Re: [Dovecot] Question regarding Postfix and Dovecot
Replying back to the list.
On Thu, Mar 14, 2013 at 10:51 AM, mourik jan c heupink < heupink@merit.unu.edu> wrote:
please reply to the list
On 3/14/2013 11:38 AM, Daniel Reinhardt wrote:
Yes I have read everything on that, and yet postfix does not even see the dovecot virtual transport.
On Thu, Mar 14, 2013 at 10:37 AM, mourik jan c heupink <heupink@merit.unu.edu <mailto:heupink@merit.unu.edu>**> wrote:
Hi Daniel, I'm new to dovecot myself, but did you read this: http://wiki2.dovecot.org/LDA/_**_Postfix<http://wiki2.dovecot.org/LDA/__Postfix> <http://wiki2.dovecot.org/LDA/**Postfix<http://wiki2.dovecot.org/LDA/Postfix>I'm guessing that perhaps you need to configure your virtual transport? So, in master.cf <http://master.cf>, include a line like: dovecot unix - n n - - pipe flags=DRhu user=vmail:vmail argv=/usr/local/libexec/__**dovecot/dovecot-lda -f ${sender} -d ${recipient} and configure virtual_transport = dovecot in main.cf <http://main.cf> But again... I'm very new to all this myself, but perhaps it helps? Regards, MJ-- Daniel Reinhardt cryptodan@cryptodan.net <mailto:cryptodan@cryptodan.**net<cryptodan@cryptodan.net>
http://www.cryptodan.net 301-875-7018(c) 410-455-0488(h)
-- Daniel Reinhardt cryptodan@cryptodan.net http://www.cryptodan.net 301-875-7018(c) 410-455-0488(h)
Looking at your config, I notice virtual_transport = virtual
However according to http://wiki2.dovecot.org/LDA/Postfix: main.cf: dovecot_destination_recipient_limit = 1 virtual_mailbox_domains = your.domain.here virtual_transport = dovecot
master.cf: dovecot unix - n n - - pipe flags=DRhu user=vmail:vmail argv=/usr/local/libexec/dovecot/dovecot-lda -f ${sender} -d ${recipient}
But again, I'm new to all this. Here postfix/dovecot/virtual works fine, and all I did was follow the docs.
On 3/14/2013 11:53 AM, Daniel Reinhardt wrote:
Replying back to the list.
On Thu, Mar 14, 2013 at 10:51 AM, mourik jan c heupink < heupink@merit.unu.edu> wrote:
please reply to the list
On 3/14/2013 11:38 AM, Daniel Reinhardt wrote:
Yes I have read everything on that, and yet postfix does not even see the dovecot virtual transport.
On Thu, Mar 14, 2013 at 10:37 AM, mourik jan c heupink <heupink@merit.unu.edu <mailto:heupink@merit.unu.edu>**> wrote:
Hi Daniel, I'm new to dovecot myself, but did you read this: http://wiki2.dovecot.org/LDA/_**_Postfix<http://wiki2.dovecot.org/LDA/__Postfix> <http://wiki2.dovecot.org/LDA/**Postfix<http://wiki2.dovecot.org/LDA/Postfix>I'm guessing that perhaps you need to configure your virtual transport? So, in master.cf <http://master.cf>, include a line like: dovecot unix - n n - - pipe flags=DRhu user=vmail:vmail argv=/usr/local/libexec/__**dovecot/dovecot-lda -f ${sender} -d ${recipient} and configure virtual_transport = dovecot in main.cf <http://main.cf> But again... I'm very new to all this myself, but perhaps it helps? Regards, MJ-- Daniel Reinhardt cryptodan@cryptodan.net <mailto:cryptodan@cryptodan.**net<cryptodan@cryptodan.net>
http://www.cryptodan.net 301-875-7018(c) 410-455-0488(h)
I looked at my config in main.cf and I do not see it. Can you point it out?
On Thu, Mar 14, 2013 at 11:02 AM, mourik jan c heupink < heupink@merit.unu.edu> wrote:
Looking at your config, I notice virtual_transport = virtual
However according to http://wiki2.dovecot.org/LDA/**Postfix<http://wiki2.dovecot.org/LDA/Postfix> : main.cf: dovecot_destination_recipient_**limit = 1 virtual_mailbox_domains = your.domain.here virtual_transport = dovecot
master.cf:
dovecot unix - n n - - pipe flags=DRhu user=vmail:vmail argv=/usr/local/libexec/**dovecot/dovecot-lda -f ${sender} -d ${recipient}
But again, I'm new to all this. Here postfix/dovecot/virtual works fine, and all I did was follow the docs.
On 3/14/2013 11:53 AM, Daniel Reinhardt wrote:
Replying back to the list.
On Thu, Mar 14, 2013 at 10:51 AM, mourik jan c heupink < heupink@merit.unu.edu> wrote:
please reply to the list
On 3/14/2013 11:38 AM, Daniel Reinhardt wrote:
Yes I have read everything on that, and yet postfix does not even see
the dovecot virtual transport.
On Thu, Mar 14, 2013 at 10:37 AM, mourik jan c heupink <heupink@merit.unu.edu <mailto:heupink@merit.unu.edu>****> wrote:
Hi Daniel, I'm new to dovecot myself, but did you read this: http://wiki2.dovecot.org/LDA/_****_Postfix<http://wiki2.dovecot.org/LDA/_**_Postfix><http://wiki2.**dovecot.org/LDA/__Postfix<http://wiki2.dovecot.org/LDA/__Postfix>
<http://wiki2.dovecot.org/LDA/****Postfix<http://wiki2.dovecot.org/LDA/**Postfix><http://wiki2.**dovecot.org/LDA/Postfix<http://wiki2.dovecot.org/LDA/Postfix>
I'm guessing that perhaps you need to configure your virtual transport? So, in master.cf <http://master.cf>, include a linelike:
dovecot unix - n n - - pipe flags=DRhu user=vmail:vmail argv=/usr/local/libexec/__****dovecot/dovecot-lda -f ${sender} -d ${recipient} and configure virtual_transport = dovecot in main.cf <But again... I'm very new to all this myself, but perhaps it helps? Regards, MJ-- Daniel Reinhardt cryptodan@cryptodan.net <mailto:cryptodan@cryptodan.****net< cryptodan@cryptodan.net>
301-875-7018(c) 410-455-0488(h)
-- Daniel Reinhardt cryptodan@cryptodan.net http://www.cryptodan.net 301-875-7018(c) 410-455-0488(h)
On 2013-03-14 7:11 AM, Daniel Reinhardt <cryptodan@gmail.com> wrote:
I looked at my config in main.cf and I do not see it. Can you point it out?
First, virtual_transport = virtual is the default setting. Since you posted -d output, that is what was shown.
With postfix (and I believe dovecot), the last config entry wins. So, if you have a certain setting specified twice in the config file, the last one (closest to the bottom) is the one that is used.
Postconf -n output will show you whether or not you have changed it.
If you are certain you have changed it, and postconf -n output doesn't show virtual_transport = dovecot, then you either have your change somewhere above the default in main.cf (meaning your setting is not the last one, and so is over-ridden), or, you are not using the main.cf file you think you are.
Show contents of master.cf (minus commented lines)...
Is your postfix chroot'd? If so (no 'n' in the chroot column), then you are probably editing the wrong main.cf.
A long time ago I decided it was easier to just create a section at the very end of main.cf, and put all of my changes there.
--
Best regards,
Charles
Thank you for point that out, so I went ahead and removed postfix, mysql, and dovecot in hopes to start fresh and follow the guide I used previously. Yet Postfix still isnt sending mail to the virtual mailbox setting defined in my dovecot config:
Here is the output of the postconf -n. I aplogize for the postconf -d, as I was not aware of the difference.
//// Postconf -n append_dot_mydomain = no biff = no broken_sasl_auth_clients = yes config_directory = /etc/postfix content_filter = amavis:[127.0.0.1]:10024 disable_vrfy_command = yes dovecot_destination_recipient_limit = 1 inet_interfaces = all mailbox_size_limit = 0 maximal_backoff_time = 8000s maximal_queue_lifetime = 7d minimal_backoff_time = 1000s mydestination = localhost, cryptodan.net, mail.cryptodan.net, mail.pandorah.net, pandorah.net, andromeda.milkyway myhostname = andromeda.milkyway mynetworks = 127.0.0.0/8 [::ffff:127.0.0.0]/104 [::1]/128 mynetworks_style = host myorigin = /etc/hostname readme_directory = no recipient_delimiter = + smtp_helo_timeout = 60s smtp_tls_note_starttls_offer = yes smtp_tls_security_level = may smtpd_banner = $myhostname ESMTP $mail_name smtpd_client_restrictions = reject_rbl_client sbl.spamhaus.org, reject_rbl_client blackholes.easynet.nl, reject_rbl_client dnsbl.njabl.org smtpd_data_restrictions = reject_unauth_pipelining smtpd_delay_reject = yes smtpd_hard_error_limit = 12 smtpd_helo_required = yes smtpd_helo_restrictions = permit_mynetworks, warn_if_reject reject_non_fqdn_hostname, reject_invalid_hostname, permit smtpd_recipient_limit = 16 smtpd_recipient_restrictions = reject_unauth_pipelining, permit_mynetworks, permit_sasl_authenticated, reject_non_fqdn_recipient, reject_unknown_recipient_domain, reject_unauth_destination, check_policy_service inet:127.0.0.1:10023, permit smtpd_sasl_auth_enable = yes smtpd_sasl_authenticated_header = yes smtpd_sasl_local_domain = smtpd_sasl_path = private/dovecot-auth smtpd_sasl_security_options = noanonymous smtpd_sasl_type = dovecot smtpd_sender_restrictions = permit_sasl_authenticated, permit_mynetworks, warn_if_reject reject_non_fqdn_sender, reject_unknown_sender_domain, reject_unauth_pipelining, permit smtpd_soft_error_limit = 3 smtpd_tls_cert_file = /etc/ssl/certs/ssl-cert-snakeoil.pem smtpd_tls_key_file = /etc/ssl/private/ssl-cert-snakeoil.key smtpd_tls_loglevel = 1 smtpd_tls_received_header = yes smtpd_tls_security_level = may smtpd_tls_session_cache_timeout = 3600s smtpd_use_tls = yes tls_random_source = dev:/dev/urandom unknown_local_recipient_reject_code = 450 virtual_alias_maps = mysql:/etc/postfix/mysql_virtual_alias_maps.cf, mysql:/etc/postfix/mysql_virtual_alias_domainaliases_maps.cf virtual_gid_maps = static:8 virtual_mailbox_base = /var/vmail virtual_mailbox_domains = mysql:/etc/postfix/mysql_virtual_domains_maps.cf virtual_mailbox_maps = mysql:/etc/postfix/mysql_virtual_mailbox_maps.cf, mysql:/etc/postfix/mysql_virtual_mailbox_domainaliases_maps.cf virtual_transport = dovecot virtual_uid_maps = static:150 ////
Here is the virtual_transport set in master.cf to dovecot:
dovecot unix - n n - - pipe flags=DRhu user=vmail:mail argv=/usr/lib/dovecot/dovecot-lda -d $(recipient)
Here is my dovecot -n output:
/////root@andromeda:/etc/postfix# dovecot -n # 2.0.19: /etc/dovecot/dovecot.conf # OS: Linux 3.2.0-38-generic-pae i686 Ubuntu 12.04.2 LTS ext4 disable_plaintext_auth = no first_valid_uid = 150 last_valid_uid = 150 mail_gid = mail mail_location = maildir:/var/vmail/%d/%n mail_uid = vmail managesieve_notify_capability = mailto managesieve_sieve_capability = fileinto reject envelope encoded-character vacation subaddress comparator-i;ascii-numeric relational regex imap4flags copy include variables body enotify environment mailbox date ihave passdb { args = /etc/dovecot/dovecot-sql.conf.ext driver = sql } plugin { sieve = ~/.dovecot.sieve sieve_dir = ~/sieve } protocols = imap pop3 sieve service auth { unix_listener /var/spool/postfix/private/dovecot-auth { group = postfix mode = 0660 user = postfix } unix_listener auth-userdb { group = mail mode = 01224 user = vmail } } ssl_cert = </etc/ssl/certs/dovecot.pem ssl_cipher_list = ALL:!LOW:!SSLv2:ALL:!aNULL:!ADH:!eNULL:!EXP:RC4+RSA:+HIGH:+MEDIUM ssl_key = </etc/ssl/private/dovecot.pem userdb { args = /etc/dovecot/dovecot-sql.conf.ext driver = sql } protocol imap { imap_client_workarounds = tb-extra-mailbox-sep mail_max_userip_connections = 10 } protocol pop3 { mail_max_userip_connections = 10 pop3_client_workarounds = outlook-no-nuls oe-ns-eoh } protocol lda { deliver_log_format = msgid=%m: %$ mail_plugins = sieve postmaster_address = postmaster quota_full_tempfail = yes rejection_reason = Your message to <%t> was automatically rejected:%n%r }
////
Thanks, Daniel Reinhardt
On Thu, Mar 14, 2013 at 11:34 AM, Charles Marcus <CMarcus@media-brokers.com>wrote:
On 2013-03-14 7:11 AM, Daniel Reinhardt <cryptodan@gmail.com> wrote:
I looked at my config in main.cf and I do not see it. Can you point it out?
First, virtual_transport = virtual is the default setting. Since you posted -d output, that is what was shown.
With postfix (and I believe dovecot), the last config entry wins. So, if you have a certain setting specified twice in the config file, the last one (closest to the bottom) is the one that is used.
Postconf -n output will show you whether or not you have changed it.
If you are certain you have changed it, and postconf -n output doesn't show virtual_transport = dovecot, then you either have your change somewhere above the default in main.cf (meaning your setting is not the last one, and so is over-ridden), or, you are not using the main.cf file you think you are.
Show contents of master.cf (minus commented lines)...
Is your postfix chroot'd? If so (no 'n' in the chroot column), then you are probably editing the wrong main.cf.
A long time ago I decided it was easier to just create a section at the very end of main.cf, and put all of my changes there.
--
Best regards,
Charles
-- Daniel Reinhardt cryptodan@cryptodan.net http://www.cryptodan.net 301-875-7018(c) 410-455-0488(h)
On 2013-03-14 6:41 PM, Daniel Reinhardt <cryptodan@gmail.com> wrote:
Here is the output of the postconf -n. I aplogize for the postconf -d, as I was not aware of the difference.
No problem... but you forgot the other most important thing...
NON-VERBOSE postfix logs exhibiting the problem...
--
Best regards,
Charles
Here are the non-verbose mode of the logs exhibiting the problem:
/// Mar 15 06:56:37 andromeda dovecot: lda(cryptodan): Fatal: setgid(8(mail) from mail_gid setting) failed with euid=1000(cryptodan), gid=1000(cryptodan), egid=1000(cryptodan): Operation not permitted (This binary should probably be called with process group set to 8(mail) instead of 1000(cryptodan))
Mar 15 06:56:38 andromeda dovecot: lda(cryptodan): Fatal: setgid(8(mail) from mail_gid setting) failed with euid=1000(cryptodan), gid=1000(cryptodan), egid=1000(cryptodan): Operation not permitted (This binary should probably be called with process group set to 8(mail) instead of 1000(cryptodan))
Mar 15 06:56:38 andromeda postfix/local[5433]: E6DD110007E: to=< cryptodan@cryptodan.net>, relay=local, delay=0.11, delays=0.07/0/0/0.04, dsn=4.3.0, status=deferred (temporary failure)
Mar 15 06:56:38 andromeda dovecot: lda(cryptodan): Fatal: setgid(8(mail) from mail_gid setting) failed with euid=1000(cryptodan), gid=1000(cryptodan), egid=1000(cryptodan): Operation not permitted (This binary should probably be called with process group set to 8(mail) instead of 1000(cryptodan))
Mar 15 06:56:38 andromeda postfix/local[5439]: 20BB2100076: to=< cryptodan@cryptodan.net>, relay=local, delay=879, delays=879/0.01/0/0.03, dsn=4.3.0, status=deferred (temporary failure)
Mar 15 06:59:56 andromeda postfix/anvil[5420]: statistics: max connection rate 1/60s for (smtp:209.85.212.41) at Mar 15 10:56:35
Mar 15 06:59:56 andromeda postfix/anvil[5420]: statistics: max connection count 1 for (smtp:209.85.212.41) at Mar 15 10:56:35
Mar 15 06:59:56 andromeda postfix/anvil[5420]: statistics: max cache size 1 at Mar 15 10:56:35 ///
I did not remove anything, just separated the log entries to make them more readable.
dovecot unix - n n - - pipe flags=DRhu user=vmail:mail argv=/usr/lib/dovecot/dovecot-lda -d $(recipient)
On Fri, Mar 15, 2013 at 10:58 AM, Charles Marcus <CMarcus@media-brokers.com>wrote:
On 2013-03-14 6:41 PM, Daniel Reinhardt <cryptodan@gmail.com> wrote:
Here is the output of the postconf -n. I aplogize for the postconf -d, as I was not aware of the difference.
No problem... but you forgot the other most important thing...
NON-VERBOSE postfix logs exhibiting the problem...
--
Best regards,
Charles
-- Daniel Reinhardt cryptodan@cryptodan.net http://www.cryptodan.net 301-875-7018(c) 410-455-0488(h)
On 2013-03-15 7:11 AM, Daniel Reinhardt <cryptodan@gmail.com> wrote:
Here are the non-verbose mode of the logs exhibiting the problem:
/// Mar 15 06:56:37 andromeda dovecot: lda(cryptodan): Fatal: setgid(8(mail) from mail_gid setting) failed with euid=1000(cryptodan), gid=1000(cryptodan), egid=1000(cryptodan): Operation not permitted (This binary should probably be called with process group set to 8(mail) instead of 1000(cryptodan))
Ok, thanks - that should be enough for someone who knows more than me to figure out what you have done wrong...
--
Best regards,
Charles
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On Fri, 15 Mar 2013, Charles Marcus wrote:
On 2013-03-15 7:11 AM, Daniel Reinhardt <cryptodan@gmail.com> wrote:
Here are the non-verbose mode of the logs exhibiting the problem:
/// Mar 15 06:56:37 andromeda dovecot: lda(cryptodan): Fatal: setgid(8(mail) from mail_gid setting) failed with euid=1000(cryptodan), gid=1000(cryptodan), egid=1000(cryptodan): Operation not permitted (This binary should probably be called with process group set to 8(mail) instead of 1000(cryptodan))
Ok, thanks - that should be enough for someone who knows more than me to figure out what you have done wrong...
Check out mail_access_groups setting or make LDA setuid.
Steffen Kaiser -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.11 (GNU/Linux)
iQEVAwUBUUMYNl3r2wJMiz2NAQIyJgf/VhNc/VjhXAMjiDbv9U27IWw9iwabh4/t c+SDKxFocvD08pVYc0tejH6t9Q4RwAWXVukDa7a+pKBc6oOeDzX7MUA6ylsei4vN Sqlo3Ne7fdFtxZ6pKkoXUxmFmIDJ9aeF75WcgDpxZYSb0GNOYEdcNJJrzt1dNgm/ BxR2iualCro02kGGVSO/usTwxf3JRHVFzuV6kSCspJPXbF0V+D80QCGtl68UTAYm 0ypAB9K7PDk/29QjVQolME0NkLYR2YXNeFuQw4Ti9rriZwThY21FR8Nn851ott+M tmA7tgOAYTDBUd4LvzMgd+Kto+tRqsW5ZyhPdRJrh+gYKpqpYQ/0HQ== =eTP/ -----END PGP SIGNATURE-----
I got the permission issues fixed, so now I am unable to retrieve email via imap or pop3 with either thunderbird, iphone, or webmail application. Is there something I am not doing that is preventing this from working?
POP3 and IMAP logins work just fine.
On Fri, Mar 15, 2013 at 12:46 PM, Steffen Kaiser < skdovecot@smail.inf.fh-brs.de> wrote:
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On Fri, 15 Mar 2013, Charles Marcus wrote:
On 2013-03-15 7:11 AM, Daniel Reinhardt <cryptodan@gmail.com> wrote:
Here are the non-verbose mode of the logs exhibiting the problem:
/// Mar 15 06:56:37 andromeda dovecot: lda(cryptodan): Fatal: setgid(8(mail) from mail_gid setting) failed with euid=1000(cryptodan), gid=1000(cryptodan), egid=1000(cryptodan): Operation not permitted (This binary should probably be called with process group set to 8(mail) instead of 1000(cryptodan))
Ok, thanks - that should be enough for someone who knows more than me to figure out what you have done wrong...
Check out mail_access_groups setting or make LDA setuid.
- -- Steffen Kaiser -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.11 (GNU/Linux)
iQEVAwUBUUMYNl3r2wJMiz2NAQIyJg**f/VhNc/**VjhXAMjiDbv9U27IWw9iwabh4/t c+**SDKxFocvD08pVYc0tejH6t9Q4RwAWX**VukDa7a+**pKBc6oOeDzX7MUA6ylsei4vN Sqlo3Ne7fdFtxZ6pKkoXUxmFmIDJ9a**eF75WcgDpxZYSb0GNOYEdcNJJrzt1d**Ngm/ BxR2iualCro02kGGVSO/**usTwxf3JRHVFzuV6kSCspJPXbF0V+**D80QCGtl68UTAYm 0ypAB9K7PDk/**29QjVQolME0NkLYR2YXNeFuQw4Ti9r**riZwThY21FR8Nn851ott+M tmA7tgOAYTDBUd4LvzMgd+Kto+**tRqsW5ZyhPdRJrh+gYKpqpYQ/0HQ== =eTP/ -----END PGP SIGNATURE-----
-- Daniel Reinhardt cryptodan@cryptodan.net http://www.cryptodan.net 301-875-7018(c) 410-455-0488(h)
On Sat, 2013-03-16 at 00:47 +0000, Daniel Reinhardt wrote:
I got the permission issues fixed, so now I am unable to retrieve email via imap or pop3 with either thunderbird, iphone, or webmail application. Is there something I am not doing that is preventing this from working?
POP3 and IMAP logins work just fine.
Check permissions, and "logs", if dovecot can't give you something its pretty good (most of the time) by logging it, and it can't be working fine if those applications can not "read" mail, since they are front ends talking to pop3 and imap.
Just a guess:
Operation not permitted (This
binary should probably be called with process group set to 8(mail) instead of 1000(cryptodan))
Is your dovecot-lda running as root? Because: this could mean it tries to become 8(mail) to be able to deliver, and it's not allowed to?
MJ
On 2013-03-14 6:41 PM, Daniel Reinhardt <cryptodan@gmail.com> wrote:
Here is the virtual_transport set in master.cf to dovecot:
dovecot unix - n n - - pipe flags=DRhu user=vmail:mail argv=/usr/lib/dovecot/dovecot-lda -d $(recipient)
I don't totally understand this stuff, but mine is different, and contains:
dovecot unix - n n - - pipe flags=DROhu user=postfix:postfix argv=/usr/libexec/dovecot/dovecot-lda -f ${sender} -a ${recipient} -d ${user}@${nexthop}
and none of the examples on t he dovecot wiki look like yours:
http://wiki2.dovecot.org/LDA/Postfix
--
Best regards,
Charles
dovecot unix - n n - - pipe flags=DRhu user=vmail:mail argv=/usr/lib/dovecot/dovecot-lda -d $(recipient)
I don't totally understand this stuff, but mine is different, and contains:
dovecot unix - n n - - pipe flags=DROhu user=postfix:postfix argv=/usr/libexec/dovecot/dovecot-lda -f ${sender} -a ${recipient} -d ${user}@${nexthop}
And for what it's worth... here is mine:
dovecot unix - n n - - pipe flags=DRhu user=vmail:vmail argv=/usr/lib/dovecot/dovecot-lda -f ${sender} -d ${recipient}
On Fri, 2013-03-15 at 13:15 +0100, mourik jan heupink wrote:
dovecot unix - n n - - pipe flags=DROhu user=postfix:postfix argv=/usr/libexec/dovecot/dovecot-lda -f ${sender} -a ${recipient} -d ${user}@${nexthop}
And for what it's worth... here is mine:
dovecot unix - n n - - pipe flags=DRhu user=vmail:vmail argv=/usr/lib/dovecot/dovecot-lda -f ${sender} -d ${recipient}
That's the old method, if my hazy memory is serving me half right tonight, that will not permit recipient delimiter processing To accommodate that suggest using - -f ${sender} -d ${user}@${nexthop}
FWIW I also use "-e" was nicer for rejecting unknown users.
That's the old method, if my hazy memory is serving me half right tonight, that will not permit recipient delimiter processing To accommodate that suggest using - -f ${sender} -d ${user}@${nexthop} Ah interesting..! Is that perhaps why dovecot_destination_recipient_limit=1 was needed, here..?
On Sat, 2013-03-16 at 15:33 +0100, mourik jan heupink wrote:
destination_recipient_limit
Not sure what happened there but evolution did not like all the chars in your post when invoking reply... probably time to update this darn thing, its the last ubuntu POS that hasn't been updated to opensuse yet.
" Ah interesting..! Is that perhaps why dovecot_destination_recipient_limit=1 was needed, here..? "
No, it was to reduce the possibility of some other little quirks rearing their nasty heads IIRC.
I really find the lack of error logging, and the virtual lack of documentation for Dovecot very disturbing. I am so close to dropping this side project of being able to support multiple domains on a mail server. It is in my utmost respectful opinion to have multiple files to edit just to get this working in basic mode. Why cant dovecot combine all configuration parameters into one config like that of postfix main.cf for server config, and the master.cf for sockets and listeners?
If there are multiple files to be edited then the error logging should mention what file contains the log, and where so you can easily locate the issue. I find this software to be very lack luster and very difficult to use.
Dovecot has given me nothing but headache and grief, and as far as I am concerned not a finished product.
On Sun, Mar 17, 2013 at 12:29 AM, Noel Butler <noel.butler@ausics.net>wrote:
On Sat, 2013-03-16 at 15:33 +0100, mourik jan heupink wrote:
destination_recipient_limit
Not sure what happened there but evolution did not like all the chars in your post when invoking reply... probably time to update this darn thing, its the last ubuntu POS that hasn't been updated to opensuse yet.
" Ah interesting..! Is that perhaps why dovecot_destination_recipient_limit=1 was needed, here..? "
No, it was to reduce the possibility of some other little quirks rearing their nasty heads IIRC.
-- Daniel Reinhardt cryptodan@cryptodan.net http://www.cryptodan.net 301-875-7018(c) 410-455-0488(h)
On 3/17/2013 3:20 AM, Daniel Reinhardt wrote:
I really find the lack of error logging, and the virtual lack of documentation for Dovecot very disturbing. I am so close to dropping this side project of being able to support multiple domains on a mail server. It is in my utmost respectful opinion to have multiple files to edit just to get this working in basic mode. Why cant dovecot combine all configuration parameters into one config like that of postfix main.cf for server config, and the master.cf for sockets and listeners?
If there are multiple files to be edited then the error logging should mention what file contains the log, and where so you can easily locate the issue. I find this software to be very lack luster and very difficult to use.
Dovecot has given me nothing but headache and grief, and as far as I am concerned not a finished product.
We are very sorry you are not satisfied with the software. Please feel free to return the software for a full refund of all the money you paid.
If you do not like the multiple config files, you can always combine them.
Documentation is online at http://www.dovecot.org/documentation.html
Welcome to the world of open source. Software is provided through the generous contributions of many people, all of whom have regular jobs and do this because they enjoy it. With open source, if there is something you do not like, you can change the software to your liking. Try that with proprietary software.
And just a friendly word of advice: change your attitude or you will find your cries fall on deaf ears.
Dem
On 3/17/2013 5:25 AM, Professa Dementia wrote:
We are very sorry you are not satisfied with the software. Please feel free to return the software for a full refund of all the money you paid.
You do not speak on behalf of Dovecot, Timo, this list, and certainly not for me. Do not use "we" in this manner. And there's no reason to be a smart ass.
Welcome to the world of open source. Software is provided through the generous contributions of many people, all of whom have regular jobs and do this because they enjoy it. With open source, if there is something you do not like, you can change the software to your liking. Try that with proprietary software.
The OP is fully familiar with the open source model. Being open source and developed by volunteer effort is NEVER an excuse for crappy software or documentation. The OP has every right, and *duty* to voice his opinion, whether it be praise or critique. Without critique software doesn't get improved.
I absolutely agree with him. While the split config setup may make sense to a developer, and it may work better with some automated tools, it is counter intuitive for the majority of UNIX users. It should have been made optional, not the default. Doing this was pretty stupid and will confuse, possibly infuriate, new users, as in this case, and likely some of those upgrading as well. Things like this, as well as lackluster documentation, tend to retard adoption, or worse, drive current users to competing solutions.
And just a friendly word of advice: change your attitude or you will find your cries fall on deaf ears.
The OP's attitude is fine. It is yours that is the problem. Dissent and critique should be cherished, not attacked. We're not a bunch of little Fascists walking in lock step under de Fuhrer Timo, so stop acting like one.
-- Stan
On 2013-03-17 10:13 AM, Stan Hoeppner <stan@hardwarefreak.com> wrote:
On 3/17/2013 5:25 AM, Professa Dementia wrote:
We are very sorry you are not satisfied with the software. Please feel free to return the software for a full refund of all the money you paid. You do not speak on behalf of Dovecot, Timo, this list, and certainly not for me. Do not use "we" in this manner. And there's no reason to be a smart ass.
I'd call it more 'tongue-in-cheek'...
But the fact is, while not perfect, Dovecot is pretty well documented, and Timo is very good about fixing it when errors are pointed out.
Also, the documentation link is very prominent on dovecots home page, so the complaint about the 'virtual lack of documentation' is flat out wrong and deserves to be called out.
Also, since it (the docs) is a wiki, and since this is 'free software', there is nothing wrong with some level of expectation that others help out if/when they encounter anything that lacks.
The OP is fully familiar with the open source model.
So you know him personally? If you don't, then how can you be so sure? I suggest that his false claim of the 'virtual lack of documentation' suggests otherwise. People familiar with the open source model are also familiar with how to look for documentation for open source software.
Being open source and developed by volunteer effort is NEVER an excuse for crappy software or documentation.
So now *you* falsely claim that dovecot is 'crappy software with crappy documentation?
The nice thing about open source software is you are free to use it, make it better (either through code contributions, documentation efforts, providing support on mail lists, etc), or, don't use it at all.
I suggest that your time would be better spent doing one of the above, rather than just spewing false claims.
The OP has every right, and *duty* to voice his opinion, whether it be praise or critique. Without critique software doesn't get improved.
Does he have every right to *lie* about it? Opinions vary, but facts are facts, and the fact is, the claim that dovecot has a 'virtual lack of documentation' is an outright lie.
It may not be as good as you or others might like, and there may be certain places where it is a bit lacking, but rather than spewing false claims, maybe your time would be better spent improving it?
I absolutely agree with him.
Then you are free to do one of the above...
While the split config setup may makesense to a developer, and it may work better with some automated tools, it is counter intuitive for the majority of UNIX users.
There are those who disagree with you. I happen to *not* be one of them, I was certainly very confused by the split config when I first peeked at it, but again, as has already been pointed out, you are free to put everything in your own single config file. In fact, It is very easy to do, and also (contrary to the false claims presented) well documented:
http://wiki2.dovecot.org/BasicConfiguration
In fact, I now really like doing it this way, because I can just add all of my settings to a file named /etc/dovecot/conf.d/99-mysettings.conf, and know that they will over-ride any settings in any other files.
This makes it really easy to manage my settings.
I do something similar in postfix - I add a new section at the very end of main.cf:
*** Bgn My Custom Settings ***
my settings here
*** End My Custom Settings
This makes it very easy (for me) to manage changes and updates.
Of course, what I like may seem silly or confusing to someone else.
Isn't freedom cool? :)
It should have been made optional, not the default.
This is a design decision. Timo is the primary dovecot author, so it is his decision. You are free to disagree with it, but his choice doesn't make dovecot 'crappy software'.
Doing this was pretty stupid and will confuse, possibly infuriate, new users, as in this case, and likely some of those upgrading as well.
As long as they rtfm - and again, it *is* fairly well documented, although personally I personally think it could be made more prominent. And again - you are free to improve the documentation.
Regardless, it certainly isn't 'stupid', and is only likely to confuse and/or infuriate new users who refuse to rtfm.
Things like this, as well as lackluster documentation, tend to retard adoption, or worse, drive current users to competing solutions.
I would agree in general, but not with respect to dovecot in particular, because I don't see dovecot as having 'lackluster' documentation.
And just a friendly word of advice: change your attitude or you will find your cries fall on deaf ears.
Back at ya Stan. Normally you are pretty level-headed, and some of your posts regarding details of enterprise hardware are extremely informative, personally I think this post is way beneath you.
The OP's attitude is fine. It is yours that is the problem. Dissent and critique should be cherished, not attacked. We're not a bunch of little Fascists walking in lock step under de Fuhrer Timo, so stop acting like one.
Dissent is ok to a point, as long as it is done politely and without unnecessary flaming, and includes some minimal amount of reasonable argument.
Flame-baiting (ie making false claims that dovecot is 'crappy software with lackluster or ), on the other hand, is absolutely *not* ok, and imnsho, this is all the OP - and you - were engaging in.
When it boils down to a simple question of personal preference (ie the 'split config' issue), by all means, provide *constructive* criticism (ie, make your argument against making the split config the default) - but just name-calling and outright false-isms is *not* OK, and I for one would appreciate it if you'd re-think your comments.
Personally, I think both of you owe Timo an apology.
--
Best regards,
Charles
Seriously. Stop it. Carry on your personal vendettas with Stan and others in private, please.
- bdh
On Mar 17, 2013, at 11:39 AM, Charles Marcus <CMarcus@Media-Brokers.com> wrote:
On 2013-03-17 10:13 AM, Stan Hoeppner <stan@hardwarefreak.com> wrote:
On 3/17/2013 5:25 AM, Professa Dementia wrote:
We are very sorry you are not satisfied with the software. Please feel free to return the software for a full refund of all the money you paid. You do not speak on behalf of Dovecot, Timo, this list, and certainly not for me. Do not use "we" in this manner. And there's no reason to be a smart ass.
I'd call it more 'tongue-in-cheek'...
But the fact is, while not perfect, Dovecot is pretty well documented, and Timo is very good about fixing it when errors are pointed out.
Also, the documentation link is very prominent on dovecots home page, so the complaint about the 'virtual lack of documentation' is flat out wrong and deserves to be called out.
Also, since it (the docs) is a wiki, and since this is 'free software', there is nothing wrong with some level of expectation that others help out if/when they encounter anything that lacks.
The OP is fully familiar with the open source model.
So you know him personally? If you don't, then how can you be so sure? I suggest that his false claim of the 'virtual lack of documentation' suggests otherwise. People familiar with the open source model are also familiar with how to look for documentation for open source software.
Being open source and developed by volunteer effort is NEVER an excuse for crappy software or documentation.
So now *you* falsely claim that dovecot is 'crappy software with crappy documentation?
The nice thing about open source software is you are free to use it, make it better (either through code contributions, documentation efforts, providing support on mail lists, etc), or, don't use it at all.
I suggest that your time would be better spent doing one of the above, rather than just spewing false claims.
The OP has every right, and *duty* to voice his opinion, whether it be praise or critique. Without critique software doesn't get improved.
Does he have every right to *lie* about it? Opinions vary, but facts are facts, and the fact is, the claim that dovecot has a 'virtual lack of documentation' is an outright lie.
It may not be as good as you or others might like, and there may be certain places where it is a bit lacking, but rather than spewing false claims, maybe your time would be better spent improving it?
I absolutely agree with him.
Then you are free to do one of the above...
While the split config setup may makesense to a developer, and it may work better with some automated tools, it is counter intuitive for the majority of UNIX users.
There are those who disagree with you. I happen to *not* be one of them, I was certainly very confused by the split config when I first peeked at it, but again, as has already been pointed out, you are free to put everything in your own single config file. In fact, It is very easy to do, and also (contrary to the false claims presented) well documented:
http://wiki2.dovecot.org/BasicConfiguration
In fact, I now really like doing it this way, because I can just add all of my settings to a file named /etc/dovecot/conf.d/99-mysettings.conf, and know that they will over-ride any settings in any other files.
This makes it really easy to manage my settings.
I do something similar in postfix - I add a new section at the very end of main.cf:
*** Bgn My Custom Settings ***
my settings here
*** End My Custom Settings
This makes it very easy (for me) to manage changes and updates.
Of course, what I like may seem silly or confusing to someone else.
Isn't freedom cool? :)
It should have been made optional, not the default.
This is a design decision. Timo is the primary dovecot author, so it is his decision. You are free to disagree with it, but his choice doesn't make dovecot 'crappy software'.
Doing this was pretty stupid and will confuse, possibly infuriate, new users, as in this case, and likely some of those upgrading as well.
As long as they rtfm - and again, it *is* fairly well documented, although personally I personally think it could be made more prominent. And again - you are free to improve the documentation.
Regardless, it certainly isn't 'stupid', and is only likely to confuse and/or infuriate new users who refuse to rtfm.
Things like this, as well as lackluster documentation, tend to retard adoption, or worse, drive current users to competing solutions.
I would agree in general, but not with respect to dovecot in particular, because I don't see dovecot as having 'lackluster' documentation.
And just a friendly word of advice: change your attitude or you will find your cries fall on deaf ears.
Back at ya Stan. Normally you are pretty level-headed, and some of your posts regarding details of enterprise hardware are extremely informative, personally I think this post is way beneath you.
The OP's attitude is fine. It is yours that is the problem. Dissent and critique should be cherished, not attacked. We're not a bunch of little Fascists walking in lock step under de Fuhrer Timo, so stop acting like one.
Dissent is ok to a point, as long as it is done politely and without unnecessary flaming, and includes some minimal amount of reasonable argument.
Flame-baiting (ie making false claims that dovecot is 'crappy software with lackluster or ), on the other hand, is absolutely *not* ok, and imnsho, this is all the OP - and you - were engaging in.
When it boils down to a simple question of personal preference (ie the 'split config' issue), by all means, provide *constructive* criticism (ie, make your argument against making the split config the default) - but just name-calling and outright false-isms is *not* OK, and I for one would appreciate it if you'd re-think your comments.
Personally, I think both of you owe Timo an apology.
--
Best regards,
Charles
First of all the wiki articles on dovecots site are poorly written compared to apache and postfix. That is what I mean by lack luster the error logging is lack luster as it doesn't specify the file or the line error is on like many very well supported applications like apache and postfix and that makes dovecot not very user friendly. I owe no one an apology for stating my opinion and I have over 10 years of using open source software and dovecot is the application that I have used that given me such headache and grief to the point I have given up on this learning experience.
Daniel Reinhardt 301-875-7018(c) 410-455-0488(h)
On 2013-03-17, at 16:46, Brian Hayden <bdh@machinehum.com> wrote:
Seriously. Stop it. Carry on your personal vendettas with Stan and others in private, please.
- bdh
On Mar 17, 2013, at 11:39 AM, Charles Marcus <CMarcus@Media-Brokers.com> wrote:
On 2013-03-17 10:13 AM, Stan Hoeppner <stan@hardwarefreak.com> wrote:
On 3/17/2013 5:25 AM, Professa Dementia wrote:
We are very sorry you are not satisfied with the software. Please feel free to return the software for a full refund of all the money you paid. You do not speak on behalf of Dovecot, Timo, this list, and certainly not for me. Do not use "we" in this manner. And there's no reason to be a smart ass.
I'd call it more 'tongue-in-cheek'...
But the fact is, while not perfect, Dovecot is pretty well documented, and Timo is very good about fixing it when errors are pointed out.
Also, the documentation link is very prominent on dovecots home page, so the complaint about the 'virtual lack of documentation' is flat out wrong and deserves to be called out.
Also, since it (the docs) is a wiki, and since this is 'free software', there is nothing wrong with some level of expectation that others help out if/when they encounter anything that lacks.
The OP is fully familiar with the open source model.
So you know him personally? If you don't, then how can you be so sure? I suggest that his false claim of the 'virtual lack of documentation' suggests otherwise. People familiar with the open source model are also familiar with how to look for documentation for open source software.
Being open source and developed by volunteer effort is NEVER an excuse for crappy software or documentation.
So now *you* falsely claim that dovecot is 'crappy software with crappy documentation?
The nice thing about open source software is you are free to use it, make it better (either through code contributions, documentation efforts, providing support on mail lists, etc), or, don't use it at all.
I suggest that your time would be better spent doing one of the above, rather than just spewing false claims.
The OP has every right, and *duty* to voice his opinion, whether it be praise or critique. Without critique software doesn't get improved.
Does he have every right to *lie* about it? Opinions vary, but facts are facts, and the fact is, the claim that dovecot has a 'virtual lack of documentation' is an outright lie.
It may not be as good as you or others might like, and there may be certain places where it is a bit lacking, but rather than spewing false claims, maybe your time would be better spent improving it?
I absolutely agree with him.
Then you are free to do one of the above...
While the split config setup may makesense to a developer, and it may work better with some automated tools, it is counter intuitive for the majority of UNIX users.
There are those who disagree with you. I happen to *not* be one of them, I was certainly very confused by the split config when I first peeked at it, but again, as has already been pointed out, you are free to put everything in your own single config file. In fact, It is very easy to do, and also (contrary to the false claims presented) well documented:
http://wiki2.dovecot.org/BasicConfiguration
In fact, I now really like doing it this way, because I can just add all of my settings to a file named /etc/dovecot/conf.d/99-mysettings.conf, and know that they will over-ride any settings in any other files.
This makes it really easy to manage my settings.
I do something similar in postfix - I add a new section at the very end of main.cf:
*** Bgn My Custom Settings ***
my settings here
*** End My Custom Settings
This makes it very easy (for me) to manage changes and updates.
Of course, what I like may seem silly or confusing to someone else.
Isn't freedom cool? :)
It should have been made optional, not the default.
This is a design decision. Timo is the primary dovecot author, so it is his decision. You are free to disagree with it, but his choice doesn't make dovecot 'crappy software'.
Doing this was pretty stupid and will confuse, possibly infuriate, new users, as in this case, and likely some of those upgrading as well.
As long as they rtfm - and again, it *is* fairly well documented, although personally I personally think it could be made more prominent. And again - you are free to improve the documentation.
Regardless, it certainly isn't 'stupid', and is only likely to confuse and/or infuriate new users who refuse to rtfm.
Things like this, as well as lackluster documentation, tend to retard adoption, or worse, drive current users to competing solutions.
I would agree in general, but not with respect to dovecot in particular, because I don't see dovecot as having 'lackluster' documentation.
And just a friendly word of advice: change your attitude or you will find your cries fall on deaf ears.
Back at ya Stan. Normally you are pretty level-headed, and some of your posts regarding details of enterprise hardware are extremely informative, personally I think this post is way beneath you.
The OP's attitude is fine. It is yours that is the problem. Dissent and critique should be cherished, not attacked. We're not a bunch of little Fascists walking in lock step under de Fuhrer Timo, so stop acting like one.
Dissent is ok to a point, as long as it is done politely and without unnecessary flaming, and includes some minimal amount of reasonable argument.
Flame-baiting (ie making false claims that dovecot is 'crappy software with lackluster or ), on the other hand, is absolutely *not* ok, and imnsho, this is all the OP - and you - were engaging in.
When it boils down to a simple question of personal preference (ie the 'split config' issue), by all means, provide *constructive* criticism (ie, make your argument against making the split config the default) - but just name-calling and outright false-isms is *not* OK, and I for one would appreciate it if you'd re-think your comments.
Personally, I think both of you owe Timo an apology.
--
Best regards,
Charles
On Sun, 17 Mar 2013 17:11:18 +0000 Daniel articulated:
First of all the wiki articles on dovecots site are poorly written compared to apache and postfix. That is what I mean by lack luster the error logging is lack luster as it doesn't specify the file or the line error is on like many very well supported applications like apache and postfix and that makes dovecot not very user friendly. I owe no one an apology for stating my opinion and I have over 10 years of using open source software and dovecot is the application that I have used that given me such headache and grief to the point I have given up on this learning experience.
First of all, I don't think any sane person would argue that the Dovecot documentation is in a league with Postfix or even Apache. For that matter, I know of no other open source software that has documentation as detailed as Postfix. Try deciphering the OpenSSL documentation sometime. However, I think it is obvious that the poster was simply "venting" his frustrations out on a convenient audience.
The logging could be "tightened" up and the documentation does need work. With that said, I think we can put this matter to bed. Nothing is going to come from it so why waste time arguing over it.
-- Jerry ♔
Disclaimer: off-list followups get on-list replies or get ignored. Please do not ignore the Reply-To header.
On 17.3.2013, at 18.11, Daniel <cryptodan@gmail.com> wrote:
First of all the wiki articles on dovecots site are poorly written compared to apache and postfix.
It's the best I can do myself. I have no idea how they could be improved in any major way. They say that the software developer himself is the worst possible person to write its documentation, because he can't understand what others find difficult..
That is what I mean by lack luster the error logging is lack luster as it doesn't specify the file or the line error is on like many very well supported applications like apache and postfix and that makes dovecot not very user friendly.
If there is a syntax error, Dovecot shows the file and line number. After that it should always mention the setting name that is causing trouble, which I'd think should be easy to grep from the configs.. I guess it could be useful to show the file+line for it, but that's quite a lot of code to add just to avoid a grep. It's also a bit tricky to do without wasting more memory (wasting memory in config / doveconf process is fine, but not elsewhere, and some settings won't get processed until later).
I owe no one an apology for stating my opinion and I have over 10 years of using open source software and dovecot is the application that I have used that given me such headache and grief to the point I have given up on this learning experience.
Quickly browsing through this thread, I guess this is the main problem? :
Mar 15 06:56:37 andromeda dovecot: lda(cryptodan): Fatal: setgid(8(mail) from mail_gid setting) failed with euid=1000(cryptodan), gid=1000(cryptodan), egid=1000(cryptodan): Operation not permitted (This binary should probably be called with process group set to 8(mail) instead of 1000(cryptodan))
Yes, this is something I've been annoyed at for a long time. But it's also not easy to make that error any better, except maybe by creating a wiki page explaining the whole thing and linking to it. (There are a ton of mails about this exact thing in Dovecot list archies.) There's also no setting that is specifically related to this (the problem is a mismatch between Dovecot/Postfix configuration). There is a super easy solution though: use LMTP instead of LDA, and there are no permission troubles. Maybe that's what the LDA wiki page should say.. Done: http://wiki2.dovecot.org/LDA
On Sun, 17 Mar 2013 21:50:34 +0100 Timo Sirainen articulated:
It's the best I can do myself. I have no idea how they could be improved in any major way. They say that the software developer himself is the worst possible person to write its documentation, because he can't understand what others find difficult..
So very true Timo. That is why many fortune 500 companies use private authors to write the documentation for their products. Someone who knows nothing about it is usually the best one to write the instructions. Of course, they are guided along by competent instructors (hopefully). Unfortunately, the actual inventor or writer of an item usually takes everything for granted. That is why I have enjoyed the "Dummies" series of books. They break things down for the beginner as well as being (in most cases) detailed enough for the more experienced user.
-- Jerry ♔
Disclaimer: off-list followups get on-list replies or get ignored. Please do not ignore the Reply-To header.
Timo,
First thank you for taking the time to reply to this, but I tried using various LDA Command line suggestions from various places on the net, and Postfix is not even seeing that dovecot is listed in the master.cf file as a unix socket.
I have tried using LMTP and again it is not seeing that postfix is configured to use dovecot-lmtp as a mail delivery agent for postfix. The way postfix works with dovecot is via mail-stack-delivery and use of mailbox_command that is set up within main.cf If I comment that out postfix sends to default mode which is Maildir in the users /home directory.
I have configured SQL Lookups successfully for authentication within postfix and dovecot, its just the mail delivery agent that is not working. I even tried the sudo method, and that also failed.
Don't get me wrong Dovecot is a nice piece of software, but maybe getting rid of the multiple files could make things easier, and a couple of sample configs could then be included in the source.
And I am here trying to learn something new that I could put to use in a future job if I were ever told to setup an email server to replace an exchange system to handle multiple domains for various people.
So thanks for allowing me to learn another process of setting up a server with email.
Sincerely, Daniel Reinhardt
On Sun, Mar 17, 2013 at 8:50 PM, Timo Sirainen <tss@iki.fi> wrote:
On 17.3.2013, at 18.11, Daniel <cryptodan@gmail.com> wrote:
First of all the wiki articles on dovecots site are poorly written compared to apache and postfix.
It's the best I can do myself. I have no idea how they could be improved in any major way. They say that the software developer himself is the worst possible person to write its documentation, because he can't understand what others find difficult..
That is what I mean by lack luster the error logging is lack luster as it doesn't specify the file or the line error is on like many very well supported applications like apache and postfix and that makes dovecot not very user friendly.
If there is a syntax error, Dovecot shows the file and line number. After that it should always mention the setting name that is causing trouble, which I'd think should be easy to grep from the configs.. I guess it could be useful to show the file+line for it, but that's quite a lot of code to add just to avoid a grep. It's also a bit tricky to do without wasting more memory (wasting memory in config / doveconf process is fine, but not elsewhere, and some settings won't get processed until later).
I owe no one an apology for stating my opinion and I have over 10 years of using open source software and dovecot is the application that I have used that given me such headache and grief to the point I have given up on this learning experience.
Quickly browsing through this thread, I guess this is the main problem? :
Mar 15 06:56:37 andromeda dovecot: lda(cryptodan): Fatal: setgid(8(mail) from mail_gid setting) failed with euid=1000(cryptodan), gid=1000(cryptodan), egid=1000(cryptodan): Operation not permitted (This binary should probably be called with process group set to 8(mail) instead of 1000(cryptodan))
Yes, this is something I've been annoyed at for a long time. But it's also not easy to make that error any better, except maybe by creating a wiki page explaining the whole thing and linking to it. (There are a ton of mails about this exact thing in Dovecot list archies.) There's also no setting that is specifically related to this (the problem is a mismatch between Dovecot/Postfix configuration). There is a super easy solution though: use LMTP instead of LDA, and there are no permission troubles. Maybe that's what the LDA wiki page should say.. Done: http://wiki2.dovecot.org/LDA
-- Daniel Reinhardt cryptodan@cryptodan.net http://www.cryptodan.net 301-875-7018(c) 410-455-0488(h)
Am 18.03.2013 00:27, schrieb Daniel Reinhardt:
First thank you for taking the time to reply to this, but I tried using various LDA Command line suggestions from various places on the net, and Postfix is not even seeing that dovecot is listed in the master.cf file as a unix socket.
I have tried using LMTP and again it is not seeing that postfix is configured to use dovecot-lmtp as a mail delivery agent for postfix
YOU ARE RESPONSIBLE TO CONFIGURE POSTFIX MASTER.CF NOT DOVECOT NOT ANYBODY ELSE
YOU ARE RESPONSIBLE TO SET POSTFIX-TRANSPORTS TO USE LMTP NOT DOVECOT NOT ANYBODY ELSE
RTFM - NOT THE DOVECOT THE POSTFIX ONES
On 3/17/2013 3:50 PM, Timo Sirainen wrote:
It's the best I can do myself. I have no idea how they could be improved in any major way. They say that the software developer himself is the worst possible person to write its documentation, because he can't understand what others find difficult..
I don't know who "they" is. Wietse writes all the Postfix documentation himself. It comes naturally when one performs formal software development, not ad hoc, because documentation precedes coding. I would assume you do ad hoc development like most 20 somethings, coding on the fly when you get an idea, no formal definitions, no flow charting, no pseudo code, etc. Correct? If so this is 99% of the reason the documentation suffers, and this is typical of today's crop of young developers, unfortunately.
For highly technical material the author is the only person qualified to write the docs. Having a 3rd party do it has a prerequisite of a Vulcan mind meld. Otherwise you talk and they type, which is slower than you doing it yourself.
A few things could improve the current docs in a major way.
- Create man(ual) documentation, preferably with
- A man page like postconf (5) which contains every single Dovecot configuration parameter and text explaining it
- This man page published online
- Publish sample conf file(s) online
- Make these things accessible from the main Dovecot page, not buried down in the index hierarchy
I've always perceived the dovecot wiki docs with the hierarchical book, chapter, verse, mini how-to layout as a dessert you assemble from the buffet--a little cake, some pudding, a dab of whipped cream, chopped nuts, and a cherry on top. You end up with a dessert, empty calories, not a complete meal. You can't get full and keep going back, assembling another dessert each time.
Typical UNIX documentation is steak and potatoes, veggies, and a dinner roll. You sit down, eat, and you're full. No running around collecting your food as you've got everything you need on one plate, and it's a complete meal.
-- Stan
On Mon, 18 Mar 2013, Stan Hoeppner wrote:
On 3/17/2013 3:50 PM, Timo Sirainen wrote:
It's the best I can do myself. I have no idea how they could be improved in any major way. They say that the software developer himself is the worst possible person to write its documentation, because he can't understand what others find difficult..
I don't know who "they" is. Wietse writes all the Postfix documentation himself. It comes naturally when one performs formal software development, not ad hoc, because documentation precedes coding. I would assume you do ad hoc development like most 20 somethings, coding on the fly when you get an idea, no formal definitions, no flow charting, no pseudo code, etc. Correct? If so this is 99% of the reason the documentation suffers, and this is typical of today's crop of young developers, unfortunately.
For highly technical material the author is the only person qualified to write the docs. Having a 3rd party do it has a prerequisite of a Vulcan mind meld. Otherwise you talk and they type, which is slower than you doing it yourself.
Software needs two types of documentation. It needs overview type documentation that describes what it is. This is for the person who is looking for a product and has no idea if this specific product can meet the need.
Second, it needs reference documentation that provides syntax for commands, config files, and the like. This is for the person who is or will be using the product and needs the details to use it properly for the need.
Somewhere in between is documentation that point the user to the right reference section. Providing detailed documentation of a command is worthless if nothing tells me that that command is what I should use for my need.
The reference documentation may well best be written by the software author who knows exactly what syntax, etc. is needed. It's also relatively (and I emphasize "relatively") easy since it can be done as you go.
But the overview documentation may well best be written by someone who knows nothing about it. The expert writing overview documentation may assume the reader knows things he doesn't. It can make sense to the author but leaves the reader without a clue as to what is being discussed. When the author is someone unfamiliar with the product, he asks questions until it makes sense and there is less tendancy for the documentation to assume knowledge by the reader that is not there.
Overview documentation is a lot tougher to write well and it needs someone with good writing skills (not good programming skills). Particularly in the open-source world where enhancements can come quickly, it can be out of date as soon as it's written.
-- Larry Stone lstone19@stonejongleux.com
On Mon, 2013-03-18 at 10:33 -0500, Stan Hoeppner wrote:
On 3/17/2013 3:50 PM, Timo Sirainen wrote:
It's the best I can do myself. I have no idea how they could be improved in any major way. They say that the software developer himself is the worst possible person to write its documentation, because he can't understand what others find difficult..
I don't know who "they" is. Wietse writes all the Postfix documentation himself. It comes naturally when one performs formal software development, not ad hoc, because documentation precedes coding. I would assume you do ad hoc development like most 20 somethings, coding on the fly when you get an idea, no formal definitions, no flow charting, no pseudo code, etc. Correct? If so this is 99% of the reason the documentation suffers, and this is typical of today's crop of young developers, unfortunately.
Because it significantly increases development times, and when you're basically doing everything yourself there's nobody else reading those anyway.
The more complex a feature is, the more I think about it, do pseudo code and testing. For example the redesigned dsync in v2.2 required months of thinking, pseudo coding and testing. Few features in Dovecot are that complex though. Mostly coding is the easy part, while figuring out how the configuration should be done is the difficult part.
Anyway, the plan is to hire more Dovecot developers and the development process is likely to change then. But now? I'm way too busy implementing things that were supposed to be finished half a year ago.
A few things could improve the current docs in a major way.
- Create man(ual) documentation, preferably with
- A man page like postconf (5) which contains every single Dovecot configuration parameter and text explaining it
- This man page published online
- Publish sample conf file(s) online
- Make these things accessible from the main Dovecot page, not buried down in the index hierarchy
So basically you're saying that the major documentation improvement = an index listing/describing all settings. Sure, would be useful, but I don't see having time to write that anytime soon.
On Mon, 2013-03-18 at 18:37 +0200, Timo Sirainen wrote:
I don't know who "they" is. Wietse writes all the Postfix documentation himself. It comes naturally when one performs formal software development, not ad hoc, because documentation precedes coding. I would assume you do ad hoc development like most 20 somethings, coding on the fly when you get an idea, no formal definitions, no flow charting, no pseudo code, etc. Correct? If so this is 99% of the reason the documentation suffers, and this is typical of today's crop of young developers, unfortunately.
Because it significantly increases development times, and when you're basically doing everything yourself there's nobody else reading those anyway.
Or actually for the larger changes I do write design docs and usually send them to this mailing list, e.g.:
http://www.dovecot.org/list/dovecot/2012-February/064114.html http://www.dovecot.org/list/dovecot/2012-February/063665.html http://www.dovecot.org/list/dovecot/2010-November/055196.html http://www.dovecot.org/list/dovecot/2010-July/050832.html http://www.dovecot.org/list/dovecot/2010-January/046148.html
So it's not all ad hoc..
On 18.03.2013, at 17:37, Timo Sirainen <tss@iki.fi> wrote:
Because it significantly increases development times, and when you're basically doing everything yourself there's nobody else reading those anyway.
From my point of view: do continue the very same way every since writing your first code regarding dovecot! I don't see that skilled admins won't be able to config dovecot with all that available information accessible by skilled admins. I honestly couldn't care less regarding all those unskilled admins, though.
The more complex a feature is, the more I think about it, do pseudo code and testing. For example the redesigned dsync in v2.2 required months of thinking, pseudo coding and testing.
From my point of view: no need for "excuses"! And, coming back to dovecot's documentation: it can't be that worse because dovecot is running at quite some numerous sites ;-)
From my point of view: just continue to push dovecot the way you did and do, please, don't become distracted by discussions like that one, sigh.
With kind regards, Michael
On 3/18/2013 11:37 AM, Timo Sirainen wrote:
So basically you're saying that the major documentation improvement = an index listing/describing all settings. Sure, would be useful, but I don't see having time to write that anytime soon.
The time issue is perfectly understandable Timo.
My suggestion may not be the gold or platinum improvement to the docs, but I think it would help a lot of people, especially since most using Dovecot are also using Postfix, and since man is the standard UNIX documentation format/interface. I think some similarity/consistency would help quite a bit as many people are so used to this format.
Do you have a way to simply dump all the current conf file parameter names from 2.x into a single column text file? I'll sort it and start adding the legal parameter values and writing the parameter definitions from information currently available in source and wiki pages. When I hit the point I can't find reference material for the rest of the parameters, we can dump it to a wiki page or similar so others with the knowledge can jump in and help finish it. Once it's done, myself, or someone else if they already have the experience, can create the man page from this to be included in the source. And you can create an update mechanism/batch process so that updating the 'master' document automatically updates the source man page and other published versions, making documentation updates simple when you add/change parameters.
We could do the wiki bazaar style editing from the beginning, but I'd rather not. I'd like to get it started with a framework/layout and style of prose typical of UNIX documentation, for other editors to follow. The definition text prose needs to be consistent all the way through, or readers may be confused by the different writing styles of ~50 different people who may speak different 'dialects' of English or have different writing styles. This consistency is one of the hallmarks of good technical writing.
Like I said previously, the one thing I'm able to contribute more than anything at this point is time. And my writing skills aren't completely horrible--I have been published, FWIW, but not recently. But my knowledge of the parameters, and a lot of Dovecot features in general is lacking. So if others are willing to contribute where I fall short, I'd be glad to give this a go and get it started, and hopefully put a decent sized dent in it so there's not so much left for others to do. Obviously you have final review/edit authority, and if you have a particular preference on writing style, etc, I'll certainly honor that.
If this is acceptable to you Timo, let me know. If so send me the aforementioned file, any preferences/thoughts you have, and I'll get started on the first draft.
-- Stan
El 19/03/13 05:15, Stan Hoeppner escribió:
On 3/18/2013 11:37 AM, Timo Sirainen wrote:
So basically you're saying that the major documentation improvement = an index listing/describing all settings. Sure, would be useful, but I don't see having time to write that anytime soon.
The time issue is perfectly understandable Timo.
My suggestion may not be the gold or platinum improvement to the docs, but I think it would help a lot of people, especially since most using Dovecot are also using Postfix, and since man is the standard UNIX documentation format/interface. I think some similarity/consistency would help quite a bit as many people are so used to this format.
Do you have a way to simply dump all the current conf file parameter names from 2.x into a single column text file? I'll sort it and start adding the legal parameter values and writing the parameter definitions from information currently available in source and wiki pages. When I hit the point I can't find reference material for the rest of the parameters, we can dump it to a wiki page or similar so others with the knowledge can jump in and help finish it. Once it's done, myself, or someone else if they already have the experience, can create the man page from this to be included in the source. And you can create an update mechanism/batch process so that updating the 'master' document automatically updates the source man page and other published versions, making documentation updates simple when you add/change parameters.
We could do the wiki bazaar style editing from the beginning, but I'd rather not. I'd like to get it started with a framework/layout and style of prose typical of UNIX documentation, for other editors to follow. The definition text prose needs to be consistent all the way through, or readers may be confused by the different writing styles of ~50 different people who may speak different 'dialects' of English or have different writing styles. This consistency is one of the hallmarks of good technical writing.
Like I said previously, the one thing I'm able to contribute more than anything at this point is time. And my writing skills aren't completely horrible--I have been published, FWIW, but not recently. But my knowledge of the parameters, and a lot of Dovecot features in general is lacking. So if others are willing to contribute where I fall short, I'd be glad to give this a go and get it started, and hopefully put a decent sized dent in it so there's not so much left for others to do. Obviously you have final review/edit authority, and if you have a particular preference on writing style, etc, I'll certainly honor that.
If this is acceptable to you Timo, let me know. If so send me the aforementioned file, any preferences/thoughts you have, and I'll get started on the first draft.
Definitely, something like man 5 postconf would be really useful. I would like to collaborate with that, but I think that my English writing skill are not good enough.
On Tue, 19 Mar 2013 13:36:53 +0100 Joseba Torre articulated:
{snip}
Definitely, something like man 5 postconf would be really useful. I would like to collaborate with that, but I think that my English writing skill are not good enough.
I would be willing to assist in a project like that. If we could get a few knowledgeable people -- including Timo -- I think it would be a truly worthwhile project.
-- Jerry ♔
Disclaimer: off-list followups get on-list replies or get ignored. Please do not ignore the Reply-To header.
On 19 March 2013 15:20, Jerry <jerry@seibercom.net> wrote:
On Tue, 19 Mar 2013 13:36:53 +0100 Joseba Torre articulated:
{snip}
Definitely, something like man 5 postconf would be really useful. I would like to collaborate with that, but I think that my English writing skill are not good enough.
I would be willing to assist in a project like that. If we could get a few knowledgeable people -- including Timo -- I think it would be a truly worthwhile project.
I can't code, but I can proof-read/write. And if *I* understand the instructions/config examples you have winning documentation - the ultimate dummy test, so to speak. So, this could be my opportunity to contribute to FOSS.
Simon
On Mon, 2013-03-18 at 23:15 -0500, Stan Hoeppner wrote:
Do you have a way to simply dump all the current conf file parameter names from 2.x into a single column text file?
With "doveconf -d" you get all the settings and also the defaults. The docs probably should mention the defaults also.
I'll sort it and start adding the legal parameter values and writing the parameter definitions from information currently available in source and wiki pages. When I hit the point I can't find reference material for the rest of the parameters, we can dump it to a wiki page or similar so others with the knowledge can jump in and help finish it. Once it's done, myself, or someone else if they already have the experience, can create the man page from this to be included in the source. And you can create an update mechanism/batch process so that updating the 'master' document automatically updates the source man page and other published versions, making documentation updates simple when you add/change parameters.
Yes, definitely something that generates all the docs from a single source. There is of course still going to be some duplication with a) example config files and b) the more context-specific wiki pages.
I guess once that reference doc is done, the example config could be put to web with all the settings as links to the reference.
I think the reference should also have pointers to the more generic wiki pages about the subject, such as ssl_* settings having a pointer to the "SSL" wiki page. That pointer could be a generic small icon in the HTML/wiki version, not sure about the man version.
If this is acceptable to you Timo, let me know. If so send me the aforementioned file, any preferences/thoughts you have, and I'll get started on the first draft.
OK.
On Mon, Mar 18, 2013 at 11:33 AM, Stan Hoeppner <stan@hardwarefreak.com> wrote:
A few things could improve the current docs in a major way.
FWIW I've found the exemplary postconf(1) almost indispensable both for exploring the Postfix configuration and for applying impromptu changes.
On 2013-03-18 12:38 PM, Ajax <aajaxx@gmail.com> wrote:
FWIW I've found the exemplary postconf(1) almost indispensable both for exploring the Postfix configuration and for applying impromptu changes.
I think most everyone would agree that postfix is in a class by itself when it comes to code quality and documentation...
I was going to just bite my tongue, but couldn't let this go...
On 2013-03-18 11:33 AM, Stan Hoeppner <stan@hardwarefreak.com> wrote:
On 3/17/2013 3:50 PM, Timo Sirainen wrote:
It's the best I can do myself. I have no idea how they could be improved in any major way. They say that the software developer himself is the worst possible person to write its documentation, because he can't understand what others find difficult.. I don't know who "they" is.
I think it was pretty obvious that he meant 'the documentation'... talk about being a smart ass.
Wietse writes all the Postfix documentation himself. It comes naturally when one performs formal software development, not ad hoc, because documentation precedes coding.
<snip>
Really nice Stan. You just proved the old axiom of what happens when you 'ass-u-me'.
Writing good docs may come easy to some people (I can't speak for Wietse as to whether or not this is true for him), but I'd argue that it may not be as easy as you seem to think.
Too bad the rest of your decent suggestions got clobbered by the condescending attitude. And I wonder how much time you have spent helping Timo get the documentation in order...
On 3/17/2013 11:39 AM, Charles Marcus wrote:
On 2013-03-17 10:13 AM, Stan Hoeppner <stan@hardwarefreak.com> wrote:
On 3/17/2013 5:25 AM, Professa Dementia wrote:
And just a friendly word of advice: change your attitude or you will find your cries fall on deaf ears.
Back at ya Stan. Normally you are pretty level-headed, and some of your posts regarding details of enterprise hardware are extremely informative, personally I think this post is way beneath you.
Follow the threading Charles. The comment above belongs to Dementia, not me.
The OP's attitude is fine. It is yours that is the problem. Dissent and critique should be cherished, not attacked. We're not a bunch of little Fascists walking in lock step under de Fuhrer Timo, so stop acting like one.
Dissent is ok to a point, as long as it is done politely and without unnecessary flaming, and includes some minimal amount of reasonable argument.
The only flaming here was done by Dementia. Note that Daniel stated in his complaint: "in my utmost respectful opinion"
Flame-baiting (ie making false claims that dovecot is 'crappy software with lackluster or ), on the other hand, is absolutely *not* ok, and imnsho, this is all the OP - and you - were engaging in.
You perceived an attack and picked up sword and shield to defend Dovecot and Timo. Unfortunately your adrenaline got the best of you. Neither Daniel nor I engaged in flame baiting. You would do well to drop the sword and shield and re-read both of our posts dispassionately.
I did not call Dovecot crappy software as you accuse me of here. If that were my opinion I certainly would not be using it and I wouldn't be participating on this mailing list. Dementia made the case that users are responsible for fixing anything they don't like about open source software, suggesting it's ok to publish crappy software and make it the users' responsibility to fix it.
When it boils down to a simple question of personal preference (ie the 'split config' issue), by all means, provide *constructive* criticism (ie, make your argument against making the split config the default) -
We both did.
but just name-calling and outright false-isms is *not* OK, and I for one would appreciate it if you'd re-think your comments.
Neither of us engaged in name calling nor made false claims. We did both express frustration. And yes I did use the word stupid. And Linus told nVidia "Fuck you!" on camera. I'd say "stupid" is very tame, blunt, and to the point. Anyone who is so thin skinned as to take offense to "stupid" isn't tough enough to participate on the interwebs.
Personally, I think both of you owe Timo an apology.
Personally I would prefer you let the air out of your chest Charles, drop the sword and the shield, and re-read the email exchange dispassionately. ;) I think you know me well enough through our extensive interaction that I would not make ad hominem attacks or make baseless accusation against Dovecot or Timo.
I think Timo's skin is sufficiently thick to take a little criticism and not lose sleep over it. Apologies are for personal attacks. I made no personal attacks.
-- Stan
All of this said (and much of it not worth repeating), one problem that seems to affect all software as it grows is that as documentation is "patched" to describe new features, it becomes too complex for the new user who just wants to do something simple to figure how to do that simple stuff. For the user who has been along for the long ride since the software started, it makes sense but the new user is overwhelmed. Rewriting documentation is no easy task but it can help for someone to take a look at it the way a new user might who knows nothing about the software.
I don't know the history of Dovecot but my guess would be the Dovecot LDA was added after the Dovecot POP/IMAP server component. Why? Because the www.dovecot.org Overview says "Dovecot is an open source IMAP and POP3 email server for Linux/UNIX-like systems" without any mention of the Dovecot LDA anywhere on that front page. Longtime users know about the Dovecot LDA but they rarely read that first page and it's harder to notice something is missing than it is to notice something is wrong.
I recently replaced UW-IMAP with Dovecot. Once I set down to do it, it was fairly easy. But getting there was tough thanks to misinformation including a failed attempt a few years back using a package that tried to do too much (including forcing a conversion to Maildir - it may be the preferred way and it might be a good goal to get there but why force it when Dovecot was perfectly content to work with the existing mbox mailboxes).
Coupled with that is the problem of people who like to give complex answers to simple questions. A poster wants to know how to do A and gets an answer that instead of doing A (which would require a simple configuration change), they're better off doing B which requires new software and/or a complicated conversion (I was just reading something in an archive where someone asked about locking of mbox files and instead of answering the question about how to configure that was told they're better off using Maildir. Perhaps true in the long run but a config change takes a few seconds; converting to Maildir can be a multi-hour or day project). Giving complex answers to simple questions creates the impression that the Dovecot is far more complex than it needs to be.
-- Larry Stone lstone19@stonejongleux.com http://www.stonejongleux.com/
Am 17.03.2013 21:12, schrieb Larry Stone:
Giving complex answers to simple questions creates the impression that the Dovecot is far more complex than it needs to be.
"mail" isnt simple in general, if you wanna help getting better or more easy documentation at whatever feel free to write it, your welcome
Actions speak louder than words.
Best Regards MfG Robert Schetterer
-- [*] sys4 AG
http://sys4.de, +49 (89) 30 90 46 64 Franziskanerstraße 15, 81669 München
Sitz der Gesellschaft: München, Amtsgericht München: HRB 199263 Vorstand: Patrick Ben Koetter, Axel von der Ohe, Marc Schiffbauer Aufsichtsratsvorsitzender: Joerg Heidrich
On Mar 17, 2013, at 4:50 PM, Robert Schetterer <rs@sys4.de> wrote:
Am 17.03.2013 21:12, schrieb Larry Stone:
Giving complex answers to simple questions creates the impression that the Dovecot is far more complex than it needs to be.
"mail" isnt simple in general,
True. But your needs aren't my needs. And while some systems may need a complex solution, not every one does.
My mail server is at home serving four users, all family members. Daily mail volume is in the hundreds. Have I incorporated every high-performance trick I can? Heck, no because I don't need it. My server sits idle most of the day. It would be a waste of my time to optimize it. To the extent I have a bottle-neck, it's the Internet connection (not all that fast but fast enough).
So for now I keep it simple. Can I do something more complex later? Maybe. If I have time, if I think it's worthwhile. Meanwhile, I would never go suggesting that the server environment I have with my four users and hundreds of messages per day is appropriate for an environment with thousands of users and millions of messages per day.
-- Larry Stone lstone19@stonejongleux.com http://www.stonejongleux.com/
Am 18.03.2013 01:02, schrieb Larry Stone:
On Mar 17, 2013, at 4:50 PM, Robert Schetterer <rs@sys4.de> wrote:
Am 17.03.2013 21:12, schrieb Larry Stone:
Giving complex answers to simple questions creates the impression that the Dovecot is far more complex than it needs to be.
"mail" isnt simple in general,
True. But your needs aren't my needs. And while some systems may need a complex solution, not every one does.
My mail server is at home serving four users, all family members. Daily mail volume is in the hundreds. Have I incorporated every high-performance trick I can? Heck, no because I don't need it. My server sits idle most of the day. It would be a waste of my time to optimize it. To the extent I have a bottle-neck, it's the Internet connection (not all that fast but fast enough).
So for now I keep it simple. Can I do something more complex later? Maybe. If I have time, if I think it's worthwhile. Meanwhile, I would never go suggesting that the server environment I have with my four users and hundreds of messages per day is appropriate for an environment with thousands of users and millions of messages per day.
Larry , this isnt a whatever software related question, it a questions of what default first setup settings are choosen, to make things work quick.
There maybe tons of meanings about that, you will never find the ultimate answer to that, its a more or less senseless disput.
I.e good example using default chroot for postfix in debian ( which gives newbies a hard time at first configure), so at last the orginal hacker of code has less to to do what linux distro releases use as default settings.
And however your are using only small setup you have to understand and learn the whole "mail" procedures ,to some "minimum stage", cause i.e you are able to send trillions of spam mails configure an open relay etc which may hurt lots of other users worldwide.
Open software mostly does not focus on promising you an easy live with i.e clicking on guis and wizards as most comercial products, it gives you the chance to learn about how things are working, and invites you making things better which you may find could be made better. Share what you have found and do constructive questions etc. Its not only a tec question, its a way of social knowledge sharing.
Dovecot itself may run in very different "modes", what should the choosen as prime default easy mode, compare i.e with samba and modes acting as member server or printer spooler and so on. Samba parameters are filling books, same in postfix etc, and there is intensive coding work at all this stuff, solving users real live problems always end in complex stuff, by needs which you and the orig coder never thinked of as starting time of the project.
So again, if you find something bad , youre able to make things better investing time and work on it, dont waste time in "bark at the moon". .
Best Regards MfG Robert Schetterer
-- [*] sys4 AG
http://sys4.de, +49 (89) 30 90 46 64 Franziskanerstraße 15, 81669 München
Sitz der Gesellschaft: München, Amtsgericht München: HRB 199263 Vorstand: Patrick Ben Koetter, Axel von der Ohe, Marc Schiffbauer Aufsichtsratsvorsitzender: Joerg Heidrich
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On 03/17/2013 07:13 AM, Stan Hoeppner wrote:
On 3/17/2013 5:25 AM, Professa Dementia wrote:
We are very sorry you are not satisfied with the software. Please feel free to return the software for a full refund of all the money you paid.
This is a formulaic response.
You do not speak on behalf of Dovecot, Timo, this list, and certainly not for me. Do not use "we" in this manner. And there's no reason to be a smart ass.
Right.
Welcome to the world of open source. Software is provided through the generous contributions of many people, all of whom have regular jobs and do this because they enjoy it. With open source, if there is something you do not like, you can change the software to your liking. Try that with proprietary software.
More formulaic response.
The OP is fully familiar with the open source model. Being open source and developed by volunteer effort is NEVER an excuse for crappy software or documentation. The OP has every right, and *duty* to voice his opinion, whether it be praise or critique. Without critique software doesn't get improved.
I don't know if the "[original poster] is fully familiar with the open source model" or not, but I would save the rest of this response as a gem. The formulaic response, which I have seen again and again in the over ten years I've been working with open source software, really assumes that all users are programmers--or should be programmers--and are responsible for submitting fixes when something is wrong.
Because what it really says is, if you don't like it, fix it yourself. Too many times, I've heard this referred to as "freedom." And like some other notions of freedom advanced in our society, it is only freedom for a limited class of people.
There are lots of these kinds of assumptions throughout our society. But the fact is that not all of our talents align in the same directions. That's why we have specialization of labor.
The next logical step in this assumption is that a dissatisfied user should, having become a programmer, fork the project, maintain and develop that fork, and integrate it at least with whatever distribution/variety of UNIX-like operating system s/he is using. Not all of us have time to do this. Again, that's why we have specialization of labor.
<snip>
The OP's attitude is fine. It is yours that is the problem. Dissent and critique should be cherished, not attacked. We're not a bunch of little Fascists walking in lock step under de Fuhrer Timo, so stop acting like one.
More words to treasure. (Not that Timo envisions himself any kind of Fuehrer.) This reaches to a certain deification of certain individuals and organizations in the community--whether they themselves would be deified or not; I know examples of both--that really ought to be looked at more critically.
-----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.19 (GNU/Linux) Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/
iQIcBAEBAgAGBQJRRj9gAAoJELJhbl/uPb4SWtoQAJm1/aHlmQQu20o/8EvOiB9I a04lh7aLDD2DDq/09M0wj4Rjd02g1PP9AYpaHkPbnIPGm6TsiMy9d6uN7VHlPMN8 2ylc4aVUSpu564LvuMolLQxuWszdt2VHA4T2DYxdTBEYOpZKI2Yn8oBniP1MA3VW 7OP1Jr0BYtn2tffU6RHguRT+/ApSDKv/owI0aJorL0ql0QA9WlXKlKz4A5SvvRHU siZT7CyoxtaVn2SDIZRMrV4T/5gSZyxUatiIw8brNkhyaJb7TptLfusxgueUHEgT ZR84qyK2uenVP9LiFG7Ur8tsSU1ANmNSnGJyQLL8FJefsX4a1s0fhTb0U6ZenJsh u1OtOIF3KVtIvIb0tlBdX2n3nDDlXanvQW6Wg5qIBI0Fpk3BupDrxnKYicKCpaMc rj5tgAR5xNOpvGVRlaIspMw6+64xD53rMxWur3U7QUHqNBr/IodDc7ySJzbx9IJi /pG2ll+2TtT2fj+zIT5nLiPyl3R7hs4SE5JM52adQBtLmmS6vZwWEeMSTT6iBokn eXytWgY1+hdE8ldZdWropPd3tu0kfiPZSHE+bc9/Yj09P53FbrHD7QxE8j3HeG+v PtjKvdSmC2dDzuRp44tUoMkS6u3Y3NjT8ZYRrYHFlXLURLcKqkfziXEXjYEYYsIG YNbYpbWeIg5/z1yXnQI4 =I801 -----END PGP SIGNATURE-----
On 3/17/2013 3:10 PM, David Benfell wrote:
I don't know if the "[original poster] is fully familiar with the open source model" or not, but I would save the rest of this response as a gem. The formulaic response, which I have seen again and again in the over ten years I've been working with open source software, really assumes that all users are programmers--or should be programmers--and are responsible for submitting fixes when something is wrong.
Because what it really says is, if you don't like it, fix it yourself. Too many times, I've heard this referred to as "freedom." And like some other notions of freedom advanced in our society, it is only freedom for a limited class of people.
Imagine this:
You live near a mechanic. He has an extra car that he got from the junk yard and fixed up. It works fairly well and you need a car so he generously gives it to you for free. After a time, something breaks. What do you do?
If you are mechanically inclined you can fix it yourself - that is, if you are part of that limited class of people known as mechanics.
If not, you can go back to the original mechanic or maybe a completely different mechanic. However, if you scream at him, call him names, tell him the car is total garbage that was never fixed right in the first place, then demand he fix it, how do you think he would respond?
It is much better to ask politely.
Everything you say above is perfectly valid. However, I have been part of quite a few open source projects and this list has much more than its share of trolls and flame wars. Why is that? A little civility goes a long way, especially if you are not part of the "limited class of people" and are asking for their help with something that you were given for free to begin with.
Dem
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On 03/17/2013 06:08 PM, Professa Dementia wrote:
If not, you can go back to the original mechanic or maybe a completely different mechanic. However, if you scream at him, call him names, tell him the car is total garbage that was never fixed right in the first place, then demand he fix it, how do you think he would respond?
It is much better to ask politely.
As I was looking for the response that I wanted to reply to, I noticed there had in fact been quite a number of messages as the original poster had attempted to resolve his problem.
Yes, he got frustrated. Yes, he lost his temper. But this summation takes a short cut in ignoring all that happened before. It also ignores--and if I read (and remember) Timo correctly, he doesn't like this either--that all too often, error messages are too vague to be helpful.
The standard response is also borne of frustration--a frustration that seemingly expects people to worship the ground that developers walk on.
So what we have here is frustration meeting frustration. It pushes my buttons because I've been watching this in a number of venues for years and I've occasionally been that frustrated user who couldn't get something working.
One of the things I came across in studying indigenous ways of knowing that may be applicable here is a caution against displaying knowledge. The point of that is that some people answer not so much to be helpful as to show off their own prowess and expertise. That gets ego involved. The standard response follows from this, making the situation worse by conveying an arrogance that undermines any claim to have been sincerely helpful before.
We all want good software. That requires more of a partnership than we often see.
-----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.19 (GNU/Linux) Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/
iQIcBAEBAgAGBQJRRphOAAoJELJhbl/uPb4SThcQAI8ipyEw8OVyVhzZLkrpU5fv 2X4uz6WAJQMgkAOCG8M5RX/FDdm0b4puxcwalYsx4V0Pv2mZa49vJPHwooFjHM/s h6x34zeip5pd+nq519pCav+et0MLD5c49iUrPL4IEID2eEiPrBKq5LCK79q7HoJ4 lwPwqRpyyQUqznScq1seLzgMBFs/isNs2+uCj5i9DjHGON7mFURP+lHwcDhEvAKC F4y1rh54vuxTChGGROYBKHgzISJwcomuS3OROjQU7nPnhb05SDMrcn3j2Zz2xAhC +rFP5Gp3mvIAJLL7jybgFph8DgeBx2RaN5VW41HKvGC159bBwIDXIY34rWEh8mk+ KoBYQUIH2YbCBBFWOAAArKr3JUCq8cb/lb+3wblSbyAeOXYhrQWOLSGC7AU+KKxZ g+x5O8fzBwibHE3i7l+NQFJLlLyXJbdHgL1ZEO128xl5cTEewXE0u/xzPyNdPqmH S1vIe2uo8qPaoodGqUOEwQIhcCfSPLVmFjdn9kOI4VqjFa7sRqL6O0VFXnUcGhju FFw1+vctHDjPnZmuNKAiddZh1gwF5l/tGPQ/5T8rwxuleAuMesefV8R2vYTzocQV mTejoxGQe3GOzRqB9gHX19Gk8aJChHR2TXwqkdo/9IGFrYa3ZTYzsrpuwx6CcCOr Z5GGtVgZ9YedSzPIcP/g =OOy+ -----END PGP SIGNATURE-----
On Sun, 2013-03-17 at 21:30 -0700, David Benfell wrote:
The standard response is also borne of frustration--a frustration that seemingly expects people to worship the ground that developers walk on.
One thing that shows a good developer, is one who listens, to the negatives as well as positives, Timo thus far has mostly done this (still waiting on the mysql failover change talked about nearly 2 yrs ago ;) ).
A good developer knows that without users feedback, both good and bad, their project can die a swift and painful death.
We all want good software. That requires more of a partnership than we often see.
True
On 3/17/2013 5:10 PM, David Benfell wrote:
More words to treasure. (Not that Timo envisions himself any kind of Fuehrer.) This reaches to a certain deification of certain individuals and organizations in the community--whether they themselves would be deified or not; I know examples of both--that really ought to be looked at more critically.
To be crystal clear, since some people tend to take things out of context, I am in no way likening Timo to Adolf Hitler for Pete's sake. Anyone who could perceive such has personal issues to work out. This was quickest analogy I could come up with to describe the situation, as it seemed anyone criticizing "the party" or it's "leader" was being treated as "the enemy", which is indeed what the Nazi party did to dissenters. It ties in well with my "sword and shield" comment, and the "militarization" of some users in response to a perceived "attack" in this thread.
It's actually quite silly how this thread devolved so quickly simply because two users expressed frustration with the software and documentation. I'd expect something like this on a political forum, but not a technical mailing list.
-- Stan
On Sun, 2013-03-17 at 09:13 -0500, Stan Hoeppner wrote:
On 3/17/2013 5:25 AM, Professa Dementia wrote:
We are very sorry you are not satisfied with the software. Please feel free to return the software for a full refund of all the money you paid.
You do not speak on behalf of Dovecot, Timo, this list, and certainly not for me. Do not use "we" in this manner. And there's no reason to be a smart ass.
I never ever in my wildest dreams thought I'd be saying this, but I wholeheartedly agree with Stan
On Sun, 2013-03-17 at 10:20 +0000, Daniel Reinhardt wrote:
I really find the lack of error logging, and the virtual lack of documentation for Dovecot very disturbing. I am so close to dropping this side project of being able to support multiple domains on a mail server.
I know I came in late having been away for a bit, but I did not note any reference to this guide you mentioned you used, perhaps show us a URL for it, maybe it is borked, and you'll chase your tail forever...
It is in my utmost respectful opinion to have multiple files to edit just to get this working in basic mode. Why cant dovecot combine all configuration parameters into one config like that of postfix main.cf for server config, and the master.cf for sockets and listeners?
It used to, I strongly disagreed with the conf.d/blah blah blah splitting, its messy, cluttery, clumsy, and not time-management efficient, but, Timo's choice to do it.
I use one single file, its about or less than 4K in size, and you can too, simple copy any million changes of other cruddy confs into dovecot.conf and comment out any includes, even delete them.
The only separate file I have, is dovecot-sql.conf
Dovecot has given me nothing but headache and grief, and as far as I am concerned not a finished product.
it runs on some pretty big networks, handling millions of users, granted 2.x was, well, I always considered 2.0 to be beta, and 2.1 up until recently, it is not a good sign when there is patches every second other week, but, its been really good for a few months now, sad I know, 1.2 was like a tank, couldn't break it no mater what you through at it, infact I still have two servers using 1.2 as a fail safe.
But, that said, you need to show us this guide you have followed, if you posted it, my bad for not seeing it, if not, your bad for not linking to it ;)
participants (20)
-
Ajax
-
Brian Hayden
-
Charles Marcus
-
Daniel
-
Daniel Reinhardt
-
David Benfell
-
Jerry
-
Joseba Torre
-
Larry Stone
-
Michael Grimm
-
mourik jan c heupink
-
mourik jan heupink
-
Noel Butler
-
Professa Dementia
-
Reindl Harald
-
Robert Schetterer
-
Simon Brereton
-
Stan Hoeppner
-
Steffen Kaiser
-
Timo Sirainen