- dovecot - dovecot.org

Re: [ceph-users] dovecot + cephfs - sdbox vs mdbox
by Webert de Souza Lima 04 Oct '18

04 Oct '18

Thanks Jack. That's good to know. It is definitely something to consider. In a distributed storage scenario we might build a dedicated pool for that and tune the pool as more capacity or performance is needed. Regards, Webert Lima DevOps Engineer at MAV Tecnologia *Belo Horizonte - Brasil* *IRC NICK - WebertRLZ* On Wed, May 16, 2018 at 4:45 PM Jack <ceph(a)jack.fr.eu.org> wrote: > On 05/16/2018 09:35 PM, Webert de Souza Lima wrote: > > We'll soon do benchmarks of sdbox vs mdbox over cephfs with bluestore > > backend. > > We'll have to do some some work on how to simulate user traffic, for > writes > > and readings. That seems troublesome. > I would appreciate seeing these results ! > > > Thanks for the plugins recommendations. I'll take the change and ask you > > how is the SIS status? We have used it in the past and we've had some > > problems with it. > > I am using it since Dec 2016 with mdbox, with no issue at all (I am > currently using Dovecot 2.2.27-3 from Debian Stretch) > The only config I use is mail_attachment_dir, the rest lies as default > (mail_attachment_min_size = 128k, mail_attachment_fs = sis posix, > ail_attachment_hash = %{sha1}) > The backend storage is a local filesystem, and there is only one Dovecot > instance > > > > > Regards, > > > > Webert Lima > > DevOps Engineer at MAV Tecnologia > > *Belo Horizonte - Brasil* > > *IRC NICK - WebertRLZ* > > > > > > On Wed, May 16, 2018 at 4:19 PM Jack <ceph(a)jack.fr.eu.org> wrote: > > > >> Hi, > >> > >> Many (most ?) filesystems does not store multiple files on the same > block > >> > >> Thus, with sdbox, every single mail (you know, that kind of mail with 10 > >> lines in it) will eat an inode, and a block (4k here) > >> mdbox is more compact on this way > >> > >> Another difference: sdbox removes the message, mdbox does not : a single > >> metadata update is performed, which may be packed with others if many > >> files are deleted at once > >> > >> That said, I do not have experience with dovecot + cephfs, nor have made > >> tests for sdbox vs mdbox > >> > >> However, and this is a bit out of topic, I recommend you look at the > >> following dovecot's features (if not already done), as they are awesome > >> and will help you a lot: > >> - Compression (classic, https://wiki.dovecot.org/Plugins/Zlib) > >> - Single-Instance-Storage (aka sis, aka "attachment deduplication" : > >> https://www.dovecot.org/list/dovecot/2013-December/094276.html) > >> > >> Regards, > >> On 05/16/2018 08:37 PM, Webert de Souza Lima wrote: > >>> I'm sending this message to both dovecot and ceph-users ML so please > >> don't > >>> mind if something seems too obvious for you. > >>> > >>> Hi, > >>> > >>> I have a question for both dovecot and ceph lists and below I'll > explain > >>> what's going on. > >>> > >>> Regarding dbox format (https://wiki2.dovecot.org/MailboxFormat/dbox) > >> when > >>> using sdbox, a new file is stored for each email message. > >>> When using mdbox, multiple messages are appended to a single file until > >> it > >>> reaches/passes the rotate limit. > >>> > >>> I would like to understand better how the mdbox format impacts on IO > >>> performance. > >>> I think it's generally expected that fewer larger file translate to > less > >> IO > >>> and more troughput when compared to more small files, but how does > >> dovecot > >>> handle that with mdbox? > >>> If dovecot does flush data to storage upon each and every new email is > >>> arrived and appended to the corresponding file, would that mean that it > >>> generate the same ammount of IO as it would do with one file per > message? > >>> Also, if using mdbox many messages will be appended to a said file > >> before a > >>> new file is created. That should mean that a file descriptor is kept > open > >>> for sometime by dovecot process. > >>> Using cephfs as backend, how would this impact cluster performance > >>> regarding MDS caps and inodes cached when files from thousands of users > >> are > >>> opened and appended all over? > >>> > >>> I would like to understand this better. > >>> > >>> Why? > >>> We are a small Business Email Hosting provider with bare metal, self > >> hosted > >>> systems, using dovecot for servicing mailboxes and cephfs for email > >> storage. > >>> > >>> We are currently working on dovecot and storage redesign to be in > >>> production ASAP. The main objective is to serve more users with better > >>> performance, high availability and scalability. > >>> * high availability and load balancing is extremely important to us * > >>> > >>> On our current model, we're using mdbox format with dovecot, having > >>> dovecot's INDEXes stored in a replicated pool of SSDs, and messages > >> stored > >>> in a replicated pool of HDDs (under a Cache Tier with a pool of SSDs). > >>> All using cephfs / filestore backend. > >>> > >>> Currently there are 3 clusters running dovecot 2.2.34 and ceph Jewel > >>> (10.2.9-4). > >>> - ~25K users from a few thousands of domains per cluster > >>> - ~25TB of email data per cluster > >>> - ~70GB of dovecot INDEX [meta]data per cluster > >>> - ~100MB of cephfs metadata per cluster > >>> > >>> Our goal is to build a single ceph cluster for storage that could > expand > >> in > >>> capacity, be highly available and perform well enough. I know, that's > >> what > >>> everyone wants. > >>> > >>> Cephfs is an important choise because: > >>> - there can be multiple mountpoints, thus multiple dovecot instances > on > >>> different hosts > >>> - the same storage backend is used for all dovecot instances > >>> - no need of sharding domains > >>> - dovecot is easily load balanced (with director sticking users to the > >>> same dovecot backend) > >>> > >>> On the upcoming upgrade we intent to: > >>> - upgrade ceph to 12.X (Luminous) > >>> - drop the SSD Cache Tier (because it's deprecated) > >>> - use bluestore engine > >>> > >>> I was said on freenode/#dovecot that there are many cases where SDBOX > >> would > >>> perform better with NFS sharing. > >>> In case of cephfs, at first, I wouldn't think that would be true > because > >>> more files == more generated IO, but thinking about what I said in the > >>> beginning regarding sdbox vs mdbox that could be wrong. > >>> > >>> Any thoughts will be highlt appreciated. > >>> > >>> Regards, > >>> > >>> Webert Lima > >>> DevOps Engineer at MAV Tecnologia > >>> *Belo Horizonte - Brasil* > >>> *IRC NICK - WebertRLZ* > >>> > >>> > >>> > >>> _______________________________________________ > >>> ceph-users mailing list > >>> ceph-users(a)lists.ceph.com > >>> http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com > >>> > >> > >> _______________________________________________ > >> ceph-users mailing list > >> ceph-users(a)lists.ceph.com > >> http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com > >> > > > >

1 1

Re: Bug reports - auth is broken in Dovecot 2.3.3
by Berindeie Avram-Teodor 04 Oct '18

04 Oct '18

> > I do not have downloaded the source from GitHub. > > On Wed, Oct 3, 2018 at 1:53 PM Timo Sirainen <tss(a)iki.fi> wrote: > >> On 3 Oct 2018, at 13.22, Berindeie Avram-Teodor <berindeie.teo(a)gmail.com> >> wrote: >> > >> > I applied the patch and recompiled but nothing resolved. >> > What else can I do? >> >> You also need to run autogen.sh & configure again. >> >>

4 5

Re: Dovecot User Listing Error - getpwent() failed: Invalid Argument
by Reuben Farrelly 04 Oct '18

04 Oct '18

Sure: https://www.reub.net/files/dovecot/lightning-dovecot.conf https://www.reub.net/files/dovecot/thunderstorm-dovecot.conf Updated nightly. Reuben On 1/09/2018 12:26 am, Aki Tuomi wrote: > Can you provide doveconf -n? > > > > --- > Aki Tuomi > Dovecot oy > > -------- Original message -------- > From: Reuben Farrelly <reuben-dovecot(a)reub.net> > Date: 31/08/2018 17:12 (GMT+02:00) > To: Aki Tuomi <aki.tuomi(a)dovecot.fi>, Dovecot Mailing List > <dovecot(a)dovecot.org> > Subject: Re: Dovecot User Listing Error - getpwent() failed: Invalid > Argument > > No. Neither of those are installed on either system. > > Reuben > > On 1/09/2018 12:09 am, Aki Tuomi wrote: > > Could apparmor or selinux be causing this! > > > > > > > > --- > > Aki Tuomi > > Dovecot oy > > > > -------- Original message -------- > > From: Reuben Farrelly <reuben-dovecot(a)reub.net> > > Date: 31/08/2018 16:50 (GMT+02:00) > > To: Dovecot Mailing List <dovecot(a)dovecot.org> > > Subject: Dovecot User Listing Error - getpwent() failed: Invalid Argument > > > > Hi, > > > > I'm running dovecot-2.3 git and seeing on 2 of my Dovecot installations, > > the following message logged quite frequently: > > > > Aug 31 16:55:53 lightning.reub.net dovecot[7698]: auth-worker(7707): > > Error: getpwent() failed: Invalid argument > > Aug 31 16:55:53 lightning.reub.net dovecot[7698]: replicator: Error: > > User listing returned failure > > Aug 31 16:55:53 lightning.reub.net dovecot[7698]: replicator: Error: > > listing users failed, can't replicate existing data > > > > It appears that this could be having an effect on replication, but it's > > unclear because the secondary/replica doesn't see much traffic. > > > > Authentication is via PAM and the system is Gentoo Linux x86_64. passb > > driver = pam, userdb driver = static . > > > > doveadm user -u '*' shows the system users listed twice like this (is > > this a problem?) and then prints: > > > > ---- > > reuben > > liam > > reuben > > liam > > <some other users> > > Error: User listing returned failure > > Fatal: user listing failed > > ---- > > > > To me that doesn't look right. > > > > Can anyone suggest what could be causing this? I don't recall seeing > > this message some time ago so I suspect it's a recent change in either > > dovecot, or pam or glibc etc.. > > > > Thanks, > > Reuben

2 10

Authenticate users using their firstname
by Fady AL HAYALI 03 Oct '18

03 Oct '18

Hi, I'm setting up a Postfic and Dovecot with LDAP email server. My users in LDAP is like this: dn: uid=firstname,ou=People,dc=domain,dc=com uid: firstname uidNumber: 4025 gidNumber: 4025 givenName: firstname objectClass: top objectClass: person objectClass: posixAccount objectClass: shadowAccount objectClass: organizationalPerson objectClass: inetOrgPerson loginShell: /bin/bash homeDirectory: /home/firstname cn: firstname lastname mail: firstname.lastname(a)domain.com<mailto:firstname.lastname@domain.com> This is how I connect Dovecot with LDAP hosts = ldapserver ldap_version = 3 base = ou=People,dc=domain,dc=com deref = never scope = subtree user_attrs = user_filter = (&(objectclass=inetOrgPerson)(uid=%n) pass_attrs = uid=user,userPassword=password pass_filter = (&(objectclass=inetOrgPerson)(uid=%n)) default_pass_scheme = SSHA When I enter a user's email address and password as the following: email: firstname.lastname(a)domain.com<mailto:firstname.lastname@domain.com> password: password and according to my setting which I used "%n" as you see above, the username used to authenticate is "firstname.lastname". I checked the Dovecot variables but I couldn't find something useful in this case to manipulate the "%n" variable. I would like to keep using email addresses as "firstname.lastname(a)domain.com"<mailto:firstname.lastname@domain.com> but authenticate users using their first name. I really hit a wall here and any help will be much appreciated.

6 6

Possible architecture ?
by Alexandre Ellert 03 Oct '18

03 Oct '18

Hi, I'd like to achieve the following setup whit dovecot using multiple servers : - one server dedicated to all client IMAP (TLS) connections (i <http://mail.numeezy.com>map.mymaindomain.com, see below ) - each backend server has it's own local storage. no replication - each backend server responsible of a few domains - each backend server has it's own Mysql local database for user's passwords. ===> Server 1 : domains A, B and C ====> i <http://mail.numeezy.com>map.mymaindomain.com ===> Server 2 : domains D, E and F (143 TLS / 993 SSL) ===> Server 3 : domains G, H For example, if a user connects from domain E to i <http://mail.numeezy.com> map.mymaindomain.com, will Dovecot be able to use password database hosted on Server 2 ? Thank you ! Alex

3 3

Delete vs. Expunge in Public Namespace
by Chris 03 Oct '18

03 Oct '18

All, my IMAP client [1] marks messages as deleted, but doesn't expunge. Dovecot (2.2.27-3+deb9u2) expunges them. Is this because the mailbox is part of public namespace? .... ::1.38247-::1.00143: MALP6 STORE 48,49,50,51,52,53,54,55,56,57,58,59,60,61 +FLAGS (\Deleted) ::1.00143-::1.38247: * 48 FETCH (FLAGS (\Deleted \Seen)) * 49 FETCH (FLAGS (\Deleted \Seen)) * 50 FETCH (FLAGS (\Deleted \Seen)) * 51 FETCH (FLAGS (\Deleted \Seen)) * 52 FETCH (FLAGS (\Deleted \Seen)) * 53 FETCH (FLAGS (\Deleted \Seen)) * 54 FETCH (FLAGS (\Deleted \Seen)) * 55 FETCH (FLAGS (\Deleted \Seen)) * 56 FETCH (FLAGS (\Deleted \Seen)) * 57 FETCH (FLAGS (\Deleted \Seen)) * 58 FETCH (FLAGS (\Deleted \Seen)) * 59 FETCH (FLAGS (\Deleted \Seen)) * 60 FETCH (FLAGS (\Deleted \Seen)) * 61 FETCH (FLAGS (\Deleted \Seen)) MALP6 OK Store completed. ::1.38247-::1.00143: MALP7 SEARCH DELETED .... Oct 2 10:18:03 hyptest dovecot: imap(chris): delete: box=MailingListen/test125, uid=180, msgid=<740a9ba3-d637-3fa0-1b87-ff8d79491bf0@ hyp.postbox.xyz>, size=1509 Oct 2 10:18:03 hyptest dovecot: imap(chris): expunge: box=MailingListen/test125, uid=167, msgid=<740a9ba3-d637-3fa0-1b87-ff8d79491bf0 @hyp.postbox.xyz>, size=1509 namespace { list = yes location = maildir:/var/vmail/public/:LAYOUT=fs:INDEXPVT=%h/.public-indexes prefix = MailingListen/ separator = / subscriptions = no type = public } - Chris [1] https://raw.githubusercontent.com/quentinsf/IMAPdedup/master/imapdedup.py

2 3

Bug reports - auth is broken in Dovecot 2.3.3
by Berindeie Avram-Teodor 03 Oct '18

03 Oct '18

I applied the patch and recompiled but nothing resolved. What else can I do?

3 3

Bug reports - auth is broken in Dovecot 2.3.3
by Berindeie Avram-Teodor 03 Oct '18

03 Oct '18

Hello, I have compiled Dovecot 2.3.3 from sources but I have problems with authentication. I tried to rebuild Dovecot 2.3.2.1 but the error persists. I attach a .txt file with the information requested to report a bug. My main mail server is fallen, (the same for the test server) I will send the report from a Gmail address. If you need additional information I will be at your disposal.

2 1

Need to convert mbox to Maildir
by Mark Foley 02 Oct '18

02 Oct '18

I have a mbox file of emails. I want to convert this to Maildir giving me individual message files per email. I've looked at dsync, but as far as I can tell this wants a specific target user and it appears that it will "distribute" the converted messages into that user's INBOX. I don't want to put these mbox messages into any particular user's Maildir hierarchy, just export to file-per-message format to a destination directory of my choosing. Is this possible? THX --Mark

2 2

Help for UID THREAD and SORT optimization
by Alessio Cecchi 02 Oct '18

02 Oct '18

Hello, we are developing a library to show last arrived messages of all threads in a folder with more than 300k messages. As per the RFC 5256, the IMAP thread command has only the option to specify the grouping criteria (REFERENCES vs ORDEREDSUBJECT). So we implemented an algorithm which gets the full UID SORT ordered by date in reverse order then gets all thread relations and then post process all outputs to obtain the final result. Once we have both output we loop through the first uids set (which is ordered) and for every item of it we loop through the other uids set with all the thread relations. If the uid of the first set is contained in one of the elements of the thread messages, we pick it up and push it in the resulting array. This is a pseudo-code that shows what we are doing: array_of_latest_messages_in_thread = [] array_of_sorted_uids = [n,n,n,n,...] // UID SORT (REVERSE DATE) UTF-8 ALL array_of_thread_uids = [n,[n,n],n,[n,n,n],[n,n],n,n,...] // UID THREAD (REFERENCES) UTF-8 ALL foreach(array_of_sorted_uids as s_uid){ foreach(array_of_thread_uids as t_uids){ if(t_uids contains s_uid){ // or a function to loop recursively t_uids to search if a leaf is equal to s_uid array_of_latest_messages_in_thread.push(s_uid) break } } } We have also made some little optimizations in the above code like for example skipping from the outer loop the uids already processed in the inner loop that have not being selected. Anyway, the loop is very expensive in term of computation and we are wondering if there is a better approach to this issue directly using IMAP instead of post processing the output of the two commands at application level. Thanks. -- Alessio Cecchi Postmaster @ http://www.qboxmail.it https://www.linkedin.com/in/alessice

2 2