Re: [Dovecot] Can anyone tell me quick, if the howto on dovecot.org is actual:
Marko Weber said the following on 2/1/12 7:11 AM: Is this Howto http://wiki2.dovecot.org/HowTo/DovecotLDAPostfixAdminMySQL actual also for Dovecot 2.x ? Cause i read Dovecot 1.x in the Howto Thank you for any hints / tipps My gut feeling is version 1.2 because it was last modified on 2010-06-29 11:37:39(see bottom of page) which is in the 1.2 days(http://www.dovecot.org/oldnews.html). -Jonathan smime.p7s Description: S/MIME Cryptographic Signature
[Dovecot] [dovecot] Enable logging of all client commands in dovecot-1.2.alpha3
Hello, I would like to log all IMAP client commands sent to dovecot. The format would be time pid command arguments. I reviewed http://wiki.dovecot.org/Logging and started digging through dovecot-1.2.alpha3/src/master . I don't need this turned on all the time, just enough to see how clients do things and I don't need to see passwords. Any tips would be appreciated. -Jonathan smime.p7s Description: S/MIME Cryptographic Signature
[Dovecot] [dovecot] INDEX variable and mbox_snarf plugin
Is there a way to tell the dovecot mbox_snarf plugin to use an alternate location for the index/cache files? It doesn't seem to want to use the INDEX variable. I'm guessing the answer is no because of all the ties in lib-storage/index/mbox/*c to the directory where the file(that is the inbox) lives. The filesystem where I keep the inbox file doesn't have a directory for the user to own. The user only owns the mbox file. Thanks, Jonathan smime.p7s Description: S/MIME Cryptographic Signature
Re: [Dovecot] [dovecot] INDEX variable and mbox_snarf plugin
Since I didn't get any bytes for this question, I'll pose it a different way. From http://wiki.dovecot.org/MailLocation#indexfiles : == Index files Index files are by default stored under the same directory as mails. With maildir they are stored in the actual maildirs, with mbox they are stored under .imap/ directory. You may want to change the index file location if you're using NFS or if you're setting up shared mailboxes. You can change the index file location by adding :INDEX= to mail_location. For example: mail_location = maildir:~/Maildir:INDEX=/var/indexes/%u The index directories are created automatically, but note that it requires that Dovecot has actually access to create the directories. Either make sure that the index root directory (/var/indexes in the above example) is writable to the logged in user, or create the user's directory with proper permissions before the user logs in. If you really want to, you can also disable the index files completely by appending :INDEX=MEMORY. == How are people setting the INDEX for mbox_snarf if you are using NFS? I'm guessing I will just be rewriting lib-storage/index/mbox/ to have a choice of directories or creating 2 new files for mbox_snarf that look alot like the mbox driver index code. Was there a reason to not have a variable for index files for mbox_snarf? Thanks, Jonathan smime.p7s Description: S/MIME Cryptographic Signature
Re: [Dovecot] How to specify /var/spool/mail/m/i/mike
On Nov 10, 2008, at 12:25 PM, Network Operations wrote: I was wondering if somebody could tell me how I can tell dovecot (IMAP) to read users' mail boxes at /var/spool/mail/{1st char}/{2nd char}/username . For example: Mailbox for user 'mike' would be located at /var/spool/mail/m/i/mike 'karen' would be at /var/spool/mail/k/a/karen etc... Look at http://wiki.dovecot.org/Variables. I think the notation would be %1u/%1.1u/%u -Jonathan smime.p7s Description: S/MIME cryptographic signature
[Dovecot] [dovecot] Pre-populate index files for mbox
Is there a program written that can be used to build the index files for a given mbox file without using IMAP/POP? If not I'll be happy to donate it when I'm done. The args would be the userid , the full path to the mbox and where to put the index files. -Jonathan smime.p7s Description: S/MIME Cryptographic Signature
Re: [Dovecot] [dovecot] Pre-populate index files for mbox
Mark Zealey said the following on 11/25/08 11:31 AM: As a quick hack, surely you could deliver a dummy file to the inbox and then login over pop/imap and remove it? Mark Sure. This is for 85k users with about 50 folders each.. smime.p7s Description: S/MIME Cryptographic Signature
Re: [Dovecot] [dovecot] Pre-populate index files for mbox
Timo Sirainen said the following on 11/25/08 12:01 PM: On Nov 25, 2008, at 6:28 PM, Jonathan Siegle wrote: Is there a program written that can be used to build the index files for a given mbox file without using IMAP/POP? If not I'll be happy to donate it when I'm done. The args would be the userid , the full path to the mbox and where to put the index files. There's a difference between IMAP and POP3: With POP3 you really want to have message sizes cached. With IMAP you might not want to cache any fields, but depending on the client you might want to cache many different fields. That's not really easy to figure out beforehand (unless you just cache pretty much everything). Anyway assuming you want POP3 to be fast for most people, this'll do it: export MAIL=mbox:/tmp:INBOX=/path/to/mbox/file echo "quit" | /usr/local/libexec/dovecot/pop3 I'm actually doing this step for our Webmail application. It has its own set of toc files and also has native access to the filesystems. I want to migrate to the toc files in dovecot instead. I still don't think it would be a bad idea to create a mailutil type application for dovecot. Thanks, Jonathan smime.p7s Description: S/MIME Cryptographic Signature
[Dovecot] Imap logging and inetd
When I ran /usr/local/sbin/dovecot, the variable login_log_format_elements from dovecot.conf was honored. Now when I run /usr/local/libexec/dovecot/imap-login from /etc/inetd.conf, it isn't. How do I get imap-login to write log lines that use login_log_format_elements for the format? This is dovecot-1.2.alpha3 soon to be alpha4. Thanks, Jonathan smime.p7s Description: S/MIME Cryptographic Signature
Re: [Dovecot] Imap logging and inetd
Jonathan Siegle said the following on 12/2/08 7:16 AM: When I ran /usr/local/sbin/dovecot, the variable login_log_format_elements from dovecot.conf was honored. Now when I run /usr/local/libexec/dovecot/imap-login from /etc/inetd.conf, it isn't. How do I get imap-login to write log lines that use login_log_format_elements for the format? This is dovecot-1.2.alpha3 soon to be alpha4. Thanks, Jonathan I found where to get my desired result in src/master/mail-process.c. -Jonathan smime.p7s Description: S/MIME Cryptographic Signature
[Dovecot] Roadmap
I've been running 1.2alpha3 for weeks now without issue. However, some people in my testing department object to even testing "alpha" code. Are there specific features/bugs that need resolved before alpha reaches beta? Thanks, Jonathan smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] Imap logging and inetd
On Dec 13, 2008, at 10:54 PM, Timo Sirainen wrote: On Tue, 2008-12-02 at 13:48 -0500, Jonathan Siegle wrote: Jonathan Siegle said the following on 12/2/08 7:16 AM: When I ran /usr/local/sbin/dovecot, the variable login_log_format_elements from dovecot.conf was honored. Now when I run /usr/local/libexec/dovecot/imap-login from /etc/inetd.conf, it isn't. How do I get imap-login to write log lines that use login_log_format_elements for the format? This is dovecot-1.2.alpha3 soon to be alpha4. Thanks, Jonathan I found where to get my desired result in src/master/mail-process.c. Huh? mail-process.c only affects post-login imap/pop3. What did you change? The goal was to print a line with the user and the IP address with a date/time stamp. So I send those to syslog from here. Is that bad information? I haven't gotten bad info from it yet.. -Jonathan smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] SSL cert problems.
On Dec 29, 2008, at 2:31 PM, Geoff Sweet wrote: So my conf looks similar to yours: # Disable SSL/TLS support. #ssl_disable = no ssl_cert_file = /etc/pki/dovecot/certs/pop.x10.com.cer ssl_key_file = /etc/pki/dovecot/private/pop.x10.com.key # If key file is password protected, give the password here. Alternatively # give it when starting dovecot with -p parameter. #ssl_key_password = # File containing trusted SSL certificate authorities. Usually not needed. # The CAfile should contain the CA-certificate(s) followed by the matching # CRL(s). CRL checking is new in dovecot .rc1 ssl_ca_file = /etc/pki/verisign/intermediate_ca.cer Reading the openssl book on page 120(chapter 5) it says that you should have the whole chain in one file. I see that if you are using the SSL_CTX_use_certificate_chain_file function(as dovecot1.2alpha4 ./login-common/ssl-proxy-openssl.c does), you just need to put the whole chain in one file with the intermediate SECOND and your certificate FIRST. The book also claims that you should put the root certificate in here. I have seen conflicting documentation on putting the root cert in here because as another poster mentioned , you will never send it out. I may have missed a post that had my info above so sorry if I'm giving already provided information. -Jonathan # Request client to send a certificate. #ssl_verify_client_cert = no and the ssl_ca_file is a copy and past from this: http://www.verisign.com/support/verisign-intermediate-ca/extended-validation/index.html Yet the cert still doesn't work. And the OpenSSL people are telling me this is an issue with my application, dovecot. For reference this is all that is in my /etc/pki/verisign/intermediate_ca.cer: -BEGIN CERTIFICATE- MIIFEzCCBHygAwIBAgIQV7/7A/ssRtThns7g10N/EzANBgkqhkiG9w0BAQUFADBf MQswCQYDVQQGEwJVUzEXMBUGA1UEChMOVmVyaVNpZ24sIEluYy4xNzA1BgNVBAsT LkNsYXNzIDMgUHVibGljIFByaW1hcnkgQ2VydGlmaWNhdGlvbiBBdXRob3JpdHkw HhcNMDYxMTA4MDAwMDAwWhcNMjExMTA3MjM1OTU5WjCByjELMAkGA1UEBhMCVVMx FzAVBgNVBAoTDlZlcmlTaWduLCBJbmMuMR8wHQYDVQQLExZWZXJpU2lnbiBUcnVz dCBOZXR3b3JrMTowOAYDVQQLEzEoYykgMjAwNiBWZXJpU2lnbiwgSW5jLiAtIEZv ciBhdXRob3JpemVkIHVzZSBvbmx5MUUwQwYDVQQDEzxWZXJpU2lnbiBDbGFzcyAz IFB1YmxpYyBQcmltYXJ5IENlcnRpZmljYXRpb24gQXV0aG9yaXR5IC0gRzUwggEi MA0GCSqGSIb3DQEBAQUAA4IBDwAwggEKAoIBAQCvJAgIKXo1nmAMqudLO07cfLw8 RRy7K+D+KQL5VwijZIUVJ/XxrcgxiV0i6CqqpkKzj/i5Vbext0uz/o9+B1fs70Pb ZmIVYc9gDaTY3vjgw2IIPVQT60nKWVSFJuUrjxuf6/WhkcIzSdhDY2pSS9KP6HBR TdGJaXvHcPaz3BJ023tdS1bTlr8Vd6Gw9KIl8q8ckmcY5fQGBO+QueQA5N06tRn/ Arr0PO7gi+s3i+z016zy9vA9r911kTMZHRxAy3QkGSGT2RT+rCpSx4/VBEnkjWNH iDxpg8v+R70rfk/Fla4OndTRQ8Bnc+MUCH7lP59zuDMKz10/NIeWiu5T6CUVAgMB AAGjggHeMIIB2jAPBgNVHRMBAf8EBTADAQH/MDEGA1UdHwQqMCgwJqAkoCKGIGh0 dHA6Ly9jcmwudmVyaXNpZ24uY29tL3BjYTMuY3JsMA4GA1UdDwEB/wQEAwIBBjBt BggrBgEFBQcBDARhMF+hXaBbMFkwVzBVFglpbWFnZS9naWYwITAfMAcGBSsOAwIa BBSP5dMahqyNjmvDz4Bq1EgYLHsZLjAlFiNodHRwOi8vbG9nby52ZXJpc2lnbi5j b20vdnNsb2dvLmdpZjA9BgNVHSAENjA0MDIGBFUdIAAwKjAoBggrBgEFBQcCARYc aHR0cHM6Ly93d3cudmVyaXNpZ24uY29tL2NwczAdBgNVHQ4EFgQUf9Nlp8Ld7Lvw MAnzQzn6Aq8zMTMwNAYDVR0lBC0wKwYJYIZIAYb4QgQBBgpghkgBhvhFAQgBBggr BgEFBQcDAQYIKwYBBQUHAwIwgYAGA1UdIwR5MHehY6RhMF8xCzAJBgNVBAYTAlVT MRcwFQYDVQQKEw5WZXJpU2lnbiwgSW5jLjE3MDUGA1UECxMuQ2xhc3MgMyBQdWJs aWMgUHJpbWFyeSBDZXJ0aWZpY2F0aW9uIEF1dGhvcml0eYIQcLrkHRDZKTS2OMp7 A8y6vzANBgkqhkiG9w0BAQUFAAOBgQCpe2YpMPfVtKaWEtDucvBYEWkVVV9B/9IS hBOk2QNm/6ngTMntjHKLtNdVOykVYMg8Ie9ELpM9xgsMjSQ/HvsBWnrdg2YU0cf9 MFNIUYWFE6hU4e52ookY05eJesb9s72UYVo6CM8Uk72T/Qmpe1bIALhEWOneW3e9 BxxsCzAwxw== -END CERTIFICATE- Like I said, just a copy and paste from the Verisign site. Any thoughts? -Geoff smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] New SSL certificate problem
On Jan 5, 2009, at 2:50 PM, Stewart Dean wrote: Although I was told by Digicert that the order of chained certs in / var/ssl/certs/dovecot.pem should make no difference, after I put our public cert first, followed by Digicert's intermediate cert, dovecot started up fine. Of course, there were so many things I looked into, it might have been something else I touched.. Stewart, I posted this answer last week in another thread(12/29/2008 Subject SSL cert problems.). Yes order seems to be important. I found this answer in the Openssl book on page 120. -Jonathan Stewart Dean wrote: Our DC has been using a Verisign certificate. Over the past year, we've been using a Digicert Wildcard Plus certificate for almost all of our machines, and I wanted to switched over our DC mailserver. I used the following command to generate the CSR and key: openssl req -new -newkey rsa:1024 -nodes -out star_bard_edu.csr - keyout star_bard_edu.key -subj "/C=US/ST=NY/L=ourtown/O=Bard College IT/OU=Bard College /CN=*.bard.edu" The resultant CSR verified and I submitted it to digicert and got back our cert, plus their intermediate and Trusted root certs. I killed the root instance of dovecot and waited for all the children to die I put together the intermediate cert (first) and our cert (second) into /usr/ssl/certs/dovecot.pem I put the key star_bard_edu.key in /var/ssl/private/dovecot.pem I restarted dovecot, but the imap login instances didn't appear, so I shifted back to the original combined cert file and key, restarted dovecot and it came up OK I check the syslog and saw these error messages: Jan 5 10:19:49 mercury mail:err|error dovecot: imap-login: Can't load private k ey file /var/ssl/private/dovecot.pem: error:0B080074:x509 certificate routines:X 509_check_private_key:key values mismatch Jan 5 10:19:49 mercury mail:err|error last message repeated 8 times Jan 5 10:19:49 mercury mail:err|error dovecot: child 4051108 (login) returned e rror 89 Jan 5 10:19:49 mercury mail:err|error dovecot: child 4231382 (login) returned e rror 89 I checked my key and it has the same time stamp as my CSR, so I didn't somehow get the wrong key. Both the old and new key are 600; if the old one works based on perms, the new one should too. Would some kind soul tell me what I'm missing? Or is there a problem using wild card certificate with DC? Is there an openssl command to verify the key. Or is it that the key is unencrypted? -- Once upon a time, the Internet was a friendly, neighbors- helping-neighbors small town, and no one locked their doors. Now it's like an apartment in Bed-Stuy: you need three heavy duty pick- proof locks, one of those braces that goes from the lock to the floor, and bars on the windows Stewart Dean, Unix System Admin, Bard College, New York 12504 sd...@bard.edu voice: 845-758-7475, fax: 845-758-7035 smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] [dovecot] INDEX variable and mbox_snarf plugin
On Jan 15, 2009, at 6:47 PM, Timo Sirainen wrote: Sorry for the late reply. On Thu, 2008-11-06 at 13:39 -0500, Jonathan Siegle wrote: Is there a way to tell the dovecot mbox_snarf plugin to use an alternate location for the index/cache files? It doesn't seem to want to use the INDEX variable. I'm guessing the answer is no because of all the ties in lib-storage/index/mbox/*c to the directory where the file(that is the inbox) lives. The filesystem where I keep the inbox file doesn't have a directory for the user to own. The user only owns the mbox file. It's not possible to specify the index location because of how the plugin has been implemented. The best you could easily do by modifying sources is to disable indexes for INBOX. Other than that it would require larger changes, perhaps a full rewrite. That's what I gleaned from the source. For now, I will go forward with adding another 300k files to the filesystem.. Thanks, Jonathan smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] I've moved to US
On Feb 6, 2009, at 9:49 PM, Timo Sirainen wrote: On Feb 6, 2009, at 8:29 PM, Ron Wilhoite wrote: Congratulations! Wow, Finland to Blacksburg. That could make for some interesting 'culture shock' posts. Actually I find Blacksburg to be very similar to Finland. I haven't really had any shocks. Just some small annoyances how some things are better/easier in Finland :) Wow! Welcome, my new neighbor to the south.. http://maps.google.com/maps?f=d&source=s_d&saddr=State+College,+PA&daddr=Blacksburg,+Va&hl=en&geocode=&mra=ls&sll=37.0625,-95.677068&sspn=53.167773,96.503906&ie=UTF8&z=7 smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] I've moved to US
On Feb 10, 2009, at 4:24 PM, Jerry wrote: On Tue, 10 Feb 2009 15:44:09 -0500 (EST) Kyle George wrote: On Mon, 9 Feb 2009, Timo Sirainen wrote: All bread tastes weird, I'm not sure why. You have to get "Arthur Avenue" bread. See if you can locate a local bakery. The regular 'commercial' bread is loaded with preservatives. Agreed wrt preservatives. I'm partial to Martin's Famous Whole Wheat Potato Bread (http://www.potatoroll.com/pages/products.asp) when I didn't have time to make my own bread for the week. It could be the water. But there are some positive things, like grocery stores being open 24h and when driving turning to right is allowed on red light :) Careful: there's no right-on-red in New York City (and there are no signs about it!). There are on the West Side Drive. I see them every morning when I have to drive into the city. They are usually located at the city borders. -- Jerry ges...@yahoo.com Technological progress has merely provided us with more efficient means for going backwards. Aldous Huxley smime.p7s Description: S/MIME cryptographic signature
[Dovecot] mbox snarf plugin + idle
I'm having a problem with mbox snarf not looking at /var/spool/mail/ when in idle mode thus never giving me a RECENT line even though there are new messages in /var/spool/mail/ . Here are the imap commands to reproduce the problem: 1 login userid password 2 select inbox 3 idle When I run "select inbox" it does see my messages in /var/spool/mail/ and moves them over fine. When I truss the process, I see it only running stat calls on my "mbox- snarf" file. To get new messages I issue DONE, CLOSE, and SELECT INBOX. I'm not sure why it is reporting alpha5 two lines below. I did an hg pull just the other day and see 1.2.beta1 in the output of hg tags. # /usr/ladmin2/sbin/dovecot -n # 1.2.alpha5: /usr/ladmin2/etc/dovecot.conf Warning: fd limit 2000 is lower than what Dovecot can use under full load (more than 4224). Either grow the limit or change login_max_processes_count and max_mail_processes settings # OS: AIX 3 0001112AD300 syslog_facility: local0 protocols: imap listen: *: ssl: no disable_plaintext_auth: no login_dir: /usr/ladmin2/var/run/dovecot/login login_executable: /usr/ladmin2/libexec/dovecot/imap-login login_greeting: Dovecot ready. login_processes_count: 30 max_mail_processes: 4096 mail_location: mbox:%h/new:INBOX=/var/spool/mail/14/%u mmap_disable: yes dotlock_use_excl: no mbox_write_locks: fcntl mbox_lazy_writes: no mail_plugins: mbox_snarf imap_client_workarounds: delay-newmail auth default: mechanisms: plain gssapi krb5_keytab: /etc/krb5/dovecot.keytab gssapi_hostname: $ALL verbose: yes debug: yes passdb: driver: pam userdb: driver: passwd plugin: mbox_snarf: /gpfs/inbox/14/%u Thanks, Jonathan smime.p7s Description: S/MIME cryptographic signature
[Dovecot] Fwd: mbox snarf plugin + idle
Ok. I went back and reread rfc 2177 and now I understand. Another bug in mail.app. I'll go tell Apple. " The IDLE command is sent from the client to the server when the client is ready to accept unsolicited mailbox update messages. The server requests a response to the IDLE command using the continuation ("+") response. The IDLE command remains active until the client responds to the continuation, and as long as an IDLE command is active, the server is now free to send untagged EXISTS, EXPUNGE, and other messages at any time." Mail.app doesnt respond to the "+" response. Begin forwarded message: From: Jonathan Siegle Date: February 12, 2009 7:55:58 AM EST To: Dovecot Mailing List Subject: [Dovecot] mbox snarf plugin + idle X-Psu-Spam-Hits: -102.599 I'm having a problem with mbox snarf not looking at /var/spool/mail/ when in idle mode thus never giving me a RECENT line even though there are new messages in /var/spool/mail/ . Here are the imap commands to reproduce the problem: 1 login userid password 2 select inbox 3 idle When I run "select inbox" it does see my messages in /var/spool/ mail/ and moves them over fine. When I truss the process, I see it only running stat calls on my "mbox-snarf" file. To get new messages I issue DONE, CLOSE, and SELECT INBOX. I'm not sure why it is reporting alpha5 two lines below. I did an hg pull just the other day and see 1.2.beta1 in the output of hg tags. # /usr/ladmin2/sbin/dovecot -n # 1.2.alpha5: /usr/ladmin2/etc/dovecot.conf Warning: fd limit 2000 is lower than what Dovecot can use under full load (more than 4224). Either grow the limit or change login_max_processes_count and max_mail_processes settings # OS: AIX 3 0001112AD300 syslog_facility: local0 protocols: imap listen: *: ssl: no disable_plaintext_auth: no login_dir: /usr/ladmin2/var/run/dovecot/login login_executable: /usr/ladmin2/libexec/dovecot/imap-login login_greeting: Dovecot ready. login_processes_count: 30 max_mail_processes: 4096 mail_location: mbox:%h/new:INBOX=/var/spool/mail/14/%u mmap_disable: yes dotlock_use_excl: no mbox_write_locks: fcntl mbox_lazy_writes: no mail_plugins: mbox_snarf imap_client_workarounds: delay-newmail auth default: mechanisms: plain gssapi krb5_keytab: /etc/krb5/dovecot.keytab gssapi_hostname: $ALL verbose: yes debug: yes passdb: driver: pam userdb: driver: passwd plugin: mbox_snarf: /gpfs/inbox/14/%u Thanks, Jonathan smime.p7s Description: S/MIME cryptographic signature
[Dovecot] dovecot1.2beta2( hg tags yields 8834:5284f45c249a) fetch error
Steps to reproduce 1 login testuser testpw 2 select inbox 3 fetch 1 body.peek[HEADER.FIELDS (date)] I get the error 3 BAD Error in IMAP command FETCH: Unknown FETCH modifier This is AIX 5.3 with mbox files. -Jonathan smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] dovecot1.2beta2( hg tags yields 8834:5284f45c249a) fetch error
On Mar 18, 2009, at 2:00 PM, Timo Sirainen wrote: On Wed, 2009-03-18 at 13:06 -0400, Jonathan Siegle wrote: Steps to reproduce 1 login testuser testpw 2 select inbox 3 fetch 1 body.peek[HEADER.FIELDS (date)] I get the error 3 BAD Error in IMAP command FETCH: Unknown FETCH modifier So it seems. Strange that no one had noticed it before. I thought Evolution did that, at least it used to.. Anyway, fixed: http://hg.dovecot.org/dovecot-1.2/rev/dc6880dcbbba I've noticed it for a little. But I didn't want to send noise. pine/ alpine use this when you go get a message that is postponed. I've just taken the time to learn howto fire up pine in debug and get that fetch statement out of the .pine-debug files.. Thanks this works now. 2 fetch 1 body.peek[HEADER.FIELDS (date)] * 1 FETCH (BODY[HEADER.FIELDS (DATE)] {41} Date: Wed, 18 Feb 2009 15:28:46 + smime.p7s Description: S/MIME cryptographic signature
[Dovecot] LIST command claims children exist in empty folder
Dovecot 1.2 (8834:5284f45c249a) Should list return \HasChildren if no folders exist under it? I'm using mbox format. 2 create testfolder/ 2 OK Create completed. 3 list "testfolder/" * * LIST (\Noselect \HasChildren) "/" "testfolder/" 3 OK List completed. 4 list "testfolder/" % * LIST (\Noselect \HasChildren) "/" "testfolder/" 4 OK List completed. 5 list "testfolder" % * LIST (\Noselect \HasChildren) "/" "testfolder" 5 OK List completed. 6 list "testfolder" * * LIST (\Noselect \HasChildren) "/" "testfolder" Thanks, Jonathan smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] LIST command claims children exist in empty folder
On Mar 24, 2009, at 11:36 AM, Jonathan Siegle wrote: Dovecot 1.2 (8834:5284f45c249a) Should list return \HasChildren if no folders exist under it? I'm using mbox format. 2 create testfolder/ 2 OK Create completed. 3 list "testfolder/" * * LIST (\Noselect \HasChildren) "/" "testfolder/" 3 OK List completed. 4 list "testfolder/" % * LIST (\Noselect \HasChildren) "/" "testfolder/" 4 OK List completed. 5 list "testfolder" % * LIST (\Noselect \HasChildren) "/" "testfolder" 5 OK List completed. 6 list "testfolder" * * LIST (\Noselect \HasChildren) "/" "testfolder" Thanks, Jonathan Here is my patch. diff -r d975ed910613 src/lib-storage/index/mbox/mbox-storage.c --- a/src/lib-storage/index/mbox/mbox-storage.c Wed Mar 25 07:34:55 2009 -0400 +++ b/src/lib-storage/index/mbox/mbox-storage.c Wed Mar 25 07:44:52 2009 -0400 @@ -888,7 +888,7 @@ path = t_strconcat(dir, "/", fname, NULL); if (stat(path, &st) == 0) { if (S_ISDIR(st.st_mode)) - *flags |= MAILBOX_NOSELECT | MAILBOX_CHILDREN; + *flags |= MAILBOX_NOSELECT ; else { *flags |= MAILBOX_NOINFERIORS | STAT_GET_MARKED(st); if (is_inbox_file(ctx->list, path, fname) && 3 list "testfolder/" * * LIST (\Noselect \HasChildren) "/" "testfolder/" 3 OK List completed. 4 list "testfolder" % * LIST (\Noselect \HasNoChildren) "/" "testfolder" 4 OK List completed. 5 list "testfolder" % * LIST (\Noselect \HasNoChildren) "/" "testfolder" 5 OK List completed. 6 list "testfolder" * * LIST (\Noselect \HasNoChildren) "/" "testfolder" 6 OK List completed. So slightly different behavior. Is this correct ? smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] LIST command claims children exist in empty folder
On Mar 25, 2009, at 8:42 PM, Timo Sirainen wrote: On Wed, 2009-03-25 at 07:52 -0400, Jonathan Siegle wrote: On Mar 24, 2009, at 11:36 AM, Jonathan Siegle wrote: Dovecot 1.2 (8834:5284f45c249a) Should list return \HasChildren if no folders exist under it? I'm using mbox format. Is this really a problem?.. alpine/pine can't delete empty folders because the empty folder \HasChildren. Here is my patch. Problem with that is that it doesn't return any children flags when using LISTEXT command: 1 list (subscribed) "" % return (children) Fixing this would require adding new code to fs_list_subs() to scan the subdirectory if children flags are missing. list_file_subdir() handles that for non-subscription listing, but it can't be directly used for subscription listing. ok thanks. I'll look at that today. smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] Dovecot v2.0 hg tree
On Apr 23, 2009, at 8:04 PM, Timo Sirainen wrote: http://hg.dovecot.org/dovecot-2.0/ I just did the initial commit for master process rewrite, which marks the beginning of Dovecot v2.0. Several things are still missing/ broken, but at least I was just able to successfully log in using imap :) I left v1.3 hg tree there for now, but once v2.0 tree is fully usable I'll just delete the v1.3 tree. Timo, Is there any reason to follow the 1.3 tree? Thanks, Jonathan smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] Timeout leak with dovecot version dovecot1.2(8985:f43bebab3dac)
On Apr 29, 2009, at 2:58 PM, Timo Sirainen wrote: On Wed, 2009-04-29 at 14:52 -0400, jsie...@psu.edu wrote: This is 64bit AIX 5.3. Looking through previous versions of dovecot, I see this warning. I didn't realize this was something bad until today. It's not exactly bad. It gets logged only when the process is exiting. But it shouldn't be happening either. local0.log.20090429:Apr 29 12:41:16 hostname dovecot: IMAP(jsiegle): Timeout leak: 1100054c0 How easily can you reproduce this? For example if you do: telnet localhost 143 1 login user pass 2 select inbox 3 logout Does it get logged? What if you select some other mailbox instead? Yes it gets logged. I did your steps and reproduced. I also did login;select;close;logout and that also gave me the error. smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] Timeout leak with dovecot version dovecot1.2(8985:f43bebab3dac)
On Apr 30, 2009, at 3:06 PM, Timo Sirainen wrote: On Thu, 2009-04-30 at 15:04 -0400, Jonathan Siegle wrote: telnet localhost 143 1 login user pass 2 select inbox 3 logout Does it get logged? What if you select some other mailbox instead? Yes it gets logged. I did your steps and reproduced. I also did login;select;close;logout and that also gave me the error. What about another mailbox than INBOX? If it happens only with INBOX, the problem is mbox-snarf plugin. It doesn't do it on folders. so login;select folder;logout doesn't produce the error. smime.p7s Description: S/MIME cryptographic signature
[Dovecot] mbox-snarf plugin and istream-mail-stats.c revision 9000:b02c642b4e51
I'm getting this error: May 1 09:09:30 tr27n12.aset.psu.edu syslog: PSU mbox-snarf name is INBOX May 1 09:09:30 tr27n12.aset.psu.edu dovecot: Panic: IMAP(tstem38): file istream-mail-stats.c: line 75: assertion failed: (ret != -1 || stream->istream.eof || stream->istream.stream_errno != 0) Will this error go away before the general 1.2 release? Thanks, Jonathan
Re: [Dovecot] mbox-snarf plugin and istream-mail-stats.c revision 9000:b02c642b4e51
On May 1, 2009, at 11:38 AM, Timo Sirainen wrote: On May 1, 2009, at 11:37 AM, Jonathan Siegle wrote: May 1 09:09:30 tr27n12.aset.psu.edu syslog: PSU mbox-snarf name is INBOX May 1 09:09:30 tr27n12.aset.psu.edu dovecot: Panic: IMAP(tstem38): file istream-mail-stats.c: line 75: assertion failed: (ret != -1 || stream->istream.eof || stream->istream.stream_errno != 0) Will this error go away before the general 1.2 release? Hopefully.. You can reproduce this always by selecting INBOX? Yeah. This is on my development server. I can't get into my inbox anymore.
Re: [Dovecot] mbox-snarf plugin and istream-mail-stats.c revision 9000:b02c642b4e51
On May 1, 2009, at 2:50 PM, Timo Sirainen wrote: On Fri, 2009-05-01 at 11:37 -0400, Jonathan Siegle wrote: I'm getting this error: May 1 09:09:30 tr27n12.aset.psu.edu syslog: PSU mbox-snarf name is INBOX May 1 09:09:30 tr27n12.aset.psu.edu dovecot: Panic: IMAP(tstem38): file istream-mail-stats.c: line 75: assertion failed: (ret != -1 || stream->istream.eof || stream->istream.stream_errno != 0) http://hg.dovecot.org/dovecot-1.2/rev/06bd1266f0c7 Thank you so much for the two fixes(this + : http://hg.dovecot.org/dovecot-1.2/rev/66b6cd495702) . I've done basic testing with no errors. Have a good weekend! smime.p7s Description: S/MIME cryptographic signature
[Dovecot] dovecot 2.0 (revision 9271:d467712aee77) compile problems on AIX 5.3
Having some problems compiling on AIX 5.3 with IBM vac version 8. Programs that I had problems building: test-mail test-imap test-index To fix the undefined symbols problem: ld: 0711-317 ERROR: Undefined symbol: .charset_to_utf8_end ld: 0711-317 ERROR: Undefined symbol: .charset_to_utf8_begin ld: 0711-317 ERROR: Undefined symbol: .charset_to_utf8 ld: 0711-317 ERROR: Undefined symbol: .charset_is_utf8 ld: 0711-317 ERROR: Undefined symbol: .charset_to_utf8_str ld: 0711-317 ERROR: Undefined symbol: .message_header_decode_utf8 ld: 0711-317 ERROR: Undefined symbol: .rfc822_parser_init ld: 0711-317 ERROR: Undefined symbol: .rfc822_skip_lwsp ld: 0711-317 ERROR: Undefined symbol: .rfc822_parse_content_type ld: 0711-317 ERROR: Undefined symbol: .rfc2231_parse ld: 0711-317 ERROR: Undefined symbol: .rfc822_parse_mime_token ld: 0711-317 ERROR: Undefined symbol: .rfc822_parse_atom ld: 0711-317 ERROR: Undefined symbol: .message_address_parse ld: 0711-317 ERROR: Undefined symbol: .iconv ld: 0711-317 ERROR: Undefined symbol: .iconv_open ld: 0711-317 ERROR: Undefined symbol: .iconv_close I add this ../lib-charset/.libs/libcharset.a -liconv to the Makefiles (./src/lib-imap/Makefile;./src/lib-index/Makefile;./src/lib-mail/ Makefile) below: $ find . -name Makefile -exec egrep -p libcharset.a {} \; -print clean-noinstPROGRAMS: @list='$(noinst_PROGRAMS)'; for p in $$list; do \ f=`echo $$p|sed 's/$(EXEEXT)$$//'`; \ echo " rm -f $$p $$f"; \ rm -f $$p $$f ; \ done test-imap$(EXEEXT): $(test_imap_OBJECTS) $(test_imap_DEPENDENCIES) @rm -f test-imap$(EXEEXT) $(LINK) $(test_imap_LDFLAGS) $(test_imap_OBJECTS) $ (test_imap_LDADD) $(LIBS) \ ../lib-mail/.libs/libmail.a ../lib-charset/.libs/libcharset.a -liconv ./src/lib-imap/Makefile clean-noinstPROGRAMS: @list='$(noinst_PROGRAMS)'; for p in $$list; do \ f=`echo $$p|sed 's/$(EXEEXT)$$//'`; \ echo " rm -f $$p $$f"; \ rm -f $$p $$f ; \ done test-index$(EXEEXT): $(test_index_OBJECTS) $(test_index_DEPENDENCIES) @rm -f test-index$(EXEEXT) $(LINK) $(test_index_LDFLAGS) $(test_index_OBJECTS) $ (test_index_LDADD) $(LIBS) ../lib-charset/.libs/libcharset.a -liconv ./src/lib-index/Makefile clean-noinstPROGRAMS: @list='$(noinst_PROGRAMS)'; for p in $$list; do \ f=`echo $$p|sed 's/$(EXEEXT)$$//'`; \ echo " rm -f $$p $$f"; \ rm -f $$p $$f ; \ done test-mail$(EXEEXT): $(test_mail_OBJECTS) $(test_mail_DEPENDENCIES) @rm -f test-mail$(EXEEXT) $(LINK) $(test_mail_LDFLAGS) $(test_mail_OBJECTS) $ (test_mail_LDADD) $(LIBS) ../lib-charset/.libs/libcharset.a -liconv ./src/lib-mail/Makefile Then compilation occurs fine. Thanks! Jonathan smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] Possibly dumb questions about DC and user/system limits
On May 14, 2009, at 2:17 PM, Stewart Dean wrote: Warning: fd limit 2000 is lower than what Dovecot can use under full load (more than 2054). Either grow the limit or change login_max_processes_count and max_mail_processes settings So I changed the no_size and no_size_hard to 3500 and 4000 respectively in both dovecot and root AIX defines nosize:* *Sets the soft limit for the number of file descriptors a user process may have open at one time. an lsuser dovecot returns: dovecot id=417 pgrp=dovecot groups=dovecot shell=/bin/false daemon=true admin=false ... fsize=2097151 cpu=-1 data=262144 stack=65536 core=2097151 rss=65536 nofiles=3500 nofiles_hard=4000 I kill dovecot and all children and restart itsame error message What am I missing? Does the machine have to be rebooted for the no_size to be updated? AIX should not have to be rebooted for this to happen. For debug, I would change the shell to a real shell; change the fsize to 600 or something small; logout and then login(as root and then su or just as the user) and try to make large files and verify limits. Ping me off list for more debugging unless people consider this on topic. -Jonathan smime.p7s Description: S/MIME cryptographic signature
[Dovecot] namespace list not working in dovecot 1.2 revision 9027:421393827a81
In my config that has worked for some time I have: namespace private { separator = / prefix = ~/ hidden = yes list = no # for v1.1+ } 1 list "~/" * 1 OK List completed. 2 list "~/" % 2 OK List completed. I _think_ that the last time I tested this was 2 weeks ago but it might have been 1 week ago. Thanks, Jonathan smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] Possibly dumb questions about DC and user/system limits
On May 14, 2009, at 2:17 PM, Stewart Dean wrote: The only changes was that max_mail_processes went from 1024 to 1280. Now I get a error message when I start DC: Warning: fd limit 2000 is lower than what Dovecot can use under full load (more than 2054). Either grow the limit or change login_max_processes_count and max_mail_processes settings So I changed the no_size and no_size_hard to 3500 and 4000 respectively in both dovecot and root AIX defines nosize:* *Sets the soft limit for the number of file descriptors a user process may have open at one time. I found a way to recreate the problem on my side. Since I start dovecot from inetd, I must verify my shell has the proper ulimits then do stopsrc -s inetd;startsrc -s inetd to pick up the new ulimits. I didn't notice how you started dovecot. Maybe this helps. smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] dovecot 2.0 (revision 9271:d467712aee77) compile problems on AIX 5.3
On May 17, 2009, at 2:42 PM, Timo Sirainen wrote: On Wed, 2009-05-13 at 15:04 -0400, Jonathan Siegle wrote: Having some problems compiling on AIX 5.3 with IBM vac version 8. Programs that I had problems building: test-mail test-imap test-index Should be fixed in hg now? Yes fixed. Is it to "early" to be reporting stuff like this? I now have this error(rev 9321:4c4b95def1fa): "ssl-proxy.c", line 12.5: 1506-343 (S) Redeclaration of ssl_proxy_new differs from previous declaration on line 17 of "ssl-proxy.h". "ssl-proxy.c", line 12.5: 1506-376 (I) Redeclaration of ssl_proxy_new has a different number of fixed parameters than the previous declaration. "ssl-proxy.c", line 12.5: 1506-377 (I) The type "struct ssl_proxy**" of parameter 3 differs from the previous type "const struct login_settings*". "ssl-proxy.c", line 19.5: 1506-343 (S) Redeclaration of ssl_proxy_client_new differs from previous declaration on line 19 of "ssl-proxy.h". "ssl-proxy.c", line 19.5: 1506-376 (I) Redeclaration of ssl_proxy_client_new has a different number of fixed parameters than the previous declaration. "ssl-proxy.c", line 19.5: 1506-377 (I) The type "int(*)(void*)" of parameter 3 differs from the previous type "const struct login_settings*". gmake[3]: *** [ssl-proxy.lo] Error 1 gmake[3]: Leaving directory `/usr/sadmin/src/imapservers/dovecothg/ dovecot-2.0psu/src/login-common' thanks, Jonathan smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] dovecot 2.0 (revision 9271:d467712aee77) compile problems on AIX 5.3
On May 18, 2009, at 2:02 PM, Timo Sirainen wrote: On Mon, 2009-05-18 at 13:57 -0400, Jonathan Siegle wrote: Yes fixed. Is it to "early" to be reporting stuff like this? No, it's not too early. Better early than late :) I now have this error(rev 9321:4c4b95def1fa): "ssl-proxy.c", line 12.5: 1506-343 (S) Redeclaration of ssl_proxy_new differs from previous declaration on line 17 of "ssl-proxy.h". Fixed: http://hg.dovecot.org/dovecot-2.0/rev/9c6597ba9e3e Thanks. No more compilation errors to report. smime.p7s Description: S/MIME cryptographic signature
[Dovecot] Problem with pam/krb5 auth on AIX 5.3
I'm using pam to authenticate users against my krb5 realm. Here is the problem scenario: User test2 attempts to login and their password is not expired so dovecot says: 0 login test2 myfavoritepassword 0 OK [CAPABILITY IMAP4rev1 LITERAL+ SASL-IR LOGIN-REFERRALS ID ENABLE SORT THREAD=REFERENCES THREAD=REFS MULTIAPPEND UNSELECT IDLE CHILDREN NAMESPACE UIDPLUS LIST-EXTENDED I18NLEVEL=1 CONDSTORE QRESYNC ESEARCH ESORT SEARCHRES WITHIN CONTEXT=SEARCH] Logged in 1 logout * BYE Logging out 1 OK Logout completed. User test1 attempts to login, but their password is expired. So dovecot says: 0 login test1 myfavoritepassword 0 NO d expired User test2 attempts to login and their password is not expired. But dovecot still says: 0 login test2 myfavoritepassword 0 NO d expired If I kill the pid with name "dovecot-auth -w", user test2 can login just fine unless I login with the user test1 before trying user test2. So it seems like something is getting cached. I'm running imap-login out of inetd, in case that matters. In my dovecot.conf, I don't have any caching/authentication variables activated. I don't see anything obvious to type in passdb pam{ } to type. For debug, I've enable pam for telnet and tested that without error. Also, the logs show that test2 This is dovecot revision 9062:694714d59cd9 . Looking at the logs, I see user test2 authenticate correctly in all instances. thanks, Jonathan smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] Problem with pam/krb5 auth on AIX 5.3
On May 20, 2009, at 1:38 PM, Timo Sirainen wrote: On Wed, 2009-05-20 at 13:22 -0400, Jonathan Siegle wrote: I'm using pam to authenticate users against my krb5 realm. Here is the problem scenario: I guess pam_krb5 doesn't like it if the same process tries to authenticate multiple times. Use passdb pam { args = max_requests=1 } Ah yes that is the flag I need. Thanks! Jonathan smime.p7s Description: S/MIME cryptographic signature
[Dovecot] uid fetch error with revision 9112:9d634c93d28a
This command fails: 2 uid fetch somevaliduid (BODYSTRUCTURE BODY.PEEK[]) 2 BAD Error in IMAP command UID: Unknown command BODYSTRUCTURE Thanks, Jonathan smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] uid fetch error with revision 9112:9d634c93d28a
On Jun 1, 2009, at 12:27 PM, Timo Sirainen wrote: On Mon, 2009-06-01 at 11:03 -0400, Jonathan Siegle wrote: This command fails: 2 uid fetch somevaliduid (BODYSTRUCTURE BODY.PEEK[]) 2 BAD Error in IMAP command UID: Unknown command BODYSTRUCTURE http://hg.dovecot.org/dovecot-1.2/rev/f1a6c9dd4c33 ? Nope. I can tell you that it definitely works in 1.2rc3 from May 4th and I think it _might_ have worked as of early last week. This is using mbox storage. The client doing this is OSX Mail.app version 3.6(935/935.3). Thanks, Jonathan smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] uid fetch error with revision 9112:9d634c93d28a
On Jun 1, 2009, at 12:59 PM, Jonathan Siegle wrote: On Jun 1, 2009, at 12:27 PM, Timo Sirainen wrote: On Mon, 2009-06-01 at 11:03 -0400, Jonathan Siegle wrote: This command fails: 2 uid fetch somevaliduid (BODYSTRUCTURE BODY.PEEK[]) 2 BAD Error in IMAP command UID: Unknown command BODYSTRUCTURE http://hg.dovecot.org/dovecot-1.2/rev/f1a6c9dd4c33 ? Nope. I can tell you that it definitely works in 1.2rc3 from May 4th and I think it _might_ have worked as of early last week. This is using mbox storage. The client doing this is OSX Mail.app version 3.6(935/935.3). Thanks, Jonathan I can also tell you that it worked on the May 14th revision. smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] uid fetch error with revision 9112:9d634c93d28a
On Jun 1, 2009, at 1:59 PM, Timo Sirainen wrote: On Mon, 2009-06-01 at 12:59 -0400, Jonathan Siegle wrote: On Jun 1, 2009, at 12:27 PM, Timo Sirainen wrote: On Mon, 2009-06-01 at 11:03 -0400, Jonathan Siegle wrote: This command fails: 2 uid fetch somevaliduid 5 uid fetch 4 (BODYSTRUCTURE BODY.PEEK[]) 5 BAD Error in IMAP command UID: Unknown command BODYSTRUCTURE 2 BAD Error in IMAP command UID: Unknown command BODYSTRUCTURE http://hg.dovecot.org/dovecot-1.2/rev/f1a6c9dd4c33 ? Nope. I can tell you that it definitely works in 1.2rc3 from May 4th and I think it _might_ have worked as of early last week. This is using mbox storage. The client doing this is OSX Mail.app version 3.6(935/935.3). I can't seem to be able to reproduce this. What plugins do you have loaded? Can you manually try a few commands? telnet localhost 143 1 login user pass 2 select inbox 3 fetch 1:* flags 4 uid fetch 1:* flags 5 fetch 1 bodystructure Sure. The problem for me is somewhere between revision 9061(works) and 9098(doesn't work). I don't select inbox. I select a folder to try to take plugins out of the picture. I'll keep doing my binary search to find the changeset that breaks.. 2 select foo4 * FLAGS (\Answered \Flagged \Deleted \Seen \Draft $NotJunk NonJunk) * OK [PERMANENTFLAGS (\Answered \Flagged \Deleted \Seen \Draft $NotJunk NonJunk \*)] Flags permitted. * 10 EXISTS * 0 RECENT * OK [UIDVALIDITY 1] UIDs valid * OK [UIDNEXT 11] Predicted next UID * OK [HIGHESTMODSEQ 1] 2 OK [READ-WRITE] Select completed. 3 fetch 1:* flags * 1 FETCH (FLAGS (\Seen $NotJunk)) * 2 FETCH (FLAGS (\Seen $NotJunk)) * 3 FETCH (FLAGS (\Seen $NotJunk NonJunk)) * 4 FETCH (FLAGS (\Seen $NotJunk NonJunk)) * 5 FETCH (FLAGS (\Seen $NotJunk NonJunk)) * 6 FETCH (FLAGS (\Seen $NotJunk NonJunk)) * 7 FETCH (FLAGS (\Seen $NotJunk)) * 8 FETCH (FLAGS (\Seen $NotJunk)) * 9 FETCH (FLAGS (\Seen $NotJunk)) * 10 FETCH (FLAGS (\Seen $NotJunk)) 3 OK Fetch completed. 4 fetch 1 bodystructure 4 BAD Error in IMAP command FETCH: Unknown command BODYSTRUCTURE 4 uid fetch 1:* flags * 1 FETCH (UID 1 FLAGS (\Seen $NotJunk)) * 2 FETCH (UID 2 FLAGS (\Seen $NotJunk)) * 3 FETCH (UID 3 FLAGS (\Seen $NotJunk NonJunk)) * 4 FETCH (UID 4 FLAGS (\Seen $NotJunk NonJunk)) * 5 FETCH (UID 5 FLAGS (\Seen $NotJunk NonJunk)) * 6 FETCH (UID 6 FLAGS (\Seen $NotJunk NonJunk)) * 7 FETCH (UID 7 FLAGS (\Seen $NotJunk)) * 8 FETCH (UID 8 FLAGS (\Seen $NotJunk)) * 9 FETCH (UID 9 FLAGS (\Seen $NotJunk)) * 10 FETCH (UID 10 FLAGS (\Seen $NotJunk)) 4 OK Fetch completed. 5 uid fetch 4 (BODYSTRUCTURE BODY.PEEK[]) 5 BAD Error in IMAP command UID: Unknown command BODYSTRUCTURE ./dovecot -n # 1.2.rc4: /usr/ladmin3/etc/dovecot.conf # OS: AIX 3 0001112AD300 syslog_facility: local0 protocols: imap listen: *:someport ssl: no disable_plaintext_auth: no login_dir: /usr/ladmin3/var/run/dovecot/login login_executable: /usr/ladmin3/libexec/dovecot/imap-login login_greeting: Dovecot baseline ready. login_processes_count: 30 max_mail_processes: 4096 mail_location: mbox:%h mmap_disable: yes dotlock_use_excl: no mbox_write_locks: fcntl mail_plugins: mbox_snarf mail_plugin_dir: /usr/ladmin3/lib/dovecot/imap/ imap_client_workarounds: tb-extra-mailbox-sep imap_id_log: * namespace: type: private separator: / inbox: yes list: yes subscriptions: yes namespace: type: private separator: / prefix: ~/ hidden: yes list: no subscriptions: yes auth default: krb5_keytab: /etc/myfavoritekeytab verbose: yes debug: yes passdb: driver: pam userdb: driver: passwd plugin: mbox_snarf: %h/SNARF smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] uid fetch error with revision 9112:9d634c93d28a
On Jun 1, 2009, at 2:37 PM, Timo Sirainen wrote: On Mon, 2009-06-01 at 14:31 -0400, Timo Sirainen wrote: On Mon, 2009-06-01 at 14:20 -0400, Jonathan Siegle wrote: 2 uid fetch somevaliduid 5 uid fetch 4 (BODYSTRUCTURE BODY.PEEK[]) 5 BAD Error in IMAP command UID: Unknown command BODYSTRUCTURE This should fix it: http://hg.dovecot.org/dovecot-1.2/rev/ 4d2b2adfd415 And http://hg.dovecot.org/dovecot-1.2/rev/9ae55b68cf61 3 fetch 1 bodystructure * 1 FETCH (BODYSTRUCTURE ("text" "plain" ("charset" "iso-8859-1") NIL NIL "7bit" 555 13 NIL ("inline" NIL) NIL NIL)) 3 OK Fetch completed. 4 uid fetch 1 (BODYSTRUCTURE BODY.PEEK[]) * 1 FETCH (UID 1 BODYSTRUCTURE ("text" "plain" ("charset" "iso-8859-1") NIL NIL "7bit" 555 13 NIL ("inline" NIL) NIL NIL) BODY[] {3953} . . . . 4 OK Fetch completed. And there was much cheering in central Pennsylvania... thanks, Jonathan smime.p7s Description: S/MIME cryptographic signature
[Dovecot] record points outside file error with dovecot revision 9116:9ae55b68cf61
I use mbox with mbox-snarf plugin. I get errors like: dovecot: IMAP(testuser): Corrupted index cache file /full/path/ dovecot.index.cache: record points outside file The errors can be in either folders or INBOX. I put some syslog statements in src/lib-index/mail-cache-lookup.c to help understand this. if (offset + sizeof(*rec) > cache->mmap_length) { mail_cache_set_corrupted(cache, "record points outside file"); syslog(LOG_DEBUG,"mail_cache_get_record rec->size is %d ", rec->size); syslog(LOG_DEBUG,"cache->mmap_length is %d ", cache- >mmap_length); syslog(LOG_DEBUG,"offset is %d ", offset); return -1; } syslog: mail_cache_get_record rec->size is 268595472 syslog: cache->mmap_length is 4096 syslog: offset is 1630760037 Oh I forgot to print off sizeof(*rec). Would that help? Thanks, Jonathan smime.p7s Description: S/MIME cryptographic signature
[Dovecot] Panic with signal 6 core dump with revision 9116:9ae55b68cf61
Jun 2 10:05:14 hostname dovecot: IMAP(testuser): Panic: file istream- raw-mbox.c: line 380: assertion failed: (new_pos > 0) Jun 2 10:05:14 hostname dovecot: dovecot: child 544822 (imap) killed with signal 6 (dbx) where raise(??) at 0x905a68c abort() at 0x9085c2c default_fatal_finish(type = LOG_TYPE_PANIC, status = 0), line 160 in "failures.c" i_internal_fatal_handler(type = LOG_TYPE_PANIC, status = 0, fmt = "file %s: line %d: assertion failed: (%s)", args = ""), line 440 in "failures.c" i_panic(format = "file %s: line %d: assertion failed: (%s)", ... = 0x1001731b4, 0x17c, 0x100173328, 0x0, 0x15b5, 0x0, 0x0), line 207 in "failures.c" i_stream_raw_mbox_read(stream = 0x0001100ca6b0), line 380 in "istream-raw-mbox.c" i_stream_raw_mbox_read(stream = 0x0001100ca6b0), line 379 in "istream-raw-mbox.c" i_stream_read(stream = 0x0001100ca700), line 80 in "istream.c" i_stream_limit_read(stream = 0x0001100cb930), line 64 in "istream- limit.c" i_stream_read(stream = 0x0001100cb980), line 80 in "istream.c" i_stream_read_copy_from_parent(istream = 0x0001100cbb20), line 118 in "istream.c" i_stream_header_filter_read(stream = 0x0001100cbad0), line 315 in "istream-header-filter.c" i_stream_read(stream = 0x0001100cbb20), line 80 in "istream.c" i_stream_read_copy_from_parent(istream = 0x0001100cbd20), line 118 in "istream.c" i_stream_mail_stats_read_mail_stats(stream = 0x0001100cbcd0), line 47 in "istream-mail-stats.c" i_stream_read(stream = 0x0001100cbd20), line 80 in "istream.c" i_stream_read_data(stream = 0x0001100cbd20, data_r = 0x0fffecc0, size_r = 0x0fffecc8, threshold = 1), line 361 in "istream.c" message_parser_read_more(ctx = 0x0001100cc118, block_r = 0x0fffecb0, full_r = 0x0fffeb84), line 118 in "message- parser.c" parse_next_body_to_boundary(ctx = 0x0001100cc118, block_r = 0x0fffecb0), line 330 in "message-parser.c" message_parser_parse_next_block(ctx = 0x0001100cc118, block_r = 0x0fffecb0), line 768 in "message-parser.c" message_parser_parse_body(ctx = 0x0001100cc118, hdr_callback = (nil), context = (nil)), line 831 in "message-parser.c" index_mail_parse_body(mail = 0x0001100c9878, field = MAIL_CACHE_FLAGS), line 792 in "index-mail.c" index_mail_get_parts(_mail = 0x0001100c9878, parts_r = 0x0fffef28), line 224 in "index-mail.c" mail_get_parts(mail = 0x0001100c9878, parts_r = 0x0fffef28), line 71 in "mail.c" unnamed block in search_arg_match_text(args = 0x0001100c7e70, ctx = 0x0001100c95f0, ret = -1), line 647 in "index-search.c" search_arg_match_text(args = 0x0001100c7e70, ctx = 0x0001100c95f0, ret = -1), line 647 in "index-search.c" search_match_next(ctx = 0x0001100c95f0), line 1101 in "index- search.c" unnamed block in index_storage_search_next_nonblock(_ctx = 0x0001100c95f0, mail = 0x0001100c9878, tryagain_r = 0x01e0), line 1301 in "index-search.c" index_storage_search_next_nonblock(_ctx = 0x0001100c95f0, mail = 0x0001100c9878, tryagain_r = 0x01e0), line 1301 in "index-search.c" mailbox_search_next_nonblock(ctx = 0x0001100c95f0, mail = 0x0001100c9878, tryagain_r = 0x01e0), line 754 in "mail-storage.c" cmd_search_more(cmd = 0x000110022058), line 347 in "imap-search.c" cmd_search_more_callback(cmd = 0x000110022058), line 434 in "imap- search.c" unnamed block in io_loop_handle_timeouts_real(ioloop = 0x00011001f310), line 316 in "ioloop.c" io_loop_handle_timeouts_real(ioloop = 0x00011001f310), line 316 in "ioloop.c" unnamed block in io_loop_handle_timeouts(ioloop = 0x00011001f310), line 327 in "ioloop.c" io_loop_handle_timeouts(ioloop = 0x00011001f310), line 327 in "ioloop.c" io_loop_handler_run(ioloop = 0x00011001f310), line 162 in "ioloop- poll.c" io_loop_run(ioloop = 0x00011001f310), line 338 in "ioloop.c" main(argc = 1, argv = 0x0630, envp = 0x0640), line 323 in "main.c" smime.p7s Description: S/MIME cryptographic signature
[Dovecot] Running imaptest revision 209:939fa886391a built against dovecot revision 9116:9ae55b68cf61 on AIX 5.3 core dumps
# ./imaptest host=127.0.0.1 port=143 user=tstem38 pass=pass4you mbox=/ gpfs/users/t/s/tstem38/IMAP/foo4 Panic: file client.c: line 620: assertion failed: (idx >= array_count(&clients) == NULL) IOT/Abort trap(coredump) smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] record points outside file error with dovecot revision 9116:9ae55b68cf61
On Jun 2, 2009, at 1:31 PM, Timo Sirainen wrote: On Tue, 2009-06-02 at 13:24 -0400, jsie...@psu.edu wrote: dovecot: IMAP(testuser): Corrupted index cache file /full/path/ dovecot.index.cache: record points outside file So you're using AIX? Do you also happen to use NFS? Can you reproduce this error by running imaptest for a while? http://imapwiki.org/ImapTest No NFS here. The filesystem is called GPFS. It is a clustered FS. Ah, that probably explains it. Can multiple different servers modify the same mailbox? Cache file is the part of Dovecot that demands the most from the OS/filesystem. The most difficult part is probably that it writes to the file without locking. It first reserves a space and then starts writing there. Multiple processes can write to the same file at the same time. Timo, Are you saying that multiple processes on the same folder(INBOX) on the same IMAP server can cause this collision as well? Is there a difference between running multiple processes on the same folder(INBOX) on multiple IMAP servers vs running multiple processes on the same folder on a single IMAP server? This should probably be changed at some point, since it could just buffer more data to memory and then lock, write, unlock. That would also make the code simpler, since it can currently leave holes to the file because it has to guess initially how much space to reserve.. This is definitely on my wish list. Thanks, Jonathan smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] record points outside file error with dovecot revision 9116:9ae55b68cf61
On Jun 3, 2009, at 9:35 AM, Timo Sirainen wrote: On Wed, 2009-06-03 at 09:31 -0400, Jonathan Siegle wrote: Are you saying that multiple processes on the same folder(INBOX) on the same IMAP server can cause this collision as well? Is there a difference between running multiple processes on the same folder(INBOX) on multiple IMAP servers vs running multiple processes on the same folder on a single IMAP server? I don't know. That depends on how GPFS is implemented. Pick a local filesystem, say ext3? smime.p7s Description: S/MIME cryptographic signature
[Dovecot] Dovecot 1.2 + AIX setups
Is anyone running Dovecot on AIX? I'm trying to debug the "Corrupted index cache file... record points outside file" problem. I can give gory details about what I've tried, but for now I would just like to see setups on AIX. Thanks, Jonathan smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] record points outside file error with dovecot revision 9116:9ae55b68cf61
On Jun 3, 2009, at 12:17 PM, Timo Sirainen wrote: On Wed, 2009-06-03 at 12:11 -0400, Timo Sirainen wrote: On Wed, 2009-06-03 at 10:14 -0400, Jonathan Siegle wrote: On Jun 3, 2009, at 9:35 AM, Timo Sirainen wrote: On Wed, 2009-06-03 at 09:31 -0400, Jonathan Siegle wrote: Are you saying that multiple processes on the same folder(INBOX) on the same IMAP server can cause this collision as well? Is there a difference between running multiple processes on the same folder(INBOX) on multiple IMAP servers vs running multiple processes on the same folder on a single IMAP server? I don't know. That depends on how GPFS is implemented. Pick a local filesystem, say ext3? But with ext3 you can't have multiple servers accessing the same filesystem. But of course there are no problems (well, some very rare random ones maybe) having multiple processes accessing the same mailbox on the same server. Ever since I wrote my imaptest tool (a few years ago?) I've been heavily stress testing multiple connections modifying the mailbox at the same time. Timo, Can you tell me what platform/filesystem you use for testing? Oh and can you hint at what "bad things" may happen when I get the error "record points outside file"? Thanks, Jonathan smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] minimize mbox mdbox fragmentation
On Oct 20, 2010, at 10:14 PM, Timo Sirainen wrote: > On 21.10.2010, at 3.12, Denny Lin wrote: > >> On Wed, Oct 20, 2010 at 06:45:17PM +0100, Timo Sirainen wrote: >>> On Wed, 2010-10-20 at 13:32 -0400, Charles Marcus wrote: On 2010-10-20 12:53 PM, Timo Sirainen wrote: > Oh, interesting. I didn't know that was possible. And even better: Linux > has fallocate() that can do it for other filesystems than just XFS. Or > looks like it's only XFS and ext4 (ext3 doesn't support it). How about reiserfs (3, not 4)? >>> >>> Doesn't support. >> >> Is it possible with UFS and ZFS? > > Linux doesn't support either and my googling didn't find any FreeBSD or > Solaris interface for this feature, so I don't know. > AIX supports posix_fallocate but only for JFS2. For GPFS(what we use), the function is gpfs_prealloc. smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] Question about mbox_snarf and dovecot2.0
On Nov 18, 2010, at 2:08 PM, Timo Sirainen wrote: > On Thu, 2010-04-29 at 11:11 -0400, Jonathan Siegle wrote: > >> As I need this to function, I've been thinking about your words above >> and been reading http://wiki.dovecot.org/Design/Storage/Plugins. The >> mentality of it reminds me of the lazy-expunge-plugin, but this API is >> taking some time getting used to. Any pointers would be appreciated. > > So I guess you never got around to implementing it? I finally did: > http://dovecot.org/list/dovecot/2010-November/055020.html > > I did do it. I've been testing it for a few months now. Sorry. Should have said something.
[Dovecot] dovecot 2.0 revision 12532:e030df616faf: problem with Snarf plugin
this command fails when the snarf plugin is enabled. 5 status inbox (UIDNEXT MESSAGES) with the error: Dec 10 13:57:57 tr27n12.aset.psu.edu dovecot: imap(tstem38): Panic: file index-transaction.c: line 71: assertion failed: (box->opened) If it isn't easy to reproduce, I'll spit out the config. thanks, Jonathan smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] dovecot 2.0 revision 12532:e030df616faf: problem with Snarf plugin
Oops! Config problem. Typing should be second nature by now huh. 4 status inbox (UIDNEXT MESSAGES) * STATUS "inbox" (MESSAGES 4716 UIDNEXT 101501) 4 OK Status completed. 5 status inbox (UIDNEXT MESSAGES) * STATUS "inbox" (MESSAGES 4717 UIDNEXT 101502) 5 OK Status completed. On Dec 10, 2010, at 2:04 PM, Jonathan Siegle wrote: > this command fails when the snarf plugin is enabled. > > 5 status inbox (UIDNEXT MESSAGES) > > > with the error: > > Dec 10 13:57:57 tr27n12.aset.psu.edu dovecot: imap(tstem38): Panic: file > index-transaction.c: line 71: assertion failed: (box->opened) > > If it isn't easy to reproduce, I'll spit out the config. > > thanks, > Jonathan > smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] dovecot 2.0 revision 12532:e030df616faf: problem with Snarf plugin
On Dec 13, 2010, at 8:06 AM, Timo Sirainen wrote: > On Fri, 2010-12-10 at 14:45 -0500, Jonathan Siegle wrote: >> Oops! Config problem. Typing should be second nature by now huh. > .. >>> Dec 10 13:57:57 tr27n12.aset.psu.edu dovecot: imap(tstem38): Panic: file >>> index-transaction.c: line 71: assertion failed: (box->opened) > > What kind of a config problem? It still shouldn't crash. > > Actually it wasn't a config problem. I'm now able to reproduce it again. I'm using mail.local from sendmail as my lda. This is on AIX 5.3 What other info do you need? smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] question about snarf plugin in dovecot 2
On Dec 14, 2010, at 4:46 PM, Tom Lieuallen wrote: > >> After these changes, snarfing doesn't work for me, either straight or UW >> optional style. Either way I try it, when I connect, it creates an empty >> ~/mbox file but does not snarf my inbox into it. My mail client >> (thunderbird) doesn't see any messages at all in my inbox. I remove the >> snarf namespace and change my mail_location back and my inbox is, of >> course, back. Note that I have all the UW compatibility namespaces in >> there. > Once I do that, I see that the plugin does load. And now it no longer > creates an ~/mbox if it doesn't first exist. Then again, it now panics, so > perhaps that is happening before getting to the point of creating the ~/mbox > file. Since the panic is happening just after reference to > .../var/run/dovecot/empty, I'm wondering if it has something to do with that > directory. I can tell you that my config looks a little different and snarfing works. I have no need for optional snarfing. My mail_location qualifies the full path to the file where I want the mail to go. So it looks like: mail_location = mbox:%h:INBOX=/gpfs/inbox/%Ju/%u:INDEX=%h/.dovecot2.0.2 Namespace wise, I have: namespace Snarf { prefix = ~~Snarfbox/ location = mbox:/var/empty:INBOX=/var/spool/mail/%Ju/%u:INDEX=MEMORY list = no hidden = yes } and finally: plugin { snarf = ~~Snarfbox/INBOX } I have not figured out my problem with using the IMAP status command. 1 status inbox (messages) -Jonathan smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] dovecot 2.0 revision 12532:e030df616faf: problem with Snarf plugin
On Dec 13, 2010, at 9:27 AM, Jonathan Siegle wrote: > > On Dec 13, 2010, at 8:06 AM, Timo Sirainen wrote: > >> On Fri, 2010-12-10 at 14:45 -0500, Jonathan Siegle wrote: >>> Oops! Config problem. Typing should be second nature by now huh. >> .. >>>> Dec 10 13:57:57 tr27n12.aset.psu.edu dovecot: imap(tstem38): Panic: file >>>> index-transaction.c: line 71: assertion failed: (box->opened) >> >> What kind of a config problem? It still shouldn't crash. >> >> > > Actually it wasn't a config problem. I'm now able to reproduce it again. I'm > using mail.local from sendmail as my lda. This is on AIX 5.3 What other info > do you need? > > Timo, If you give me some logic to throw at this, I'll take a shot at the programming. I tested my snarf code and it does the same thing. thanks, Jonathan smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] dovecot 2.0 revision 12532:e030df616faf: problem with Snarf plugin
On Dec 17, 2010, at 8:16 AM, Timo Sirainen wrote: > On Fri, 2010-12-17 at 07:45 -0500, Jonathan Siegle wrote: >>>>>> Dec 10 13:57:57 tr27n12.aset.psu.edu dovecot: imap(tstem38): Panic: file >>>>>> index-transaction.c: line 71: assertion failed: (box->opened) > .. >> If you give me some logic to throw at this, I'll take a shot at the >> programming. I tested my snarf code and it does the same thing. > > Does http://hg.dovecot.org/dovecot-2.0/rev/b7dd7a966a3a fix it? > > Yeah thanks! smime.p7s Description: S/MIME cryptographic signature
[Dovecot] [dovecot] Question about master user and PAM with dovecot 2.0.12
Is it possible to allow master users to be authenticated against PAM? Something like: passdb { driver = pam #args = /etc/dovecot/passwd.masterusers master = yes #pass = yes } and then have a userdb which qualifies what accounts are master accounts but doesn't have passwords? Thanks, Jonathan smime.p7s Description: S/MIME cryptographic signature
[Dovecot] [dovecot] Getting duplicates when using snarf plugin with mbox backend
Running dovecot version 2.0.11. To reproduce, open two imap sessions and issue a check command from each at the same time with new mail in the queue. 0 login testuser testpw 1 select inbox -- Deliver mail -- 2 check Thanks, Jonathan smime.p7s Description: S/MIME Cryptographic Signature
Re: [Dovecot] patch: allow proxy to lookup host by name
On Mar 12, 2010, at 7:57 AM, Stan Hoeppner wrote: > > Speaking of mbox, you're not going to be dropping it any time soon are you? > I've noticed there are only like 2 people left on planet earth using it, me > and one other guy. :( > We love mbox here at PSU. -Jonathan smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] Question about mbox_snarf and dovecot2.0
On Feb 2, 2010, at 12:47 PM, Timo Sirainen wrote: > On Tue, 2010-02-02 at 12:33 -0500, Jonathan Siegle wrote: >> I verified that not using the mbox_snarf works just fine. I have no >> namespaces enabled. > > The code contains: > > //FIXME:storage->set->mail_full_filesystem_access = TRUE; > > IIRC I couldn't figure out a nice and easy way to actually make that > work. Maybe the whole plugin could be redesigned. Probably by creating a > configuration like: > > namespace { > location = mbox:~/mail > .. > } > namespace { > prefix = ~~MboxSnarf~~/ # or whatever unique string you can think of > location = mbox:~/mail:INBOX=/var/mail/%u > list = no > hidden = yes > } > plugin { > mbox_snarf = ~~MboxSnarf~~/ > } > First, I haven't found a working config to use for dovecot2.0 and mbox_snarf. As I need this to function, I've been thinking about your words above and been reading http://wiki.dovecot.org/Design/Storage/Plugins. The mentality of it reminds me of the lazy-expunge-plugin, but this API is taking some time getting used to. Any pointers would be appreciated. Thanks, Jonathan smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] Question about mbox_snarf and dovecot2.0
On Apr 29, 2010, at 1:27 PM, Timo Sirainen wrote: > On Thu, 2010-04-29 at 11:11 -0400, Jonathan Siegle wrote: >>> namespace { >>> location = mbox:~/mail >>> .. >>> } >>> namespace { >>> prefix = ~~MboxSnarf~~/ # or whatever unique string you can think of >>> location = mbox:~/mail:INBOX=/var/mail/%u >>> list = no >>> hidden = yes >>> } >>> plugin { >>> mbox_snarf = ~~MboxSnarf~~/ >>> } > > I guess my idea above was that mbox_snarf would: > > 1) Do nothing special when opening the mailbox (unlike now), except add > hooks to sync_init and mailbox_close. > Ok gotcha. > 2) when mbox_snarf_sync_init() is called the first time: > - find mbox_snarf namespace Know how to do that. > - open INBOX from that namespace and keep it open Ok sync init would open /var/mail/%u > - goto 3) > > 3) every time mbox_snarf_sync_init() is called: > - call mbox_snarf() > Oh Ok. So mbox_snarf would open the SNARF FILE and do the move and then return it like it does now. > 4) on mailbox_close() close the snarf mailbox > >> As I need this to function, I've been thinking about your words above >> and been reading http://wiki.dovecot.org/Design/Storage/Plugins. The >> mentality of it reminds me of the lazy-expunge-plugin, but this API is >> taking some time getting used to. Any pointers would be appreciated. > > Maybe the above helps? That Plugins wiki page probably won't help you > much, since the current mbox_snarf already sets up the plugin stuff > correctly. "Overriding methods" is probably the only important part > there. The other Design/* pages could be more helpful. Yeah your thoughts help here. Thanks. Do you have a working example of mailbox_opened ? Would that solve my IDLE problem(doesn't stat /var/spool at all so no new messages)? -Jonathan smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] Question about mbox_snarf and dovecot2.0
On Apr 29, 2010, at 2:42 PM, Timo Sirainen wrote: > On Thu, 2010-04-29 at 14:00 -0400, Jonathan Siegle wrote: > >>> - open INBOX from that namespace and keep it open >> >> Ok sync init would open /var/mail/%u > > Well, you would simply call mailbox_open(ns->list, "INBOX"). Dovecot > internally then figures out that INBOX is /var/mail/%u based on the > namespace's configuration. Ok. I couldn't find a mailbox_open that takes a list and char* as arguments, but I found a mail_namespace_find_inbox function which works well. > Also note that nothing here now assumes anything about what mailbox > format is being used. The whole plugin should perhaps be renamed to > "snarf", since it would work with any mailbox formats > (source/destination could even be different formats). > Where do I specify the Destination mailbox in the config file? I have a source of INBOX(/var/spool). Should I just tack on SNARF on the var list? namespace foo { prefix = ~~Snarf/ location = mbox~/mail:INBOX=/var/mail/%u:SNARF=/some/other/file list = no hidden = yes inbox = yes } Thanks, Jonathan smime.p7s Description: S/MIME cryptographic signature
[Dovecot] libdovecot not referencing libiconv
OS: AIX 5.3 with IBM c compiler. Dovecot revision: 11278:2ead7574bb08 This problem is old for me, I just kept forgetting to report. Trying to start dovecot and I get: exec(): 0509-036 Cannot load program dovecot because of the following errors: rtld: 0712-001 Symbol iconv was referenced from module /usr/local/dovecot2.0beta5/lib/dovecot/libdovecot.so(), but a runtime definition of the symbol was not found. rtld: 0712-001 Symbol iconv_open was referenced from module /usr/local/dovecot2.0beta5/lib/dovecot/libdovecot.so(), but a runtime definition of the symbol was not found. rtld: 0712-001 Symbol iconv_close was referenced from module /usr/local/dovecot2.0beta5/lib/dovecot/libdovecot.so(), but a runtime definition of the symbol was not found. To fix this I did: --- a/src/lib-dovecot/Makefile.am Mon May 10 12:05:58 2010 -0400 +++ b/src/lib-dovecot/Makefile.am Mon May 10 12:31:36 2010 -0400 @@ -16,7 +16,7 @@ libdovecot_la_LIBADD = \ $(libs) \ $(MODULE_LIBS) \ - $(LTLIBICONV) + $(LIBICONV) libdovecot_la_DEPENDENCIES = $(libs) smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] libdovecot not referencing libiconv
On May 25, 2010, at 1:24 PM, Timo Sirainen wrote: > On Mon, 2010-05-10 at 13:41 -0400, Jonathan Siegle wrote: >> To fix this I did: >> >> --- a/src/lib-dovecot/Makefile.am Mon May 10 12:05:58 2010 -0400 >> +++ b/src/lib-dovecot/Makefile.am Mon May 10 12:31:36 2010 -0400 >> @@ -16,7 +16,7 @@ >> libdovecot_la_LIBADD = \ >>$(libs) \ >>$(MODULE_LIBS) \ >> - $(LTLIBICONV) >> + $(LIBICONV) >> >> libdovecot_la_DEPENDENCIES = $(libs) > > That's going to break other systems. What do you have in LIBICONV and > LTLIBICONV in Makefile? Maybe your iconv.m4 doesn't contain support for > LTLIBICONV? > > fgrep LIBICONV src/lib-dovecot/Makefile LIBICONV = -liconv $(LIBICONV) I can't find a iconv.m4 on my system. smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] libdovecot not referencing libiconv
On May 25, 2010, at 1:44 PM, Timo Sirainen wrote: > On Tue, 2010-05-25 at 13:39 -0400, Jonathan Siegle wrote: >>> That's going to break other systems. What do you have in LIBICONV and >>> LTLIBICONV in Makefile? Maybe your iconv.m4 doesn't contain support for >>> LTLIBICONV? >>> >>> >> fgrep LIBICONV src/lib-dovecot/Makefile >> LIBICONV = -liconv >>$(LIBICONV) >> >> I can't find a iconv.m4 on my system. > > I guess it should be in /usr/share/aclocal/. You're compiling from hg, > right? Yes. After updating some Gnu tools(automake/autoconf/m4/libtool/gettext/) and installing Gnu libiconv(was using system libiconv), I can report success. Looking back now, I probably only needed to install Gnu libiconv. tr27n12.aset.psu.edu# ldd libdovecot.so.0.0.0 libdovecot.so.0.0.0 needs: /usr/local/Gnu/lib/libiconv.so /usr/lib/libc.a(shr_64.o) /unix /usr/lib/libcrypt.a(shr_64.o) fgrep LIBICON Makefile LIBICONV = /usr/ladmin/Gnu/lib//libiconv.a LTLIBICONV = -L/usr/ladmin/Gnu/lib/ -liconv $(LTLIBICONV) However, I no longer have the --enable-header-install feature: $ ./configure --prefix=/usr/local/dovecot2.0beta4test --without-ssl --with-gssapi=yes --enable-header-install configure: WARNING: unrecognized options: --enable-header-install AND /usr/local/ovecot2.0beta4test/include/dovecot is still populated. thanks, Jonathan smime.p7s Description: S/MIME cryptographic signature
[Dovecot] Write a c program to access a mailbox as a system user using dovecot version 2 includes/libs
Should this be doable with the current design? Is there a good reference to do this? Thanks, Jonathan smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] Write a c program to access a mailbox as a system user using dovecot version 2 includes/libs
On Aug 5, 2010, at 10:16 AM, Timo Sirainen wrote: > On Thu, 2010-08-05 at 10:12 -0400, Jonathan Siegle wrote: >> Should this be doable with the current design? > > Yes. > >> Is there a good reference to do this? > > doveadm and dsync are good examples. Or maybe you shouldn't write a > separate C program, but a doveadm plugin instead (see e.g. doveadm_quota > plugin). http://wiki2.dovecot.org/Design has some docs. > > The first goal is to open these mailboxes(mdbox/dbox) with alpine version 2 and the mailboxes are on localhost. So I will look at dsync. Both programs use master_service_* calls to start up something. Is that the design I should follow? smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] Write a c program to access a mailbox as a system user using dovecot version 2 includes/libs
On Aug 6, 2010, at 7:28 AM, Timo Sirainen wrote: > On 6.8.2010, at 11.45, Jonathan Siegle wrote: > >> The first goal is to open these mailboxes(mdbox/dbox) with alpine version 2 >> and the mailboxes are on localhost. > > Why? Why not just let it use IMAP? Seems like a huge amount of work to make > c-client call lib-storage. And who knows how many symbol conflicts there will > be.. I thought you were talking about some standalone program. > Well the idea is for some client to be able to do updates to a mailbox. I grabbed alpine out of the air because it knows how to use different drivers to open mailboxes using the filesystem and not going through an IMAP socket. Maybe I shouldn't be going down this road. The end goal is for my Webmail client to access dovecot mailboxes. Does it make more sense to do the more traditional route of making Webmail an IMAP client? I don't know. I thought it would make more sense to just access the filesystem and write a smart Webmail client. Right now my Webmail is a caching AJAX POP client. Thanks, Jonathan smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] Question about "target is inferior" type of message
On Jun 17, 2009, at 11:00 AM, Michael Durket wrote: We're converting UW-IMAP users here to dovecot. Some of them have seen the error message issued when Thunderbird clients try to delete folders and I understand what's going on (from previous postings) that Dovecot doesn't support mbox subfolders within folders. Hi Michael, I'm running dovecot 1.2beta5. I'm not sure which version does not support folders with subfolders, but 1.2x does. Here are some imap commands to prove it works. 2 list "mail" * * LIST (\Noselect \HasChildren) "/" "mail" * LIST (\NoInferiors \UnMarked) "/" "mail/folder1" * LIST (\NoInferiors \UnMarked) "/" "mail/folder2" * LIST (\NoInferiors \UnMarked) "/" "mail/Trash" * LIST (\Noselect \HasChildren) "/" "mail/anotherfolder" * LIST (\NoInferiors \UnMarked) "/" "mail/anotherfolder/mbox1" * LIST (\NoInferiors \UnMarked) "/" "mail/foo2" * LIST (\NoInferiors \UnMarked) "/" "mail/foo3" 2 OK List completed. 3 list "mail/anotherfolder" * * LIST (\Noselect \HasChildren) "/" "mail/anotherfolder" * LIST (\NoInferiors \UnMarked) "/" "mail/anotherfolder/mbox1" 3 OK List completed. 4 delete mail/anotherfolder/mbox1 4 OK Delete completed. 5 delete mail/anotherfolder 5 OK Delete completed. 6 create mail/anotherfolder/ 6 OK Create completed. 7 create mail/anotherfolder/mbox2 7 OK Create completed. 8 list "mail/anotherfolder" * * LIST (\Noselect \HasChildren) "/" "mail/anotherfolder" * LIST (\NoInferiors \UnMarked) "/" "mail/anotherfolder/mbox2" 8 OK List completed. Now there is config setting in dovecot.conf called tb-extra-mailbox- sep. Maybe that is what the complaint is. Tbird will put the delimiter on the back end of the delete command(delete mail/anotherfolder/). What puzzles me is that many of these users already have folders within folders (I've found typically 2 levels of nested folders within their main Mail directory). And they're not complaining that they can't read the messages in those folders. If UW-IMAP supposedly never supported this for mbox format, and Dovecot doesn't support it, how is this possible under Dovecot? (We don't have any maildir format mailboxes here at all). I have an older version of Uwash in use(2006k?) and folders with subfolders works. Maybe you are talking about folders that have messages and subfolders? -Jonathan smime.p7s Description: S/MIME cryptographic signature
[Dovecot] libsieve compilation error on AIX 5.3
Dovecot revision: 9186:f70192e7ff3c libsieve revision: 958:e86a81f4bf28 Using IBM c compiler version 8. Trying to compile libsieve(64 bit mode if it matters): "sieve-ast.c", line 302.9: 1506-045 (S) Undeclared identifier last. "sieve-ast.c", line 302.9: 1506-045 (S) Undeclared identifier result. "sieve-ast.c", line 320.9: 1506-275 (S) Unexpected text node encountered. "sieve-ast.c", line 320.9: 1506-045 (S) Undeclared identifier node. "sieve-ast.c", line 324.9: 1506-045 (S) Undeclared identifier last. "sieve-ast.c", line 324.9: 1506-045 (S) Undeclared identifier result. gmake[4]: *** [sieve-ast.lo] Error 1 Thanks, Jonathan smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] libsieve compilation error on AIX 5.3
On Jul 6, 2009, at 9:13 AM, Stephan Bosch wrote: Jonathan Siegle wrote: Dovecot revision: 9186:f70192e7ff3c libsieve revision: 958:e86a81f4bf28 Using IBM c compiler version 8. Trying to compile libsieve(64 bit mode if it matters): "sieve-ast.c", line 302.9: 1506-045 (S) Undeclared identifier last. "sieve-ast.c", line 302.9: 1506-045 (S) Undeclared identifier result. "sieve-ast.c", line 320.9: 1506-275 (S) Unexpected text node encountered. "sieve-ast.c", line 320.9: 1506-045 (S) Undeclared identifier node. "sieve-ast.c", line 324.9: 1506-045 (S) Undeclared identifier last. "sieve-ast.c", line 324.9: 1506-045 (S) Undeclared identifier result. Apparently, the typeof() operator is not supported by your C compiler. Having relatively little experience with cross-platform programming, I did not notice that the typeof operator is not part of the main ANSI C standard. I've removed it from the sources entirely (making things somewhat more ugly), so your problem should be resolved with the following changes: Ok. I added -qkeyword=typeof to my CFLAGS and the errors go away and no further issues on compilation. http://hg.rename-it.nl/dovecot-libsieve/rev/bd106aa05040 I didn't have to update to this to make it compile. I don't know if you need to keep that patch set around. Thanks! Jonathan I browsed the web a little and found that IBM C should at least support __typeof__(e), and, with the appropriate command line options, typeof as well. Timo, do you know how this works and other systems that may be affected? Or is it a better idea to avoid typeof() in general? Regards, Stephan. smime.p7s Description: S/MIME cryptographic signature
[Dovecot] sieve filtering setup
I'm looking at implementing sieve in my environment. Software is: dovecot-1.2-sieve revision 1022:3c9a22c28156 dovecot-1.2 revision 9269:a303bb82c1c9 AIX 5.3 with sendmail mta using prescribed deliver lda. I have a few questions. I'll have 110k sieve files(1 for each user). Does sieve read the file each time a new message is accepted by sendmail? Are there any measurements on cpu load for sieve filters? Thanks, Jonathan smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] sieve filtering setup
On Jul 30, 2009, at 11:12 AM, Stephan Bosch wrote: Jonathan Siegle schreef: I'm looking at implementing sieve in my environment. Software is: dovecot-1.2-sieve revision 1022:3c9a22c28156 dovecot-1.2 revision 9269:a303bb82c1c9 AIX 5.3 with sendmail mta using prescribed deliver lda. I have a few questions. I'll have 110k sieve files(1 for each user). Does sieve read the file each time a new message is accepted by sendmail? Are there any measurements on cpu load for sieve filters? If all is configured correctly, the Sieve scripts are compiled once each time they are changed or created. After that, the deliver LDA only reads the compiled binary from disk for each message. I recently found out about something called memcached. The goal of memcached(server)[1] and libmemcached(client library)[2] is to store maps in memory of tokens. So my key would be "jsiegle_sieve" and my data would be my sieve file. So instead of the 10-20 million reads to disk, we would just pull from memory. The logic looks like this: On update of sieve file, Do validation of file. Do compilation of file. Delete key if it exists and add new. On new mail, sieve plugin would call memcached_get() and get the token. I could be very wrong, but I think this is a big win. memcached is designed for small files(<1MB). So if each of my users has a 2KB file, that would only be 20MB of memory for usage. [1] http://www.danga.com/memcached/ [2] http://tangent.org/552/libmemcached.html -Jonathan smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] sieve filtering setup
On Jul 30, 2009, at 1:29 PM, Timo Sirainen wrote: On Thu, 2009-07-30 at 11:30 -0400, Jonathan Siegle wrote: I recently found out about something called memcached. The goal of memcached(server)[1] and libmemcached(client library)[2] is to store maps in memory of tokens. So my key would be "jsiegle_sieve" and my data would be my sieve file. So instead of the 10-20 million reads to disk, we would just pull from memory. Then again, if you have enough memory your OS could be doing that automatically already. Good point. AIX does love to use memory to cache filesystems. I think I'll try to do some formal testing. The hooks appear to be fairly easy to put into the sieve code. Or maybe if the Sieve plugin supports giving a separate path to downloaded scripts, the destination could be in a ramdisk or if you're using Linux: http://memcachefs.sourceforge.net/ Also I'm hopefully going to abstracting out filesystem accessing code in Dovecot's index files and dbox code. Sieve could use this same FS API, and you could implement whatever backend to actually perform the FS access, like memcached.. Are you saying that it isn't possible to put the memcached hooks in now until this work is done? It looks like I just put hooks into sieve_script_init() to read from memcached before disk. -Jonathan smime.p7s Description: S/MIME cryptographic signature
[Dovecot] dovecot-1.2.4, sendmail,deliver,sieve
I'm trying to setup deliver as an LDA for system users. I've read http://wiki.dovecot.org/LDA/Sendmail which handles virtual users. I don't see a README for what I need. I want sendmail to accept the mail and then hand off to deliver with sieve plugin. Deliver should look in the users home directory which is found in /etc/passwd for the sieve-script to see what to do with it. If the sieve-script DNE, then just deliver to INBOX(from dovecot.conf). Any help would be appreciated. Thanks, Jonathan protocols: imap disable_plaintext_auth: no login_dir: /usr/dovecot-1.2.4/var/run/dovecot/login login_executable: /usr/dovecot-1.2.4/libexec/dovecot/imap-login login_greeting: Dovecot ready. login_processes_count: 30 max_mail_processes: 4096 mail_location: mbox:%h:INBOX=/var/spool/mail/%Ju/%u mail_full_filesystem_access: yes mmap_disable: yes dotlock_use_excl: no mbox_write_locks: fcntl mbox_lazy_writes: no mail_plugins: mbox_snarf mail_plugin_dir: /usr/dovecot-1.2.4/lib/dovecot/imap/ imap_client_workarounds: tb-extra-mailbox-sep imap_id_log: * lda: postmaster_address: postmas...@localhost auth default: mechanisms: plain gssapi krb5_keytab: /etc/krb5/krb5.keytab gssapi_hostname: $ALL debug: yes worker_max_count: 2 passdb: driver: pam args: max_requests=1 userdb: driver: passwd plugin: mbox_snarf: /gpfs/inbox/%Ju/%u smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] dovecot-1.2.4, sendmail,deliver,sieve
On Aug 19, 2009, at 11:23 AM, Steffen Kaiser wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Wed, 19 Aug 2009, Jonathan Siegle wrote: I'm trying to setup deliver as an LDA for system users. I've read I use (all in one line): FEATURE(`local_procmail', `/etc/mail/smrsh/dovecot-deliver', `/etc/ mail/smrsh/dovecot-deliver -e -f $g -d $u -m $h')dnl /etc/mail/smrsh/dovecot-deliver is a symlink to the real LDA. Thanks! That did the trick. I had to comment out the socket master/ client lines in dovecot.conf and get the socket set up and then everything worked. smime.p7s Description: S/MIME cryptographic signature
[Dovecot] Sieve plugin load error
Running: dovecot revision 9333:15b4b2c904fe dovecot sieve revision: 1091:0fe00772366f AIX 5.3 64bit Deliver gives error: Aug 19 12:30:52 deliver(jsiegle): Error: dlopen(/usr/dovecot-1.2.4/lib/ dovecot/lda/lib90_sieve_plugin.so) failed: rtld: 0712-001 Symbol notify_command was referenced from module /usr/dovecot-1.2.4/lib/dovecot/lda/ lib90_sieve_plugin.so(), but a runtime definition of the symbol was not found. rtld: 0712-001 Symbol notify_operation/usr/dovecot-1.2.4/lib/dovecot/ lda/lib90_sieve_plugin.so was referenced from module /usr/dovecot-1.2.4/lib/dovecot/lda/ lib90_sieve_plugin.so(), but a runtime definition of the symbol was not found. rtld: 0712-002 fatal error: exiting. Aug 19 12:30:52 deliver(jsiegle): Fatal: Couldn't load required plugins Aug 19 12:38:53 deliver(jsiegle): Info: Loading modules from directory: /usr/dovecot-1.2.4/lib/dovecot/lda I see a local symbol referenced in the so and also an Undefined as well. tr27n12.aset.psu.edu# nm -X 64 /usr/ladmin/dovecot-1.2.4/lib/dovecot/ lda/lib90_sieve_plugin.so|grep notify_command notify_command U - notify_command d 4563447648 8 notify_operation is different: # nm lib90_sieve_plugin.so |grep ^notify_operation notify_operation U - # nm lib90_sieve_plugin.a |grep ^notify_operation notify_operation U - notify_operation d 448 8 smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] Sieve plugin load error
Jonathan Siegle said the following on 8/19/09 1:46 PM: Running: dovecot revision 9333:15b4b2c904fe dovecot sieve revision: 1091:0fe00772366f AIX 5.3 64bit Deliver gives error: Aug 19 12:30:52 deliver(jsiegle): Error: dlopen(/usr/dovecot-1.2.4/lib/ dovecot/lda/lib90_sieve_plugin.so) failed: rtld: 0712-001 Symbol notify_command was referenced from module /usr/dovecot-1.2.4/lib/dovecot/lda/ lib90_sieve_plugin.so(), but a runtime definition of the symbol was not found. rtld: 0712-001 Symbol notify_operation/usr/dovecot-1.2.4/lib/dovecot/ lda/lib90_sieve_plugin.so was referenced from module /usr/dovecot-1.2.4/lib/dovecot/lda/ lib90_sieve_plugin.so(), but a runtime definition of the symbol was not found. Ok I think I see the problem. I don't have a shared library for lib-sieve. Just an indexed library(.a). I've been lazy and using http://wiki.dovecot.org/AixPluginsSupport to solve this for dovecot. How do I solve this? $ nm libsieve.a |grep ^notify_operation notify_operation U - notify_operation d 448 8 notify_operation D5160 40 notify_operation d4696 8
[Dovecot] Help with building sieve
Am building sieve plugin on AIX. This is revision 1095:8010f1eb66b3. I see the following error when doing the build: rm -fr .libs/lib90_sieve_plugin.lax^M mkdir .libs/lib90_sieve_plugin.lax^M rm -fr .libs/lib90_sieve_plugin.lax/libsieve.a^M mkdir .libs/lib90_sieve_plugin.lax/libsieve.a^M (cd .libs/lib90_sieve_plugin.lax/libsieve.a && ar x /usr/sadmin/src/ imapservers/dovecothg/dovecot-1.2-sievepsu/src/plugins/lda-si eve/../../../src/lib-sieve/.libs/libsieve.a)^M libtool: link: warning: object name conflicts; renaming object files^M libtool: link: warning: to ensure that they will not overwrite^M Usage: uniq [-c | -d | -u] [-f Fields] [-s Chars] [-Fields] [+Chars] [Input_file [Output_file]]^M I think this is the crux of my missing notify_operation token I reported the other day. I can solve the uniq -cd problem. I don't know what libtool is trying to do there. smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] Help with building sieve
On Aug 21, 2009, at 10:09 AM, Stephan Bosch wrote: Jonathan Siegle wrote: Am building sieve plugin on AIX. This is revision 1095:8010f1eb66b3. I see the following error when doing the build: rm -fr .libs/lib90_sieve_plugin.lax^M mkdir .libs/lib90_sieve_plugin.lax^M rm -fr .libs/lib90_sieve_plugin.lax/libsieve.a^M mkdir .libs/lib90_sieve_plugin.lax/libsieve.a^M (cd .libs/lib90_sieve_plugin.lax/libsieve.a && ar x /usr/sadmin/src/ imapservers/dovecothg/dovecot-1.2-sievepsu/src/plugins/lda-si eve/../../../src/lib-sieve/.libs/libsieve.a)^M libtool: link: warning: object name conflicts; renaming object files^M libtool: link: warning: to ensure that they will not overwrite^M Usage: uniq [-c | -d | -u] [-f Fields] [-s Chars] [-Fields] [+Chars] [Input_file [Output_file]]^M I think this is the crux of my missing notify_operation token I reported the other day. I can solve the uniq -cd problem. I don't know what libtool is trying to do there. I am wondering why it breaks specifically on the notify command and opcode structs. All other extensions are defined identically and should show the same symptoms. Maybe there is a subtle difference somewhere that triggers this problem on AIX. I'll have a look, but I won't be able to reproduce anything. Timo also may have an idea. He'll be back some time next week. I'm closing in on the problem. uniq/sort on AIX are different than GNU uniq/sort. I've built gnu coreutils and am using that. ar is also different in that it won't remove duplicate symbols. I'll get it figured out and report back, but any tips are appreciated. -Jonathan smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] Help with building sieve
On Aug 21, 2009, at 1:01 PM, Jonathan Siegle wrote: On Aug 21, 2009, at 10:09 AM, Stephan Bosch wrote: Jonathan Siegle wrote: Am building sieve plugin on AIX. This is revision 1095:8010f1eb66b3. I see the following error when doing the build: rm -fr .libs/lib90_sieve_plugin.lax^M mkdir .libs/lib90_sieve_plugin.lax^M rm -fr .libs/lib90_sieve_plugin.lax/libsieve.a^M mkdir .libs/lib90_sieve_plugin.lax/libsieve.a^M (cd .libs/lib90_sieve_plugin.lax/libsieve.a && ar x /usr/sadmin/ src/imapservers/dovecothg/dovecot-1.2-sievepsu/src/plugins/lda-si eve/../../../src/lib-sieve/.libs/libsieve.a)^M libtool: link: warning: object name conflicts; renaming object files^M libtool: link: warning: to ensure that they will not overwrite^M Usage: uniq [-c | -d | -u] [-f Fields] [-s Chars] [-Fields] [+Chars] [Input_file [Output_file]]^M I think this is the crux of my missing notify_operation token I reported the other day. I can solve the uniq -cd problem. I don't know what libtool is trying to do there. I am wondering why it breaks specifically on the notify command and opcode structs. All other extensions are defined identically and should show the same symptoms. Maybe there is a subtle difference somewhere that triggers this problem on AIX. I'll have a look, but I won't be able to reproduce anything. Timo also may have an idea. He'll be back some time next week. I'm closing in on the problem. uniq/sort on AIX are different than GNU uniq/sort. I've built gnu coreutils and am using that. ar is also different in that it won't remove duplicate symbols. I'll get it figured out and report back, but any tips are appreciated. -Jonathan Ok I got it work. I just need to figure out long term. The issue is why I must use GNU ar vs AIX ar when building lib90_sieve_plugin.so, but AIX ar and not GNU ar when building libsieve.a. Plus it would be nice if GNU sort/uniq had different flags than AIX sort/uniq, but I don't see that happening.. -Jonathan smime.p7s Description: S/MIME cryptographic signature
[Dovecot] dovecot-auth stops responding
Running dovecot 1.2.4(imap only) on AIX out of inetd. I accept passwords(PAM for auth) and kerberos tickets. Occasionally we see dovecot-auth stop responding. I just kill it off and another process is spawned which works fine. It doesn't look like it is hitting any ulimits(11MB for memory). I issued a kill -11 and ran it through the debugger. Here is what it says: dbx dovecot-auth core Type 'help' for help. [using memory image in core] reading symbolic information ... Segmentation fault in __fd_poll at 0x9117634 ($t1) 0x9117634 (__fd_poll+0x98) e8410028 ld r2,0x28(r1) (dbx) where __fd_poll(??, ??, ??) at 0x9117634 ioloop-poll.poll(__listptr = 0x0001104dd330, __nfds = 145, __timeout = 1051), line 121 in "poll.h" io_loop_handler_run(ioloop = 0x00011001bfb0), line 157 in "ioloop- poll.c" io_loop_run(ioloop = 0x00011001bfb0), line 335 in "ioloop.c" main(argc = 1, argv = 0x0ad8), line 347 in "main.c" Here are my ulimits: ulimit -a time(seconds)unlimited file(blocks) 2097151 data(kbytes) 1048576 stack(kbytes)65536 memory(kbytes) 65536 coredump(blocks) 2097151 nofiles(descriptors) 32768 Any help would be appreciated. Thanks, Jonathan smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] dovecot-auth stops responding
On Sep 9, 2009, at 4:16 PM, Timo Sirainen wrote: On Wed, 2009-09-09 at 14:12 -0400, Jonathan Siegle wrote: Running dovecot 1.2.4(imap only) on AIX out of inetd. I accept passwords(PAM for auth) and kerberos tickets. Occasionally we see dovecot-auth stop responding. I just kill it off and another process is spawned which works fine. If you use PAM then I guess you should have auth worker processes too? Yes So one dovecot-auth and 1+ "dovecot-auth -w"? Yes. One of each. Does it help if you kill off the worker processes? It only helps when I kill dovecot-auth, not dovecot-auth -w. What if you kill imap-login processes instead? I don't have imap-login processes associated with inetd spawned dovecot. imap-login is what I call in inetd. I don't see them in the process table. Segmentation fault in __fd_poll at 0x9117634 ($t1) 0x9117634 (__fd_poll+0x98) e8410028 ld r2,0x28(r1) (dbx) where __fd_poll(??, ??, ??) at 0x9117634 ioloop-poll.poll(__listptr = 0x0001104dd330, __nfds = 145, This just shows that it was waiting for something to do. Does AIX have something like strace or truss? Use it to see what's happening when a request comes. Yes I have truss. So tomorrow when this happens I'll do truss -f -p on the dovecot-auth process? Oh my dovecot.conf file has this for pam: passdb pam { args = max_requests=1 } because of the expire password problem for kerberos. thanks, Jonathan smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] dovecot-auth stops responding
On Sep 10, 2009, at 2:43 PM, Timo Sirainen wrote: On Thu, 2009-09-10 at 14:32 -0400, Jonathan Siegle wrote: It only helps when I kill dovecot-auth, not dovecot-auth -w. Interesting.. What if you kill imap-login processes instead? I don't have imap-login processes associated with inetd spawned dovecot. Why do you use inetd? I'm currently wondering if I should bother adding inetd support to Dovecot v2.0. We ran imap(Uwash) for years. Is there a reason to not run it from inetd? Yes I have truss. So tomorrow when this happens I'll do truss -f -p on the dovecot-auth process? Yes. Also having auth_debug=yes enabled might show something useful. What does it log last before it stops responding? I found something in syslog today: local0.log.20090916:Sep 16 11:58:01 tr27n18.aset.psu.edu dovecot: auth (default): BUG: Worker sent reply with id 1, expected 2 local0.log.20090916:Sep 16 11:58:01 tr27n18.aset.psu.edu dovecot: auth (default): worker-server(mck2,146.186.125.214): Aborted: Worker is buggy smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] dovecot-auth stops responding
On Sep 16, 2009, at 1:35 PM, Jonathan Siegle wrote: On Sep 10, 2009, at 2:43 PM, Timo Sirainen wrote: Yes. Also having auth_debug=yes enabled might show something useful. What does it log last before it stops responding? I found something in syslog today: local0.log.20090916:Sep 16 11:58:01 dovecot: auth(default): BUG: Worker sent reply with id 1, expected 2 local0.log.20090916:Sep 16 11:58:01 dovecot: auth(default): worker- server(foo,146.186.125.214): Aborted: Worker is buggy Found thread from August("Aborted: Worker is buggy" ) and am applying patch as suggested. smime.p7s Description: S/MIME cryptographic signature
[Dovecot] Question about installing managesieve
I'm reading http://wiki.dovecot.org/ManageSieve/Install . I already use mercurial queues. These are the steps I see to install: hg clone http://hg.dovecot.org/dovecot-1.2 cd dovecot-1.2 hg clone http://hg.rename-it.nl/dovecot-1.2-managesieve-patch .hg/ patches hg qpush Do the steps change? Thanks, Jonathan smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] dovecot-auth stops responding
On Oct 27, 2009, at 7:36 PM, Timo Sirainen wrote: On Wed, 2009-09-16 at 13:50 -0400, Jonathan Siegle wrote: I found something in syslog today: local0.log.20090916:Sep 16 11:58:01 dovecot: auth(default): BUG: Worker sent reply with id 1, expected 2 local0.log.20090916:Sep 16 11:58:01 dovecot: auth(default): worker- server(foo,146.186.125.214): Aborted: Worker is buggy Found thread from August("Aborted: Worker is buggy" ) and am applying patch as suggested. Did that fix dovecot-auth hanging? We have not tested it fully yet. It should be back on the todo list next week. -Jonathan smime.p7s Description: S/MIME cryptographic signature
[Dovecot] [dovecot] Compiling dovecot 2.0 revision 10435:73679e39f072 on AIX
Trying to compile and getting this error: "process-title.c", line 139.14: 1506-045 (S) Undeclared identifier argv_memblock. "process-title.c", line 140.14: 1506-045 (S) Undeclared identifier environ_memblock. gmake[4]: *** [process-title.lo] Error 1 HAVE_SETPROCTITLE is undefined in config.h -Jonathan smime.p7s Description: S/MIME cryptographic signature
[Dovecot] [dovecot] Compile problem on AIX with dovecot 1.2 rev 10449:a164075ea33a
+ exec /usr/vac/bin/xlc -g -qarch=pwr5x -DHAVE_CONFIG_H -I. -I. -I../.. -I../../src/lib -I../../src/lib-test -I../../src/lib-mail -q64 -c -M mail-index-transaction.c -DPIC "../../src/lib/bsearch-insert-pos.h", line 39.9: 1506-236 (W) Macro name array_bsearch has been redefined. "../../src/lib/bsearch-insert-pos.h", line 39.9: 1506-358 (I) "array_bsearch" is defined on line 285 of ../../src/lib/array.h. "mail-index-transaction.c", line 34.16: 1506-045 (S) Undeclared identifier idx_r. gmake[3]: *** [mail-index-transaction.lo] Error 1 gmake[3]: Leaving directory `/usr/sadmin/src/imapservers/dovecothg/dovecot-2.0psu/src/lib-index' thanks, Jonathan smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] [dovecot] Compile problem on AIX with dovecot 1.2 rev 10449:a164075ea33a
On Dec 10, 2009, at 10:56 AM, Timo Sirainen wrote: > On Dec 10, 2009, at 9:41 AM, Jonathan Siegle wrote: > >> + exec /usr/vac/bin/xlc -g -qarch=pwr5x -DHAVE_CONFIG_H -I. -I. -I../.. >> -I../../src/lib -I../../src/lib-test -I../../src/lib-mail -q64 -c -M >> mail-index-transaction.c -DPIC >> "../../src/lib/bsearch-insert-pos.h", line 39.9: 1506-236 (W) Macro name >> array_bsearch has been redefined. >> "../../src/lib/bsearch-insert-pos.h", line 39.9: 1506-358 (I) >> "array_bsearch" is defined on line 285 of ../../src/lib/array.h. >> "mail-index-transaction.c", line 34.16: 1506-045 (S) Undeclared identifier >> idx_r. > > Fixed. > I'm not making another thread at this point, just say something if I should. Haven't compiled this in weeks... Next: + exec /usr/vac/bin/xlc -g -qarch=pwr5x -I/usr/ladmin/include/ -q64 -bexpall -brtl -o test-dsync-brain test-dsync-brain.o test-dsync-worker.o test-dsync-common.o dsync-data.o dsync-brain.o dsync-worker.o -L/usr/ladmin/lib/ ../lib-test/.libs/libtest.a ../lib-mail/.libs/libmail.a ../lib-imap/.libs/libimap.a ../lib/.libs/liblib.a -lrt -bE:.libs/libtest.exp -bE:.libs/libmail.exp -bE:.libs/libimap.exp -bE:.libs/liblib.exp ld: 0711-317 ERROR: Undefined symbol: .charset_to_utf8_end ld: 0711-317 ERROR: Undefined symbol: .charset_to_utf8_begin ld: 0711-317 ERROR: Undefined symbol: .charset_to_utf8 ld: 0711-317 ERROR: Undefined symbol: .charset_is_utf8 ld: 0711-317 ERROR: Undefined symbol: .charset_to_utf8_str smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] [dovecot] Compile problem on AIX with dovecot 1.2 rev 10449:a164075ea33a
On Dec 10, 2009, at 1:44 PM, Timo Sirainen wrote: > On Thu, 2009-12-10 at 13:40 -0500, Jonathan Siegle wrote: >> I'm not making another thread at this point, just say something if I should. >> Haven't compiled this in weeks... > > That's fine. > >> + exec /usr/vac/bin/xlc -g -qarch=pwr5x -I/usr/ladmin/include/ -q64 -bexpall >> -brtl -o test-dsync-brain test-dsync-brain.o test-dsync-worker.o >> test-dsync-common.o dsync-data.o dsync-brain.o dsync-worker.o >> -L/usr/ladmin/lib/ ../lib-test/.libs/libtest.a ../lib-mail/.libs/libmail.a >> ../lib-imap/.libs/libimap.a ../lib/.libs/liblib.a -lrt -bE:.libs/libtest.exp >> -bE:.libs/libmail.exp -bE:.libs/libimap.exp -bE:.libs/liblib.exp >> ld: 0711-317 ERROR: Undefined symbol: .charset_to_utf8_end > > Fixed now, I think? > No. I found my fix from http://www.mail-archive.com/dovecot@dovecot.org/msg18813.html and that fixed it. Here is the fix if you don't want to read the above page: I add this ../lib-charset/.libs/libcharset.a -liconv to the Makefile for src/dsync. I doubt I need -liconv at this point. I think I noticed it on the compile line. thanks for the fast commits.. -Jonathan smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] [dovecot] Compile problem on AIX with dovecot 1.2 rev 10449:a164075ea33a
On Dec 10, 2009, at 3:52 PM, Timo Sirainen wrote: > > How about now? :) > Yes! Thanks bunches. -Jonathan smime.p7s Description: S/MIME cryptographic signature
[Dovecot] Question about Pigeonhole for Dovecot v2.0 revision 1223:67f7f59671e8
I'm looking for documentation on the testsuite program. The most useful piece would be a sample scriptfile. smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] Question about Pigeonhole for Dovecot v2.0 revision 1223:67f7f59671e8
On Jan 28, 2010, at 11:42 AM, Stephan Bosch wrote: > Jonathan Siegle wrote: >> I'm looking for documentation on the testsuite program. The most useful >> piece would be a sample scriptfile. > > All test scripts are located in the tests/ directory. > Ah there they are. Ok thanks. > The vnd.dovecot.testsuite extension is currently not documented. You will > have to infer the syntax from the existing examples. The testsuite is under > active development, so script syntax may change in the near future. Also, > some parts of the syntax are downright ugly and suggestions are welcome. Sure no problem. I can read along when changes happen. Thanks Stephan! -Jonathan smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] Question about Pigeonhole for Dovecot v2.0 revision 1223:67f7f59671e8
On Jan 28, 2010, at 11:54 AM, Stephan Bosch wrote: > Jonathan Siegle wrote: >> On Jan 28, 2010, at 11:42 AM, Stephan Bosch wrote: >>> Jonathan Siegle wrote: >>>> I'm looking for documentation on the testsuite program. The most useful >>>> piece would be a sample scriptfile. >>> All test scripts are located in the tests/ directory. >>> >> Ah there they are. Ok thanks. >>> The vnd.dovecot.testsuite extension is currently not documented. You will >>> have to infer the syntax from the existing examples. The testsuite is under >>> active development, so script syntax may change in the near future. Also, >>> some parts of the syntax are downright ugly and suggestions are welcome. >> Sure no problem. I can read along when changes happen. > > Just wondering... what are you working on? :) > I'm trying to understand what changes I would have to make to roll our custom WM filters to sieve filters and what choices need to be made so our custom Webmail can work with dovecot from an API standpoint if we go that route. smime.p7s Description: S/MIME cryptographic signature
[Dovecot] [dovecot] Problems installing managesieve and then managesieve-login
Hi, Software: AIX 5.3 dovecot-2.0-pigeonhole revision 1234:7b3db97399de with dovecot revision 10621:a978089d090b I'm having problems installing managesieve. I modified the Makefile because DESTDIR was not being found. Here is what I changed to make it work: echo " $(INSTALL_PROGRAM_ENV) $(LIBTOOL) --mode=install $(dovecot_pkglibexecPROGRAMS_INSTALL) '$$p' '/usr/local/dovecot2.0beta2/bin$(dovecot_pkglibexecdir)/$$f'"; \ $(INSTALL_PROGRAM_ENV) $(LIBTOOL) --mode=install $(dovecot_pkglibexecPROGRAMS_INSTALL) "$$p" "/usr/local/dovecot2.0beta2/bin$(dovecot_pkglibexecdir)/$$f" || exit 1; \ --- Here is ORIGINAL: echo " $(INSTALL_PROGRAM_ENV) $(LIBTOOL) --mode=install $(dovecot_pkglibexecPROGRAMS_INSTALL) '$$p' '$(DESTDIR)$(dovecot_pkglibexecdir)/$$f'"; \ $(INSTALL_PROGRAM_ENV) $(LIBTOOL) --mode=install $(dovecot_pkglibexecPROGRAMS_INSTALL) "$$p" "$(DESTDIR)$(dovecot_pkglibexecdir)/$$f" || exit 1; \ Here is the error: test -z "" || /usr/local/src/imapservers/dovecothg/dovecot-2.0-pigeonhole/install-sh -d "" /bin/sh ../../libtool --mode=install /usr/ladmin/bin/install -c 'managesieve' '/managesieve' libtool: install: `' must be an absolute directory name Try `libtool --help --mode=install' for more information. gmake[3]: *** [install-dovecot_pkglibexecPROGRAMS] Error 1 managesieve-login is having the same problem. sieve-tools installed just fine. On my configure line I have: $ ./configure --with-dovecot=../dovecot-2.0 --prefix=/usr/local/dovecot2.0beta2 --with-managesieve=yes Thanks, Jonathan smime.p7s Description: S/MIME cryptographic signature
Re: [Dovecot] [dovecot] Problems installing managesieve and then managesieve-login
On Feb 1, 2010, at 10:43 AM, Timo Sirainen wrote: > > Sounds like this would help: > http://hg.dovecot.org/dovecot-2.0/rev/de9d6dae7fe5 > Fixed. Thanks! Jonathan smime.p7s Description: S/MIME cryptographic signature
[Dovecot] Question about mbox_snarf and dovecot2.0
Hello, I'm running revision 10622:de9d6dae7fe5 on AIX 5.3 with some local mods for our inbox hash function. I'm having a problem doing a "select inbox" when I use the mbox_snarf plugin. When I run truss on the process I see the following stat calls: 614528: kread(9, " 1 s e l e c t i n b".., 4096)= 15 614528: statx("/gpfs/inbox/14/tstem38", 0x02F8, 176, 0) = 0 614528: statx("/gpfs/users/t/s/tstem38/IMAP//gpfs/inbox/14/.imap/tstem38", 0x0FFFED98, 176, 0) Err#2 ENOENT 614528: statx("/gpfs/users/t/s/tstem38/IMAP//gpfs/inbox/14/tstem38", 0x0FFFEDD8, 176, 0) Err#2 ENOENT 614528: statx("/gpfs/users/t/s/tstem38/IMAP", 0x0FFFEC88, 176, 0) = 0 614528: statx("/gpfs/users/t/s/tstem38/IMAP//gpfs/inbox/14/tstem38", 0x0FFFEDD8, 176, 0) Err#2 ENOENT 614528: statx("/gpfs/users/t/s/tstem38/IMAP", 0x0FFFEC88, 176, 0) = 0 and the client gets: 1 NO [CANNOT] Invalid mailbox name For pathing I have: plugin.conf: mbox_snarf = /gpfs/inbox/%Ju/%u and mail.conf:mail_location = mbox:%h:INBOX=/var/spool/mail/%Ju/%u I verified that not using the mbox_snarf works just fine. I have no namespaces enabled. Thanks, Jonathan smime.p7s Description: S/MIME cryptographic signature