > > Operations > > I’ve decided to migrate one mailbox per command. I only have about 30 accounts on the server so this shoudn't be a huge deal. > Here is the command that I believe is good to go: > > sudo doveadm -Dv sync -u [ MBOX_TO_MIGRATE ] remote:vmail@[ SERVER ] > > The SERVER references a host stanza in root's SSH connection file: /root/.ssh/config > I need to use the config file because I (a) use a non-standard SSH port; (b) need to login as “vmail” ; (c) need to use the “vmail" identity file. > > Comments & notes > > The Dovecot 2.2 package for AWS Linux was apparently setup to install all of the sockets in /usr/local/var/run/dovecot. This must have caused authentication issues because to get it working properly, I then had to add "path = /var/run/dovecot/auth-master” statement to the “auth" stanza in dovecot.conf file. > Also, the fact that I do not use a “standard” SSH config was an issue early on but once I committed to allowing vmail to SSH into the old server, that part became moot. For security purposes, I will probably disable vmail SSH after the data is migrated. > > Questions > > First and foremost — any obvious issues here? > Next, any advice and\or lessons learned with this type of migration? > Finally, presuming this is the proper way to do this, will this migrate all mail content including Sent, Deleted, Drafts, etc.? > > > Thanks in advance for taking the time and providing any feedback — very much appreciated... > >

1 0

Dsync mail migration from v2.2 --> v2.3
by Antonio Leding 14 Jul '20

14 Jul '20

Hello all, I’ve been battling how to migrate my imap mail following a new server install that has Dovecot and Postfix upgrades. The research seems pretty clear that the preferred method is to use dsync to get the mail from Server-A to B. After several hours and several different combos, I think I might finally have this working. However, before committing to my procedure and running it against all mailboxes, I wanted to bounce this off of the community and get any feedback that might catch possible issues and\or offer some advice. Environment Old server: Dovecot 2.2.2 running on AWS linux 1 (I believe that is based on RHEL 4 or 5). New server: Dovecot 2.3.10.1 running on CentOS 7.6. All authentication is done by Dovecot against a mySQL (and Maria) database — nothing is in files. All mail data and mailboxes are owned by a dedicated user “vmail” and I’m using a static userdb. SSH is via public key file only — no PW based SSH is allowed. SSH is on a non-standard port Operations I’ve decided to migrate one mailbox per command. I only have about 30 accounts on the server so this shoudn't be a huge deal. Here is the command that I believe is good to go: sudo doveadm -Dv sync -u [ MBOX_TO_MIGRATE ] remote:vmail@[ SERVER ] The SERVER references a host stanza in root's SSH connection file: /root/.ssh/config I need to use the config file because I (a) use a non-standard SSH port; (b) need to login as “vmail” ; (c) need to use the “vmail" identity file. Comments & notes The Dovecot 2.2 package for AWS Linux was apparently setup to install all of the sockets in /usr/local/var/run/dovecot. This must have caused authentication issues because to get it working properly, I then had to add "path = /var/run/dovecot/auth-master” statement to the “auth" stanza in dovecot.conf file. Also, the fact that I do not use a “standard” SSH config was an issue early on but once I committed to allowing vmail to SSH into the old server, that part became moot. For security purposes, I will probably disable vmail SSH after the data is migrated. Questions First and foremost — any obvious issues here? Next, any advice and\or lessons learned with this type of migration? Finally, presuming this is the proper way to do this, will this migrate all mail content including Sent, Deleted, Drafts, etc.? Thanks in advance for taking the time and providing any feedback — very much appreciated...

1 0

Re: Dovecot permission denied errors on NFS after upgrade to 2.2.17
by Claudio Corvino 13 Jul '20

13 Jul '20

Hi Jochen, the ID has not changed, in addition the "permission denied" error is random, most of the time Dovecot works well. How can I check the mappings NFS uses? After the Debian upgrade I was using these NFS mount options: rw,user,rsize=8192,wsize=8192,timeo=30,intr Now trying to fix the problem I set "defaults" mount option, but nothing has changed. Thanks On 13/07/20 13:08, Jochen Bern wrote: > On 07/13/2020 12:57 PM, Claudio Corvino wrote: >> we just upgraded our mail server from Debian 8 to Debian 9, so our >> dovecot packages just got upgraded from version 2.2.13 to 2.2.17. > [...] >> Every directory and sub-dir under >> /mnt/mail-storage >> is assigned to vmail:vmail with >> drwx------ >> permission set. > Just a quick shot into the dark, you verified the ownerships on the NFS > *client*, not the NFS *server*, yes? Because NFS speaks UIDs and GIDs, > not their names, so if your Debian 9 installation assigned different IDs > to "vmail" ... > > (You might also want to check what mappings the NFS mount uses, > including possible *default* behaviour.) > > Good luck,

1 0

Dovecot permission denied errors on NFS after upgrade to 2.2.17
by Claudio Corvino 13 Jul '20

13 Jul '20

Hi everyone, we just upgraded our mail server from Debian 8 to Debian 9, so our dovecot packages just got upgraded from version 2.2.13 to 2.2.17. Now we have a lot of errors in our dovecot.log about permission denied (our Mailboxes are on NFS remote server since years without problems) like these: Jul 13 12:39:23 Error: imap(XXXXXXXXX): open(/mnt/mail-storage/XXXXXXXXX/Maildir/cur/1594378191.M653755P26498.ganimede,S=12716,W=12984:2,S) failed: Permission denied (euid=501(vmail) egid=501(vmail) missing +r perm: /mnt/mail-storage/XXXXXXXXX/Maildir/cur/1594378191.M653755P26498.ganimede,S=12716,W=12984:2,S stat(/mnt/mail-storage/XXXXXXXXX/Maildir/cur/1594378191.M653755P26498.ganimede,S=12716,W=12984:2,S) failed: Permission denied) For some user this is not a problem, for some other user this blocks their mail client (Thunderbird). We use vmail user for running Dovecot and I can't find a permission problem on the path indicated by the mail error, user vmail can read and write and uid and gid seems to be correct. Permissions on /mnt/mail-storage was drwx------ 94 vmail vmail 4096 ago 29 2019 mail-storage I just changed to drwxr-xr-x 94 root root 4096 ago 29 2019 mail-storage but nothing has changed. Every directory and sub-dir under /mnt/mail-storage is assigned to vmail:vmail with drwx------ permission set. If try to stat on /mnt/mail-storage/XXXXXXXXX/Maildir/cur/1594378191.M653755P26498.ganimede,S=12716,W=12984:2,S I get permission denied. We use LMTP and Postfix to deliver mail to mailbox. mail_debug=yes already enabled; mail_fsync is set to never (should I change to always?). Please could you give us help on the matter as I don't know where to look further. Thanks Claudio

1 0

Re: Please Help me
by Kishore Potnuru 13 Jul '20

13 Jul '20

Hi Vitalii, I am using "Maildir". Thanks Kishore Potnuru On Mon, Jul 13, 2020 at 10:32 AM Vitalii <vnagara(a)yandex.com> wrote: > On Sun, 12 Jul 2020 21:17:48 +0100 > Kishore Potnuru <kishore.reachme(a)gmail.com> wrote: > > > Hi All, > > > > I need to migrate the complete data & setup/environment from an old > > dovecot environment to a new dovecot environment. > > Hi Kishore, > I was wondering what mail format do you use Maildir/mdbox/sdbox? >

1 0

Please Help me
by Kishore Potnuru 13 Jul '20

13 Jul '20

Hi All, I need to migrate the complete data & setup/environment from an old dovecot environment to a new dovecot environment. Old environment details: ======================== Linux OS - Red Hat Enterprise Linux Server release 6.5 (Santiago) Dovecot version - 2.0.9 Postfix version - 2.6.6 New Environment details: ( 2 systems) ======================== Linux OS - Red Hat Enterprise Linux Server release 7.7 (Maipo) Dovecot version - 2.2.36 (1f10bfa63) Postfix version - 2.10.1 In the new environment, I have 2 servers. I have got 2 of the above servers with above configuration and need to create HA/Resilience. I know there is advanced versions in Redhat and Dovecot. As per my organisation's existing support/policy, that is what maximum supported RHEL and official version of dovecot as per RHEL 7.7. Now my questions are: 1. I will stop the dovecot/postfix services on the server in old environment. Is it possible to move/copy the complete configuration along with data (directories/sub-directories/users/groups/existing email in new or cur folders/complete email domains along with complete structure) to the other 2 servers in new env? Could you please suggest me the commands or steps follow that? 2. Now, With the above 2 servers, I would like to create the HA/Resilience. I know there is a Replication options which are there. Could you please provide me the steps/video, how it needs to be done? I have got the systems and I need to move/copy the environment to a new environment and create the HA/Resilience. I am not sure, how to proceed and what needs to be done clearly? I am stuck here. Please help me in this step by step procedure. Thanks, Kishore Potnuru

3 2