Re: [Dovecot] mail_max_userip_connections=10
Timo Sirainen tss at iki.fi writes:
On 19.9.2011, at 11.27, Tom Clark wrote:
Is there anyway of whitelisting an IP so that it can ignore
mail_max_userip_connections=10?
With v2.0 in theory:
remote 1.2.3.4 {
mail_max_userip_connections = 0
}
I don't know if it actually works.
Not sure I follow- in which conf file would it be appropriate to place this? Is
it possible to maintain separate values for the same config directive?
(I'm having a similar problem with iOS devices connecting to our server- they
seem to open multiple connections, easily going over 10 if they have many
subscribed directories).
Thanks,
-E-
[Dovecot] SiS
Hello, SiS is implemented and stable in the last version ? Best Regards Guy
Re: [Dovecot] Can't receive emails
On 5 October 2010 15:52, Chris spamo...@freenet.de wrote: Hello again, I searched every logfile I could find, but without any results :( How does it usually work? The mail is send by another server and on my server, does postfix receive the email or is it dovecots job? Who puts the mail into that mail directory? I guess it's postfix? Postfix will either deliver using virtual or some other delivery agent if you have set one up. Unless you are using Dovecot LDA (which I don't see in your config) then Dovecot has nothing to do with delivery. Mail logs will most likely be in /var/log/syslog or /var/log/mail.log or similar. The postfix mailing list is probably what you're looking for: http://www.postfix.org/lists.html And you should give this a quick read before posting: http://www.postfix.org/DEBUG_README.html#mail Guy -- Don't just do something...sit there!
[Dovecot] Dovecot director - silly questions
Hi, I'm currently using Dovecot 1.1.18 with storage over NFS and there's been the occasional corrupt index. I'm going to be installing a new set of servers soon and would like to start using Dovecot 2.0. I've had a look at the bit there is on the new Dovecot director, but I'm not totally sure how it's intended to be used. Should I install Dovecot on the software load balancers (IMAP/POP3 and other services balanced by haproxy) and configure the director on there? I'm also using maildrop at the moment but would like to switch to Dovecot deliver at some stage. Since mail is delivered to a random mail server (they both have access to the NFS shares), won't I run into the same index problem? Should I look at per user transports before then or can director also be used to effect Dovecot deliver? Thanks Guy -- Don't just do something...sit there!
[Dovecot] Dovecot 2.0 progress?
Hi, Been having a look at the roadmap on the wiki for Dovecot 2.0 and I was wondering how that's progressing? Are you able to offer a guesstimate on how far it is from stable Timo? Just asking because we're moving to new servers soon and I'm considering using the opportunity to switch from Dovecot/Maildrop/Maildir to Dovecot/deliver/mdbox if 2.0 is likely to be reasonably safe to use in the next month or so. Thanks Guy -- Don't just do something...sit there!
Re: [Dovecot] Dovecot not delivering mail.
2010/1/6 Eduardo M KALINOWSKI edua...@kalinowski.com.br flamebait If they could, they'd be running debian. :-) /flamebait flame insert flame from ubuntu admin ;) /flame I tried the postfix-dovecot package once and dumped it. Found it easier using the separate packages. I'd also recommend installing Dovecot from source under Ubuntu as they're lagging well behind. Karmic still only has Dovecot 1.1.11. Cheers Guy -- Don't just do something...sit there!
Re: [Dovecot] NFS random redirects
2009/10/21 Timo Sirainen t...@iki.fi On Oct 20, 2009, at 4:47 AM, Thomas Hummel wrote: If you do it, you'll most likely see some random index related errors. But are index related errors recoverable (does dovecot notice and fix it dynamically ?) or will they cause client-side corruption ? How bad would that corruption be ? (like fetching wrong message, since index store nextuid as well ? setting wrong flag ? ...) and how could a client fix it ? It's unlikely that anything bad happens, but who knows. Random unnoticed corruption can do pretty much anything. Our current setup uses two NFS mounts accessed simultaneously by two servers. Our load balancing tries to keep a user on the same server whenever possible. Initially we just had roundrobin load balancing which led to index corruption. The problems we've had with that corruption have simply been that some messages are displayed twice or not displayed at all in mail clients. Deletion of the corrupted index allowed Dovecot to recreate it correctly, so the client can't do anything about it. You'd probably have to do it manually or have some sort of web interface for users to do it themselves. I certainly wouldn't use NFS with multiple servers accessing it again for Dovecot. Looking at a clustered FS on SAN solution at the moment. Cheers Guy -- Don't just do something...sit there!
[Dovecot] Internal server
Hi, I'm looking at doing some testing for a little internal IMAP server. Basically use fetchmail (or some such thing to bring in mail) so users have a local IMAP server to connect to. I'm interested in trying out mdbox, but since I'd have to use dovecot 2.0 with it I thought I'd just check what peoples' experience with 2.0 is like so far? Is it reasonably stable already? Is there also a page showing the differences in config for dbox over maildir? Or is it simply just the change in mail_location and nothing else? Thanks Guy -- Don't just do something...sit there!
[Dovecot] Funky quota things
Hi guys, I already use Dovecot for imap/pop3 with some simple quota options. I'm starting to look at the switch from maildrop to dovecot lda and I'm wondering whether it's possible to do our unusual quota management with the quota plugin. We have the users total quota in MySQL along with a number of grace days (counts up to 7). The amount of space used is stored in maildirsize in the users maildir. What we currently some bits of voodoo which allow the user to go up to 200% of their quota. When they go over 100% they start receiving warnings and their grace days start counting down. If the account is still using over 100% of quota after the 7 days then the 100% quota is enforced so no new mail will be accepted. So is something like this burst quota possible in the Dovecot plugin? Thanks Guy -- Don't just do something...sit there!
[Dovecot] Dovecot 1.2.2
Hi, I've just tried upgrading from 1.1.8 to 1.2.2. As soon as the new version was started up the number of imap/pop3 processes began to climb to 800+, where I normally have about 300 or so connections at any given time. The load on the server also climbed to 150+. I've reverted back to 1.1.18, which is running just fine, but is there anything I can do to help trace what was causing the problem? Unfortunately I don't have a clue where to start. Thanks Guy -- Don't just do something...sit there!
[Dovecot] 1.1.8 upgrade to 1.2
Hi, I'm hoping to upgrade to 1.2 in the next couple of days. I remember seeing something mentioning some sort of config change, but unfortunately I don't remember where I saw it nor what the config option was. Are there any gotchas I need to watch out for in the change to 1.2 from 1.1.8? I've included dovecot -n below. Thanks Guy #dovecot -n # 1.1.8: /etc/dovecot/dovecot.conf # OS: Linux 2.6.24-23-server x86_64 Ubuntu 8.04.2 protocols: imap imaps pop3 pop3s listen(default): *:143 listen(imap): *:143 listen(pop3): *:110 ssl_listen(default): *:993 ssl_listen(imap): *:993 ssl_listen(pop3): *:995 ssl_cert_file: /etc/ssl/certs/imapd.pem ssl_key_file: /etc/ssl/private/imapd.pem disable_plaintext_auth: no verbose_ssl: yes login_dir: /var/run/dovecot/login login_executable(default): /usr/libexec/dovecot/imap-login login_executable(imap): /usr/libexec/dovecot/imap-login login_executable(pop3): /usr/libexec/dovecot/pop3-login login_greeting: Dovecot ready on mink.domain.org. login_process_per_connection: no login_processes_count: 5 login_max_processes_count: 1024 login_max_connections: 1024 max_mail_processes: 1024 mail_max_userip_connections: 20 verbose_proctitle: yes mail_location: maildir:%h/Maildir/ mail_debug: yes mail_full_filesystem_access: yes mmap_disable: yes dotlock_use_excl: no mail_nfs_storage: yes mail_nfs_index: yes lock_method: dotlock mail_executable(default): /usr/libexec/dovecot/imap mail_executable(imap): /usr/libexec/dovecot/imap mail_executable(pop3): /usr/libexec/dovecot/pop3 mail_plugins(default): imap_quota quota mail_plugins(imap): imap_quota quota mail_plugins(pop3): quota mail_log_max_lines_per_sec: 30 imap_client_workarounds(default): outlook-idle delay-newmail netscape-eoh imap_client_workarounds(imap): outlook-idle delay-newmail netscape-eoh imap_client_workarounds(pop3): pop3_uidl_format(default): %08Xu%08Xv pop3_uidl_format(imap): %08Xu%08Xv pop3_uidl_format(pop3): %08Xv%08Xu pop3_client_workarounds(default): pop3_client_workarounds(imap): pop3_client_workarounds(pop3): outlook-no-nuls oe-ns-eoh namespace: type: private separator: / inbox: yes list: yes subscriptions: yes namespace: type: private separator: / prefix: mail/ location: maildir:%h/Maildir/ hidden: yes subscriptions: yes auth default: cache_size: 2048 cache_ttl: 300 cache_negative_ttl: 1 username_chars: abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ01234567890.-_@' master_user_separator: * debug: yes passdb: driver: passwd-file args: /etc/dovecot/dovecot-master.pwd master: yes passdb: driver: sql args: /etc/dovecot/dovecot-mysql.conf userdb: driver: sql args: /etc/dovecot/dovecot-mysql.conf plugin: quota: maildir quota_rule: *:storage=100M quota_rule2: Trash:ignore -- Don't just do something...sit there!
[Dovecot] Dovecot + DRBD/GFS mailstore
Hi guys, I'm looking at the possibility of running a pair of servers with Dovecot LDA/imap/pop3 using internal drives with DRBD and GFS (or other clustered FS) for the mail storage and ext3 for the root drive. I'm currently using maildrop for delivery and Dovecot imap/pop3 with the stores over NFS. I'm looking for better performance but still keeping the HA element I have now with shared storage over NFS. Has anyone had experience with a setup like the one I'm suggesting? What was performance like with Dovecot using GFS? Thanks Guy -- Don't just do something...sit there!
[Dovecot] Seperate auth per protocol (pop3/imap)
Hi, During discussions with the boss we've speculated about having users having only specific services available. One of the ideas was to have users being able to specify if they wanted POP3 or IMAP and whether we could limit their access to one or the other server side. We use MySQL for the userdb and passdb lookups at the moment. Is it possible to specify separate auth files with different MySQL lookups per protocol? Thanks Guy -- Don't just do something...sit there!
[Dovecot] Intermittent certificate cannot be verified error
Hi guys, Not sure where to start looking for this. I've got a few users getting intermittent certificate cannot be verified messages when connecting through SSL to Dovecot. Connections go through haproxy to Dovecot 1.1.8 on the back end servers. I've got verbose_ssl and auth_debug enabled. All I'm seeing on the logs for the time the users reported the error is this: Jan 21 23:30:51 mink dovecot: auth(default): new auth connection: pid=28811 Jan 21 23:30:51 mink dovecot: IMAP(us...@domain1.net): Disconnected in IDLE bytes=73/4235 Jan 21 23:24:23 mink dovecot: auth(default): new auth connection: pid=28811 Jan 21 23:24:23 mink dovecot: imap-login: Disconnected (no auth attempts): rip=x.x.x.x, lip=x.x.x.x Jan 21 23:24:23 mink dovecot: auth(default): new auth connection: pid=28811 Jan 21 23:24:24 mink dovecot: IMAP(us...@domain1.net): Disconnected in IDLE bytes=89/920 Since it's so intermittent I'm not sure where to start. Since there are no real errors in the Dovecot logs I'm suspecting that haproxy is perhaps not routing every packet correctly leading to Dovecot not getting all the data needed for the connection. Are there any other possibilities I've missed? Thanks Guy r...@mink:/var/log/mail# dovecot -n # 1.1.8: /etc/dovecot/dovecot.conf # OS: Linux 2.6.24-23-server x86_64 Ubuntu 8.04.1 protocols: imap imaps pop3 pop3s listen(default): *:143 listen(imap): *:143 listen(pop3): *:110 ssl_listen(default): *:993 ssl_listen(imap): *:993 ssl_listen(pop3): *:995 ssl_cert_file: /etc/ssl/certs/imapd.pem ssl_key_file: /etc/ssl/private/imapd.pem disable_plaintext_auth: no verbose_ssl: yes login_dir: /var/run/dovecot/login login_executable(default): /usr/libexec/dovecot/imap-login login_executable(imap): /usr/libexec/dovecot/imap-login login_executable(pop3): /usr/libexec/dovecot/pop3-login login_process_per_connection: no login_processes_count: 5 login_max_processes_count: 256 max_mail_processes: 1024 verbose_proctitle: yes mail_location: maildir:%h/Maildir/ mail_full_filesystem_access: yes mmap_disable: yes dotlock_use_excl: no mail_nfs_storage: yes mail_nfs_index: yes lock_method: dotlock mail_executable(default): /usr/libexec/dovecot/rawlog /usr/libexec/dovecot/imap mail_executable(imap): /usr/libexec/dovecot/rawlog /usr/libexec/dovecot/imap mail_executable(pop3): /usr/libexec/dovecot/rawlog /usr/libexec/dovecot/pop3 mail_process_size: 128 mail_plugins(default): imap_quota quota mail_plugins(imap): imap_quota quota mail_plugins(pop3): quota mail_log_max_lines_per_sec: 30 imap_client_workarounds: outlook-idle delay-newmail pop3_uidl_format: %08Xv%08Xu pop3_client_workarounds: outlook-no-nuls oe-ns-eoh namespace: type: private separator: / inbox: yes list: yes subscriptions: yes namespace: type: private separator: / prefix: mail/ location: maildir:%h/Maildir/ hidden: yes subscriptions: yes auth default: cache_size: 2048 cache_ttl: 300 cache_negative_ttl: 1 username_chars: abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ01234567890.-_@' master_user_separator: * debug: yes worker_max_count: 5 passdb: driver: passwd-file args: /etc/dovecot/dovecot-master.pwd master: yes passdb: driver: sql args: /etc/dovecot/dovecot-mysql.conf userdb: driver: sql args: /etc/dovecot/dovecot-mysql.conf plugin: quota: maildir quota_rule: *:storage=100M quota_rule2: Trash:ignore -- Don't just do something...sit there!
[Dovecot] Error message in logs since 1.1.6 upgrade
Hi, I've got two imap boxes, one is running 1.1.5 and hasn't exhibited the error below, but the one that's been upgraded to 1.1.6 is showing them. Nov 14 16:06:31 mink dovecot: imap-login: Disconnected (no auth attempts): rip=x.x.x.x, lip=x.x.x.x, TLS handshaking: SSL_accept() failed: error:1408F119:SSL routines:SSL3_GET_RECORD:decryption failed or bad record mac If this isn't already a known error, is there any extra data I can collect that would help diagnose what's causing it? I haven't a clue where to start with this sort of debugging I'm afraid. Thanks Guy -- Don't just do something...sit there!
Re: [Dovecot] allow_nets overridden by cache
Hi Timo, 2008/10/29 Timo Sirainen [EMAIL PROTECTED]: What Dovecot version? Post your dovecot -n output? Seems to work fine with the almost-v1.1.6 (and I don't remember doing changes related to this for a long time): Sorry, should have included that before. I've just upgraded to 1.1.5, compiled from source with nothing extra added. Although I did just do a make install without doing an uninstall on the old 1.1.2, could that effect it? [EMAIL PROTECTED]:/opt/src/dovecot/1.1.5/dovecot-1.1.5# dovecot -n # 1.1.5: /etc/dovecot/dovecot.conf protocols: imap imaps pop3 pop3s listen(default): *:143 listen(imap): *:143 listen(pop3): *:110 ssl_listen(default): *:993 ssl_listen(imap): *:993 ssl_listen(pop3): *:995 ssl_cert_file: /etc/ssl/certs/imapd.pem ssl_key_file: /etc/ssl/private/imapd.pem disable_plaintext_auth: no login_dir: /var/run/dovecot/login login_executable(default): /usr/libexec/dovecot/imap-login login_executable(imap): /usr/libexec/dovecot/imap-login login_executable(pop3): /usr/libexec/dovecot/pop3-login login_process_per_connection: no login_processes_count: 10 max_mail_processes: 1024 verbose_proctitle: yes mail_location: maildir:%h/Maildir/ mail_full_filesystem_access: yes mmap_disable: yes dotlock_use_excl: no mail_nfs_storage: yes mail_nfs_index: yes lock_method: dotlock mail_executable(default): /usr/libexec/dovecot/imap mail_executable(imap): /usr/libexec/dovecot/imap mail_executable(pop3): /usr/libexec/dovecot/pop3 mail_process_size: 128 mail_plugins(default): imap_quota quota mail_plugins(imap): imap_quota quota mail_plugins(pop3): quota mail_log_max_lines_per_sec: 30 imap_client_workarounds: outlook-idle delay-newmail pop3_uidl_format: %08Xv%08Xu pop3_client_workarounds: outlook-no-nuls oe-ns-eoh namespace: type: private separator: / inbox: yes list: yes subscriptions: yes namespace: type: private separator: / prefix: mail/ location: maildir:%h/Maildir/ hidden: yes subscriptions: yes auth default: cache_size: 2048 cache_ttl: 1 cache_negative_ttl: 1 username_chars: abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ01234567890.-_@' master_user_separator: * verbose: yes worker_max_count: 5 passdb: driver: passwd-file args: /etc/dovecot/dovecot-master.pwd master: yes passdb: driver: sql args: /etc/dovecot/dovecot-mysql.conf userdb: driver: sql args: /etc/dovecot/dovecot-mysql.conf plugin: quota: maildir quota_rule: *:storage=100M quota_rule2: Trash:ignore In the section of logs that I showed you could see the client out:FAIL after the allow_nets check but a cache hit immediately afterwards would allow the account to login. Even if I set the auth_cache_ttl to 15 seconds it would still fail the allow_nets check and then get a cache hit and be logged in. If there's any other information I can provide please let me know. Thanks Guy -- Don't just do something...sit there!
[Dovecot] allow_nets overridden by cache
Hi, I've just started trying allow_nets on one of my servers. I have auth_debug and auth_verbose both enabled and the output is as follows: Oct 28 13:05:48 mink dovecot: auth-worker(default): auth([EMAIL PROTECTED],x.x.x.x): allow_nets: Matching for network 127.0.0.1/8 Oct 28 13:05:48 mink dovecot: auth-worker(default): auth([EMAIL PROTECTED],x.x.x.x): allow_nets: Matching for network 10.0.7.176/28 Oct 28 13:05:48 mink dovecot: auth-worker(default): passdb([EMAIL PROTECTED],x.x.x.x): allow_nets check failed: IP not in allowed networks Oct 28 13:05:50 mink dovecot: auth(default): client out: FAIL 1265 [EMAIL PROTECTED] Oct 28 13:05:50 mink dovecot: auth(default): cache([EMAIL PROTECTED],x.x.x.x): hit: hidden [EMAIL PROTECTED] Oct 28 13:05:50 mink dovecot: auth(default): client out: OK 1266 [EMAIL PROTECTED] auth_cache_ttl is set to 300. If I set it to 1 then the allow_nets successfully rejects. Once I set it back up to 300 the cache overrides the result from the allow_nets check and let's the account log in even though the allow_nets check fails. I've tried waiting for longer than the 300 seconds and then logged in again, but I still get the same result as above. Is there a gotcha that I'm unaware of or have I done something stupid (which seems the most likely :P)? Thanks Guy -- Don't just do something...sit there!
Re: [Dovecot] Install from source of dovecot 1.1.2
Thanks for everyone's input. I was already sudoed to root before running any of the operations so it wasn't that. May have been that --with-storages error I'd made. I tried your config John, with mods for my system and besides having to add the dovecot user/group manually it seems to be running just fine now. Thanks! 2008/8/10 John and Catherine Allen [EMAIL PROTECTED]: I also install from source on my Debian system and I keep the configure options in a little script which currently looks like this: #!/bin/bash #configure flags for compilation compatible with gentoo emerge #john allen #9.5.2005 #modified 19.11.2006 #modified 26.1.2008 for kirtley ./configure \ --with-pam \ --without-deliver \ --with-storages=maildir \ --prefix=/usr \ --mandir=/usr/share/man \ --infodir=/usr/share/info \ --datadir=/usr/share \ --sysconfdir=/etc/dovecot \ --localstatedir=/var I can't see any problem with your configure options, so as Charles says maybe there's another dovecot somewhere on your system. Have you tried the command which dovecot? John -- John Allen Bofferdange, Luxembourg [EMAIL PROTECTED] http://allenlux.dyndns.org -- Don't just do something...sit there!
[Dovecot] Install from source of dovecot 1.1.2
Hi guys, I have what I'm sure is a stupid question. I'm trying to install Dovecot 1.1.2 from source and my command sequence is this: cd dovecot-1.1.2 ./configure --prefix=/usr --exec-prefix=/usr --with-mysql --with-storagesmaildir --sysconfdir=/etc/dovecot make make install cd ../dovecot-sieve-1.1.5 ./configure --with-dovecot=../dovecot-1.1.2 make make install I don't get any errors during either install I've done the odd config on dovecot to try and keep it as close to the install dirs used by the Dovecot package included with Ubuntu. Once it's done, dovecot -n still looks in /usr/local/etc and when I try and start dovecot it looks for the libexec dir in /usr/local instead of /usr. Is there something I'm missing on the configure? I've very, very little experience (rather obviously) installing directly from source so any advice would be greatly appreciated. Thanks Guy -- Don't just do something...sit there!
[Dovecot] Corrupted transaction logs
Hi, I'm getting a lot of the following messages in my mail.err log. Jul 12 11:25:51 mink dovecot: IMAP([EMAIL PROTECTED]): Corrupted transaction log file /virtual/store1/mail/u/user1_domain1.net/Maildir/.Sent/dovecot.index.log: Append with UID 925, but next_uid = 926 Jul 12 12:44:56 mink dovecot: IMAP([EMAIL PROTECTED]): Corrupted transaction log file /virtual/store2/mail/u/user2_domain1.net/Maildir/.Trash/dovecot.index.log: Append with UID 4927, but next_uid = 4931 Jul 12 12:45:19 mink dovecot: IMAP([EMAIL PROTECTED]): Corrupted transaction log file /virtual/store2/mail/u/user2_domain1.net/Maildir/.Trash/dovecot.index.log: Append with UID 4930, but next_uid = 4931 The users get the wrong headers and bodies being displayed in Thunderbird and I'm unsure of the symptoms in other browsers. I'm running two mail servers with two NFS servers for the mail stores. Both mail servers have access to both mail stores and I'm assuming that that is what's causing the problem. I'm running dovecot 1.0.10 (Ubuntu - so I'm using only packages provides by aptitude at the moment). Should dovecot be able to handle access by multiple machines for the indexes? Thanks Guy -- Don't just do something...sit there!
Re: [Dovecot] Dovecot timeout over NFS
On 24/04/2008, Timo Sirainen [EMAIL PROTECTED] wrote: Have you set maildir_copy_with_hardlinks=yes? Yes, that's already set. I've also added: default_mail_env = maildir:%h/Maildir/:INDEX=MEMORY after reading the suggestion somewhere. The server is only running one domain so far so it's not critical to clients just yet. But I need to be sure it can handle fairly heavy usage before I transfer other clients on to it. Thanks for having a look. -- Don't just do something...sit there!
[Dovecot] Dovecot timeout over NFS
Hi, I'm using postfix with Dovecot 1.0.5 and getting the following errors when a user moves a massive folders contents to trash (65000 or so messages): Apr 21 15:31:01 mink dovecot: IMAP([EMAIL PROTECTED]): Timeout while waiting for lock for transaction log file /virtual/store1/mail/d/dev2_aluminati.net/Maildir/.Deleted Items/dovecot.index.log Is there anything I can do to compensate for situations like this? Thanks Guy -- Don't just do something...sit there!
[Dovecot] Single quote email addresses
Hi, I've just been told that I need to make our mail server cater to accounts with single quotes. Has anyone set dovecot up to do this or know of any documentation/howto on it, assuming it's possible at all. I'm running dovecot 1.0.5 on Ubuntu 7.10 if that makes much of a difference. Thanks for any help anyone can give. Guy -- Don't just do something...sit there!
