Re: Migrate mail_location

[Aki Tuomi]

> On 11/01/2022 08:10 he...@reox.at wrote:
> 
>  
> On 05.01.2022 09:34, Aki Tuomi wrote:
> > 
> >> On 03/01/2022 10:51 he...@reox.at wrote:
> >>
> >>   
> >> Hi!
> >>
> >> I have a dovecot installation, with the following configuration:
> >> mail_home = /var/vmail/%d/%n
> >> mail_location = maildir:~
> >>
> >> I would like to migrate the mail_location to maildir:~/mail, to get
> >> around the issues described in the documentation.
> >> But I can not really find a good way to do that.
> >> I played around with doveadm sync/backup, but I can only copy everything
> >> into a new location (i.e. doveadm sync maildir:~/mail), but this does
> >> not really untangle the mail_home and location.
> >> My other idea was to migrate the maildir manually, i.e. move the maildir
> >> related folders and files (new, cur, tmp, all folders starting with a
> >> dot and containing itself {new,cur,tmp}, dovecot.* files, subscriptions
> >> file) into the ~/mail folder, but I'm not sure if this is a safe method.
> >>
> >> Is there any safe method to do this using dovecot tools? If the only way
> >> is a offline migration, I'm fine with that.
> >>
> >> Best,
> >> Sebastian
> > 
> > Hi!
> > 
> > Since you're using maildir, you can just stop dovecot, move the files, and 
> > start dovecot.
> > 
> > In particular, you should **not** move .dovecot.sieve directory, or 
> > dovecot-attributes (if you have used mail_attribute_dict).
> > 
> > It's a good idea to do testing.
> > 
> > Aki
> 
> Hi!
> Thanks for the answer! Just to confirm: that means the only files which 
> will stay in mail_home are all sieve related files and folders, 
> dovecot-attributes (I do not use this currently though) and the 
> xapian-indexes folder (I use fts-xapian)? Thus the files 
> dovecot{.index,.index.cache,.index.log,.index.thread,-keywords,.list.index,.mailbox.log,-uidlist,-uidvalidity,-uidvalidity.*}
>  
> all belong to the maildir?
> For most of them I'm pretty sure they belong to the maildir (if not 
> otherwise configured).
> Must I run some re-indexing command after migration, for example also 
> for the FTS?
> 
> I'll test that on a test server before but it is also good to know what 
> files belong where.
> 
> Sebastian

If you do the move correctly, there should be no need to reindex anything. I 
think the file list looks ok.

Aki

Re: Migrate mail_location

[hello]

On 05.01.2022 09:34, Aki Tuomi wrote:



On 03/01/2022 10:51 he...@reox.at wrote:

  
Hi!


I have a dovecot installation, with the following configuration:
mail_home = /var/vmail/%d/%n
mail_location = maildir:~

I would like to migrate the mail_location to maildir:~/mail, to get
around the issues described in the documentation.
But I can not really find a good way to do that.
I played around with doveadm sync/backup, but I can only copy everything
into a new location (i.e. doveadm sync maildir:~/mail), but this does
not really untangle the mail_home and location.
My other idea was to migrate the maildir manually, i.e. move the maildir
related folders and files (new, cur, tmp, all folders starting with a
dot and containing itself {new,cur,tmp}, dovecot.* files, subscriptions
file) into the ~/mail folder, but I'm not sure if this is a safe method.

Is there any safe method to do this using dovecot tools? If the only way
is a offline migration, I'm fine with that.

Best,
Sebastian


Hi!

Since you're using maildir, you can just stop dovecot, move the files, and 
start dovecot.

In particular, you should **not** move .dovecot.sieve directory, or 
dovecot-attributes (if you have used mail_attribute_dict).

It's a good idea to do testing.

Aki


Hi!
Thanks for the answer! Just to confirm: that means the only files which 
will stay in mail_home are all sieve related files and folders, 
dovecot-attributes (I do not use this currently though) and the 
xapian-indexes folder (I use fts-xapian)? Thus the files 
dovecot{.index,.index.cache,.index.log,.index.thread,-keywords,.list.index,.mailbox.log,-uidlist,-uidvalidity,-uidvalidity.*} 
all belong to the maildir?
For most of them I'm pretty sure they belong to the maildir (if not 
otherwise configured).
Must I run some re-indexing command after migration, for example also 
for the FTS?


I'll test that on a test server before but it is also good to know what 
files belong where.


Sebastian

2.17.1 Fatal: master: service(doveadm): child 43879 killed with signal 6 (core dumped)

[Salatiel Filho]

Hi, my dovecot 2.17.1 instance is crashing a lot.
Quick question: The coredump should contain some private data, am I
correct ? Assuming so, should it be posted on a public maillist?

Panic: file dsync-ibc.c: line 166 (dsync_ibc_send_mail_request):
assertion failed: (request->guid != NULL || request->uid != 0)
...
Fatal: master: service(doveadm): child 43879 killed with signal 6 (core dumped)

Using gdb/bt  over the coredump I get:

Using host libthread_db library "/lib64/libthread_db.so.1".
Core was generated by `dovecot/doveadm-server'.
Program terminated with signal SIGABRT, Aborted.
#0  __GI_raise (sig=sig@entry=6) at ../sysdeps/unix/sysv/linux/raise.c:50
50  return ret;
(gdb) bt
#0  __GI_raise (sig=sig@entry=6) at ../sysdeps/unix/sysv/linux/raise.c:50
#1  0x7f612d382db5 in __GI_abort () at abort.c:79
#2  0x7f612db04485 in default_fatal_finish (status=0,
type=LOG_TYPE_PANIC) at failures.c:459
#3  fatal_handler_real (ctx=, format=,
args=) at failures.c:471
#4  0x7f612dbb10f7 in i_internal_fatal_handler (ctx=, format=, args=) at failures.c:872
#5  0x7f612db04126 in i_panic (format=format@entry=0x5641e4c16ba8
"file %s: line %d (%s): assertion failed: (%s)") at failures.c:524
#6  0x5641e4bcfcb2 in dsync_ibc_send_mail_request
(ibc=0x5641e50a5990, request=0x5641e51dcc70) at dsync-ibc.c:166
#7  0x5641e4bfe0bc in dsync_brain_send_mail_request
(brain=0x5641e50a8538) at dsync-brain-mails.c:215
#8  dsync_brain_sync_mails (brain=brain@entry=0x5641e50a8538) at
dsync-brain-mails.c:418
#9  0x5641e4bf9ad8 in dsync_brain_run_real
(changed_r=0x7ffcd2397a47, brain=0x5641e50a8538) at dsync-brain.c:716
#10 dsync_brain_run (changed_r=0x7ffcd2397a47, brain=0x5641e50a8538)
at dsync-brain.c:749
#11 dsync_brain_run (brain=0x5641e50a8538, changed_r=0x7ffcd2397a47)
at dsync-brain.c:737
#12 0x5641e4bfa13e in dsync_brain_run_io (context=0x5641e50a8538)
at dsync-brain.c:110
#13 0x5641e4c0b3c3 in dsync_ibc_stream_input (ibc=0x5641e50a5990)
at dsync-ibc-stream.c:232
#14 0x7f612dbc78ad in io_loop_call_io (io=0x5641e5040510) at ioloop.c:737
#15 0x7f612dbc8f19 in io_loop_handler_run_internal
(ioloop=ioloop@entry=0x5641e4f89ce0) at ioloop-epoll.c:222
#16 0x7f612dbc7950 in io_loop_handler_run (ioloop=0x5641e4f89ce0)
at ioloop.c:789
#17 0x7f612dbc7ad0 in io_loop_run (ioloop=0x5641e4f89ce0) at ioloop.c:762
#18 0x5641e4bddafa in cmd_dsync_run_remote (user=,
user=) at doveadm-dsync.c:496
#19 cmd_dsync_run (_ctx=0x5641e4f89fe8, user=) at
doveadm-dsync.c:703
#20 0x5641e4bdfc26 in doveadm_mail_next_user (ctx=0x5641e4f89fe8,
error_r=0x7ffcd2397de8) at doveadm-mail.c:462
#21 0x5641e4be0d74 in doveadm_mail_cmd_exec (wildcard_user=0x0,
ctx=0x5641e4f89fe8) at doveadm-mail.c:654
#22 doveadm_cmd_ver2_to_mail_cmd_wrapper (cctx=) at
doveadm-mail.c:983
#23 0x5641e4bf17c7 in doveadm_cmd_run_ver2 (argc=argc@entry=10,
argv=argv@entry=0x5641e4fa3788, cctx=cctx@entry=0x7ffcd2397f40) at
doveadm-cmd.c:465
#24 0x5641e4bf5d1a in doveadm_cmd_server_run_ver2
(cctx=0x7ffcd2397f40, argv=0x5641e4fa3788, argc=10,
conn=0x5641e4f81328) at client-connection-tcp.c:187
#25 doveadm_cmd_handle (cctx=0x7ffcd2397f40, argv=,
argc=, cmd_name=, conn=0x5641e4f81328)
at client-connection-tcp.c:214
#26 client_handle_command (args=, conn=0x5641e4f81328)
at client-connection-tcp.c:284
#27 client_connection_tcp_input (conn=0x5641e4f81328) at
client-connection-tcp.c:436
#28 0x7f612dbc78ad in io_loop_call_io (io=0x5641e4f88ee0) at ioloop.c:737
#29 0x7f612dbc8f19 in io_loop_handler_run_internal
(ioloop=ioloop@entry=0x5641e4f3df40) at ioloop-epoll.c:222
#30 0x7f612dbc7950 in io_loop_handler_run (ioloop=0x5641e4f3df40)
at ioloop.c:789
#31 0x7f612dbc7ad0 in io_loop_run (ioloop=0x5641e4f3df40) at ioloop.c:762
#32 0x7f612db39287 in master_service_run (service=0x5641e4f3dda0,
callback=) at master-service.c:863
#33 0x5641e4bd09f2 in main (argc=, argv=) at main.c:124

# 2.3.17.1 (476cd46418): /etc/dovecot/dovecot.conf
# Pigeonhole version 0.5.17.1 (a1a0b892)
# OS: Linux 4.18.0-348.7.1.el8_5.x86_64 x86_64 AlmaLinux release 8.5
(Arctic Sphynx) xfs
# Hostname: redacted
auth_username_format = %Ln
doveadm_password = # hidden, use -P to show it
first_valid_uid = 1000
mail_home = /var/spool/imap/%n
mail_location = maildir:/var/spool/imap/%n/mail:VOLATILEDIR=/var/spool/imap/%n
mail_plugins = " notify replication"
managesieve_notify_capability = mailto
managesieve_sieve_capability = fileinto reject envelope
encoded-character vacation subaddress comparator-i;ascii-numeric
relational regex imap4flags copy include variables body enotify
environment mailbox date index ihave duplicate mime foreverypart
extracttext
namespace {
  list = children
  location = 
maildir:/var/spool/imap/redacted/mail/:INDEXPVT=/var/spool/imap/%n/pub-seen-idx/
  mailbox * {
autoexpunge = 60 days
  }
  prefix = redacted/
  separator = /
  subscriptions = no
  type = public
}
namespace inbox {
  inbox = yes
  location =
  mailbox 

Re: Is Diffie-Hellman needed?

[Aki Tuomi]

> On 10/01/2022 18:12 dove...@ptld.com wrote:
> 
>  
> And follow up question;
> 
> The docs say you are encouraged to disable non-ECC DH algorithms completely.
> However i didn't see anything on that same page explaining how to go about 
> doing that.
> 
> Can someone point me to something explaining what that means and how to go 
> about doing it?

Bit hidden in the documentation, but 
https://doc.dovecot.org/installation_guide/upgrading/from-2.2-to-2.3/#diffie-hellman-parameters-for-ssl

Re: Is Diffie-Hellman needed?

[justina colmena ~biz]

I want better explanations of the maths.

If RSA and DSA algorithms based on standard arithmetic exponentiation modulo 
the product of two large primes are "deprecated" -- that means that there have 
been or are expected to be major mathematical and algorithmic advances in 
factoring large integers. The maths are easy for those algorithms, whereas the 
ECC algorithms are based on very advanced maths which aren't being explained 
satisfactorily to the general public, with $1,000,000 USD prizes still out for 
the so-called Birch and Swinnerton-Dyer conjecture and the Riemann Hypothesis, 
which might be more applicable to factoring the "semi-primes" of RSA/DSA/DH 
type algorithms.

On January 10, 2022 7:12:40 AM AKST, dove...@ptld.com wrote:
>And follow up question;
>
>The docs say you are encouraged to disable non-ECC DH algorithms completely.
>However i didn't see anything on that same page explaining how to go about 
>doing that.
>
>Can someone point me to something explaining what that means and how to go 
>about doing it?

--
Sent from my Android device with K-9 Mail. Please excuse my brevity.

Re: Is Diffie-Hellman needed?

[dovecot]

And follow up question;

The docs say you are encouraged to disable non-ECC DH algorithms completely.
However i didn't see anything on that same page explaining how to go about 
doing that.

Can someone point me to something explaining what that means and how to go 
about doing it?

Is Diffie-Hellman needed?

[dovecot]

Is Diffie-Hellman needed on a modern new dovecot setup?

ssl_dh is empty by default.
Both ssl_cert and ssl_key on my setup are pointing to let's encrypt certs.

However i see Diffie-Hellman related warnings in logs:
dovecot[1073]: imap-login: Error: Diffie-Hellman key exchange requested, but no 
DH parameters provided. Set ssl_dh=

Re: Spamassassin fails

[Benny Pedersen]

On 2022-01-10 08:41, Hans Brage wrote:


Spamassassin reports version 3.4.6


good

how is spamd configured ?

in gentoo i use spamd as follow options


SPAMD_OPTS="--allowed-ips=127.0.0.1 --min-children=1 --max-children=2 
--min-spare=1 --max-spare=2 --max-conn-per-child=100 --ipv4-only 
--sql-config --nouser-config --allow-tell --timing"

SPAMD_NICELEVEL=0
SPAMD_TIMEOUT=120
SPAMD_RUN_AS_ROOT=false

with that sql user prefs works

hope the remaining problems is ok