- Users - KICTANet

Roundcube and password plugin
by Helmut Fritz 05 Jan '17

05 Jan '17

Hello All, I am trying to get the Roundcube password plugin working, but I am no SQL guy and I need to know what to put for the password query line in the password/config.inc.php (I think!). It currently contains: $config['password_query'] = 'SELECT update_passwd(%c, %u)'; I understand that the plugin needs to connect to the exim4u database and change the users password there. Please correct me if I am wrong; I do have the dsn configured. The error I get is that the FUNCTION does not exist: [04-Jan-2017 19:11:21 -0800]: <ggorh7v1> DB Error: [1305] FUNCTION exim4u.update_passwd does not exist (SQL Query: SELECT update_passwd('$1$mgJJ/WAK$hSmBBmaC.Tpv/2ACIZsmq.', 'user(a)email.com')) in /us r/home/exim4u/public_html/webmail/program/lib/Roundcube/rcube_db.php on line 543 (POST /webmail/?_task =settings&_action=plugin.password-save?_task=&_action=) >From my research I have determined that the query line needs to be correct to change the password in the exim4U database. I did find one thread in the archive: http://users.exim4u.org/Which-Webmail-How-do-users-change-passwords-tp402362 4.html But the resolution was not posted. Thx for any help. Helmut

3 5

Re: [Exim4U] help with DEFER errors
by Helmut Fritz 13 Jul '16

13 Jul '16

Not sure if it is related, but my /var/spool/exim/scan and /var/spool/exim/input directories have a lot of files in them. Input: 2223 files Scan: 2223 files I am sure it is not a coincidence. The queue is empty. The files names definitely seem to correspond between the two directories. All the files in inpur are -D files (I believe that means body, -H is header). Interestingly there are n o corresponding -H header files in input. I was notified of this issue by a client who has not received any emails from the below mentioned mx.sendinghost.com [xxx.xxx.xxx.xxx]. They do get a reject message on their end of: Generating server: mx.sendinghost.com client(a)hisdomain.com #< #4.4.2 X-Proprietary; lost connection with my.host.com [xxx.xxx.xxx.xxx] while sending end of data -- message may be sent more than once> #SMTP# I did run a: exim -bh xxx.xxx.xxx.xxx and it ends up at an ACCEPT. From: Helmut Fritz [mailto:helmut(a)fritz.us.com] Sent: Monday, July 11, 2016 12:48 PM To: 'Exim4U General Discussion' Subject: help with DEFER errors Guys, 1. I have a lot of these type of messages in my mainlog: H=(mx.sendinghost.com) [xxx.xxx.xxx.xxx] Warning: ACL "warn" statement skipped: condition test deferred: failed to expand ACL string "${lookup dnsdb{a=$smtp_active_hostname}}": lookup of "a=my.host.com" gave DEFER: I am trying to look through the configs to see if I can determine where and why I am getting these. Are they a cause for any real concer? What causes them and can it be recified? I am not running a multi-ip host, but do host multiple virtual domains. 2. I am not sure if it is related (I think not), but I also get messages that correspond with the sending host that is in some of the above messages but the emails do NOT seem to come through to the users mailbox. I (think I) do see the actual emails in the /var/spool/exim/scan directory though. The only thing I find in the logs is this message: 1b4BZC-0005p9-Ek H= mx.sendinghost.com [xxx.xxx.xxx.xxx] Warning: spam-score-int: 1 (/). spamreject: 990. Any help, ideas or thoughts would be welcome. I did a google search that did not seem to provide an useful help. Thx. Helmut P.S. not yet updated to latest release of exim4u.

2 7

help with DEFER errors
by Helmut Fritz 12 Jul '16

12 Jul '16

Guys, 1. I have a lot of these type of messages in my mainlog: H=(mx.sendinghost.com) [xxx.xxx.xxx.xxx] Warning: ACL "warn" statement skipped: condition test deferred: failed to expand ACL string "${lookup dnsdb{a=$smtp_active_hostname}}": lookup of "a=my.host.com" gave DEFER: I am trying to look through the configs to see if I can determine where and why I am getting these. Are they a cause for any real concer? What causes them and can it be recified? I am not running a multi-ip host, but do host multiple virtual domains. 2. I am not sure if it is related (I think not), but I also get messages that correspond with the sending host that is in some of the above messages but the emails do NOT seem to come through to the users mailbox. I (think I) do see the actual emails in the /var/spool/exim/scan directory though. The only thing I find in the logs is this message: 1b4BZC-0005p9-Ek H= mx.sendinghost.com [xxx.xxx.xxx.xxx] Warning: spam-score-int: 1 (/). spamreject: 990. Any help, ideas or thoughts would be welcome. I did a google search that did not seem to provide an useful help. Thx. Helmut P.S. not yet updated to latest release of exim4u.

2 4

fix for anyone having trouble with passwords with a $ in them
by Chuck Cochems 16 Jun '16

16 Jun '16

in config/functions.php at line 10 foreach ($_POST as $postkey => $postval) { $_POST[$postkey] = preg_replace('/[\'";$%]/','',$postval); } replace with foreach ($_POST as $postkey => $postval) { if (($postkey <> "crypt") && ($postkey <> "clear")) $_POST[$postkey] = preg_replace('/[\'";$%]/','',$postval); } this is for exim4u current stable.

1 0

php7 and exim4u.
by Chuck Cochems 15 Jun '16

15 Jun '16

horde webmail doesn't work at all. exim4u itself seems to work fine. anyone have a replacement that's php7 compatible for webmail?

4 4

Exim4U Version 3.1.0 Released
by Gordon Dickens 14 Jun '16

14 Jun '16

1 0

Exim4u and DMARC
by Kreuder 19 May '16

19 May '16

Hi everyone, I'm currently giving DMARC a try and I'm wondering how to use it within my setup. The domains that are hosted on my exim4u based installation all use mail.foo.tld as their imap and smtp server. That is because I require my customers to use SSL/TLS connections and this way I only have to manage one central certificate. All domains have their SPF records and hostname.foo.tld has a valid DKIM record. So far so good. But: DMARC (as far as I understand the whole process) seems to check each _domain_'s DKIM, right? And the way exim4u works (again: as far as I understand it) it's the server's DKIM that used to sign outgoing mail. I signed up for dmarcian.com's DMARC reports -a service that collects and analyzes your DMARC reports- and it tells me that all domains (but foo.tld) lack DKIM signature. I've set the DMARC policy to "none" for every domain so that shouldn't be a major problem for now. Still I'm wondering if there's a way to setup exim4u to sign mails using the domain's DKIM, not the server's. Has anyone experience using DMARC? With or without exim4u? Am I missing something? Any tips or hints are highly appreciated. thanks, Mika

4 4

Best Place for manual blacklisting of senders/sender_domains
by Odhiambo Washington 03 May '16

03 May '16

Right now I am using system filter to deal with some bad senders and sender_domains, but I am wondering if we can have such functionality included in the main exim.conf with the option of adding the sender addresses and sender domains in files in /etc/exim/some-files. I love keeping my exim.conf as close to the one in the distro as possible, but for some reason I don't see the most appropriate place in exim.conf to insert the rules, such as: deny message = Users have complained that you send unsolicited mail.\n\You're therefore blacklisted.\n\ Please contact abuse(a)crownkenya.com senders = ${if exists {/etc/exim/blacklisted_senders}{/etc/exim/blacklisted_senders}{}} Thanks guys and have a great week. -- Best regards, Odhiambo WASHINGTON, Nairobi,KE +254 7 3200 0004/+254 7 2274 3223 "Oh, the cruft."

3 6

exim startup and runtime error
by Helmut Fritz 21 Apr '16

21 Apr '16

Anyone else getting this: 2016-04-20 11:54:06 WARNING: purging the environment. Suggested action: use keep_environment and add_environment. That was at startup, it also seems I get thrown at during runtime (probably moving an email through the system): 2016-04-20 10:41:38 WARNING: purging the environment. Suggested action: use keep_environment and add_environment. 2016-04-20 10:41:38 Start queue run: pid=23845 2016-04-20 10:41:38 End queue run: pid=23845 It seems it is related to a couple of new variables introduced in 4.84 after a local-to-the-system vulnerability was found, keep_environment and add_environment. But when I find some related lines as suggestions and put them into the exim.conf, exim will not start and throws: option "keep_environment" unknown stating the line in the config file that entry is on. I am definitely running 4.86_2. Thoughts? Ideas to fix? Thx. Helmut

2 2

are these being delivered
by Helmut Fritz 20 Apr '16

20 Apr '16

Or dropped? This is from mainlog: 2016-04-20 12:31:08 1asxpu-0000Is-Dh DKIM: d=reply1.ebay.com s=emarsys2007 c=relaxed/relaxed a=rsa-sha1 i=eBay(a)reply1.ebay.com [verification succeeded] 2016-04-20 12:31:10 1asxpu-0000Is-Dh H=e3uspmta148.emarsys.net [91.194.248.148] Warning: spam-score-int: 27 (++). spamreject: 100. There is nothing corresponding in rejectlog, at least I cannot find anything. searched by the d= address and the H= address and i.p. Since there is no to: address, I cannot ask an end user. I do find this in the /var/logs/maillog with a corresponding time stamp: Apr 20 12:31:08 svr6 spamd[1097]: spamd: connection from localhost [127.0.0.1]:49471 to port 783, fd 6 Apr 20 12:31:08 svr6 spamd[1097]: spamd: setuid to spamd succeeded Apr 20 12:31:08 svr6 spamd[1097]: spamd: checking message <0.0.26.9A4.1D19AFD842264CA.0(a)e3uspmta148.emarsys.net> for spamd:58 Apr 20 12:31:10 svr6 spamd[1097]: spamd: identified spam (2.7/0.0) for spamd:58 in 1.8 seconds, 60322 bytes. Apr 20 12:31:10 svr6 spamd[1097]: spamd: result: Y 2 - DATE_IN_PAST_06_12,DCC_CHECK,HTML_MESSAGE,MONEY_BACK,RCVD_IN_DNSWL_NONE,RCVD _IN_RP_SAFE scantime=1.8,size=60322,user=spamd,u id=58,required_score=0.0,rhost=localhost,raddr=127.0.0.1,rport=49471,mid=<0. 0.26.9A4.1D19AFD842264CA.0(a)e3uspmta148.emarsys.net>,autolearn=no autolearn_force=no but nothing about the disposition of the email? Thx for any help, or pointing me to the right place to research/ask questions.

1 0