- dovecot - dovecot.org

subscriptions via master user
by Przemysław Kwiatkowski 25 May '23

25 May '23

Hello, When I impersonate an regular user using master user (log in to Roundcube as "user*masteruser") - there are user's subscriptions missing. I can see all folders, but have to manually enable visibility in settings. Is this normal? Can I fix it? passdb { driver = passwd-file master = yes args = /filepath pass = yes } passdb { driver = passwd-file args = /filepath } userdb { driver = passwd-file args = /filepath default_fields = uid=vmail gid=vmail system_groups_user=vmail home=/srv/vmail/%d/%n master_user=%u } -- Przemyslaw Kwiatkowski

1 0

fts crash
by Przemysław Kwiatkowski 21 May '23

21 May '23

Hello, I started full reindex of whole emails base (doveadm index -A '*') and solr crashed due to out of memory. I had plenty of such messages: doveadm(gzt): Error: fts_solr: Indexing failed: Connection lost: read((conn:127.0.0.1:8983,id=621)) failed: EOF (Request queued 40.832 secs ago, 1 send attempts in 39.359 secs, 0.002 in http ioloop, 0.000 in other ioloops, connected 704.138 secs ago) doveadm(gzt): Error: Mailbox Sent: Precache for UID=4036 failed: Internal error occurred. Refer to server log for more information. [2023-05-20 06:01:40] doveadm(gzt): Error: Mailbox Sent: Transaction commit failed: FTS transaction commit failed: backend deinit I increased solr memory and restarted it. Do I have to restart dovedam index? The "request queued" message suggest that Dovecot is smart and I don't need to do anything. Am I right? (Any way to confirm that indeed everything was indexed?) -- Przemyslaw Kwiatkowski

2 1

Re: LMTP server client certificate validation
by Jochen Bern 19 May '23

19 May '23

On 16.05.23 14:27, Sean Gallagher wrote: > I have a created a CA for the sole purpose of signing the certificate of > the LMTPS client. I regard this as a horrible horrible kludge. ... how do you figure that? *Someone*, and that means you, has to specify which clients are allowed to connect. Which leaves three possibilities: a) You run the CA and thus, the CA can do the selection for you. Which means that in order to deal with a single-permissible-client situation, you need the CA to issue client certs to just that one entity. b) The CA does *not* and issues (client) certs to entities you do *not* want to grant access. Then the selection job remains yours and, assuming that requiring a client cert from that CA makes any sense in the first place, auth_ssl_username_from_cert is there to help you (by making sure that the username will be "the ID" from the cert so that you can base some additional filtering on that). c) The CA is willing to enrich the certs it issues with DN parts, X509v3 extensions or the like signalling access rights as granted beforehand by a *fourth* party, as in, you. Which is a terrible can of worms as soon as things, especially the number of such fourth parties to support by the same CA, start to scale up IMHO. Which promptly brings us back to you running the CA yourself ... Kind regards, -- Jochen Bern Systemingenieur Binect GmbH

3 5

Possible hack via doveadm
by Daniel L. Miller 18 May '23

18 May '23

I currently have two servers running with synchronization that I've been very happy with. However, I'm seeing the following in my logs recently: May 12 15:45:58 cloud1 dovecot: doveadm(194.165.16.78): Error: doveadm client not compatible with this server (mixed old and new binaries?) May 13 03:44:31 cloud1 dovecot: doveadm(45.227.254.48): Error: doveadm client not compatible with this server (mixed old and new binaries?) Since I don't recognize those IPs, the first is out of Panama and the other is Belize, I assume these are hostile attackers trying to exploit something. How can I defend against this? -- Daniel

5 6

pigeonhole: sieve: float score failed for comparator-i ascii-numeric
by Randy Li 18 May '23

18 May '23

Hello Following the sieve_spamtest_status_header from the document, I think I could extract the score from the spamassassin (which is from spampd lmtp). But sieve script can't never work properly with comparator "i;ascii-numeric". I have used sieve-test several times, once I remove the dot from the mail sample, the sieve script works well. I noticed that document says i;ascii-numeric is for unsigned integer only which certainly does work for this case. The score here would be signed float(all right all the float is signed) and it is a float variable from tst-spamvirustest.c. I think that is the place the bug happened. Unfortunately, I didn't found any RFC that supports a float comparator and I am not familiar with the sieve interrupter. If someone could give me a possible method to fix this, I could submit a patch. Sincerely Randy Li

2 2

Re: Incorrect saved dates in mailboxes
by Joseph Tam 18 May '23

18 May '23

> OK, that does make sense so far. Today I have checked, and the correct > dates seem to be showing up now after the initial date (ever since that > initial run of the "dovecot expunge" command was run on all mailboxes). > So in our case I have the expunge set to delete older than 30 days. I > assume if I wait 30 days from now, it will start working? Yup. If you run your script every day (and thus, run "doveadm fetch ... date.saved" as well), that will make sure any new mail put into your Trash folder will have date.saved within 24h of the "real" value. Joseph Tam <jtam.home(a)gmail.com>

1 0

Re: Incorrect saved dates in mailboxes
by Joseph Tam 18 May '23

18 May '23

Chris Szilagyi <chris(a)apex-internet.com> writes: > Recently, I noticed that our expunge script is not working, and I don't > think it ever has on this server. On further inspection, it looks like > the saved date for emails in the folders we want to expunge is set to a > recent date (yesterday) for almost all messages. For example, when I > run the command "doveadm -f tab fetch -u username date.saved mailbox > Trash", the date for almost all of the messages is yesterday at the > exact same time. I tried this on other users and they have this exact > same timestamp shown for most of their messages, too. > > I tried checking for "date.received" and that shows correct, it is > "date.saved" that is not working. > > Any idea of how or why this would not be showing the correct date? I've > used this setup in the past on other servers with dovecot and it has > worked great, no idea why we would be seeing this issue now. I recall having the same problem. I think the "date.saved" is not instantiated in the cache until you query for it. So the value you were shown is when you last dumped its value if it wasn't previously set -- your run of same values coincided when you ran "doveadm fetch". My expunge script just uses date.received instead -- it seems to work. Joseph Tam <jtam.home(a)gmail.com>

2 1

[PATCH] [WIP]: thunderbird mboxrd to mbox
by Randy Li 16 May '23

16 May '23

I was trying to restore email from local thunderbird download data. None of the script I found could work. This script is modified from superuser.com/questions/105653/how-to-convert-mboxo-mboxrd-to-mboxcl-mboxcl2 Which works for me, using this script I could use doveadm import to restore my INBOX(I don't know how to restore the Sent mailbox). If this script is useful, I would rewrite one for python3 and following the mboxrd spec. Signed-off-by: Randy Li <ayaka(a)soulik.info> --- mailman2mbox.py | 111 ++++++++++++++++++++++++++++++++++++++++++++++++ 1 file changed, 111 insertions(+) create mode 100755 mailman2mbox.py diff --git a/mailman2mbox.py b/mailman2mbox.py new file mode 100755 index 0000000..1a154b7 --- /dev/null +++ b/mailman2mbox.py @@ -0,0 +1,111 @@ +#!/usr/bin/env python2 +""" +Usage: ./mailman2mbox.py infile outfile default-to-address +""" +import sys +from time import strftime,strptime,mktime,asctime +from email.utils import parseaddr,formatdate + +if len(sys.argv) not in (3,4): + print __doc__ + sys.exit() + +out = open(sys.argv[2],"w") +listid = None +if len(sys.argv)==4: + listid = sys.argv[3] + +date_patterns = ("%b %d %H:%M:%S %Y", "%d %b %H:%M:%S %Y", "%d %b %Y %H:%M:%S", "%d %b %H:%M:%S", "%d %b %y %H:%M:%S", "%d %b %Y %H.%M.%S",'%m/%d/%y %H:%M:%S %p') + +class HeaderError(TypeError): + pass + + +def finish(headers, body): + body.append("\n") + for n,ln in enumerate(headers): + if ln.startswith("Date:"): + break + else: + raise HeaderError("No 'Date:' header:\n" + "".join(headers)+"\n") + if listid is not None: + for ln2 in headers: + if ln2.lower().startswith("list-id:"): + break + else: + headers.append("List-Id: <%s>\n" % (listid,)) + date_line = ln[5:].strip() + if date_line.endswith(')'): + date_line = date_line[:date_line.rfind('(')].rstrip() + if date_line[-5] in "+-": + date_line, tz = date_line[:-5].rstrip(), int(date_line[-5:])//100 + else: + tz = -5 + if date_line[:3] in ("Mon","Tue","Wed","Thu","Fri","Sat","Sun"): + if date_line[3:5] == ', ': + prefix = "%a, " + elif date_line[3] == ',': + prefix = "%a," + else: + prefix = "%a " + else: + prefix = "" + while True: + for p in date_patterns: + try: + date_struct = strptime(date_line, prefix+p) + except ValueError: + pass + else: + break + else: + if not date_line: + raise ValueError(headers[n]) + date_line = date_line[:date_line.rfind(' ')] + continue + break + + date_struct = list(date_struct) + try: + headers[n] = 'Date: %s\n' % (formatdate(mktime(date_struct),True)) + headers[0] = "From %s %s\n" % (headers[0][:-25].rstrip(), asctime(date_struct), ) + except ValueError: + raise ValueError(headers[n]) + + for w in headers, body: + for s in w: + out.write(s) + + +message = 0 +headers, body = None, None +for line in open(sys.argv[1]): + if line.startswith("From "): + message+=1 + header = True + if headers is not None: + try: + finish(headers, body) + except HeaderError: + message -= 1 + out.write('>') + for w in headers, body: + for s in w: + out.write(s) + headers, body = [], [] + line = line.replace(" at ", "@") + elif line == '\n': + header = False + elif header and line.startswith('From:'): + line = line.replace(" at ","@") + (headers if header else body).append(line) + +try: + finish(headers, body) +except HeaderError: + out.write('>') + for w in headers, body: + for s in w: + out.write(s) + +out.close() -- 2.40.1

2 1

Incorrect saved dates in mailboxes
by Chris Szilagyi 16 May '23

16 May '23

Hello: We have a server (Debian 11) using Dovecot 2.3 which has been in place for about a year. We are using mbox format for our mailboxes, and do not use MySQL, just plain user directories under /home. Recently, I noticed that our expunge script is not working, and I don't think it ever has on this server. On further inspection, it looks like the saved date for emails in the folders we want to expunge is set to a recent date (yesterday) for almost all messages. For example, when I run the command "doveadm -f tab fetch -u username date.saved mailbox Trash", the date for almost all of the messages is yesterday at the exact same time. I tried this on other users and they have this exact same timestamp shown for most of their messages, too. I tried checking for "date.received" and that shows correct, it is "date.saved" that is not working. Any idea of how or why this would not be showing the correct date? I've used this setup in the past on other servers with dovecot and it has worked great, no idea why we would be seeing this issue now. Thanks in advance. Chris

2 1

[feature request] SSL handshake rejection for non-SNI clients
by Serg 16 May '23

16 May '23

I would like to offer to implement a feature to reject SSL handshakes for a default certificate-key pair for efficiently discarding bot requests (i.e. such requests that provide invalid/not configured hostname or do not specify at all, like when doing request to the IP address directly). Nginx has such feature already implemented as seen here[1], and it would be beneficial if dovecot would support this too. Currently I am using the following SSL configuration snippet to mimic such behavior: > ssl_cert = </etc/ssl/dovecot/server.crt > ssl_key = </etc/ssl/dovecot/server.key > > local_name flopster.at.encryp.ch { > ssl_cert = </etc/ssl/domains/flopster.at.encryp.ch/fullchain > ssl_key = </etc/ssl/domains/flopster.at.encryp.ch/key > } But in this case the problem is that the invalid requests (for this example it is requests that don't have Server Name Indication at all or mention anything else but not flopster.at.encryp.ch) are still being replied by Dovecot with a TLS certificate rather than being simply rejected with a TLSV1_UNRECOGNIZED_NAME error code. [1]: <https://nginx.org/en/docs/http/ngx_http_ssl_module.html#ssl_reject_handshake>

3 3