Language:
switch to room list switch to menu My folders
Go to page: First ... 43 44 45 46 [47] 48 49 50 51 ... Last
[#] Thu Apr 23 2015 09:02:33 EDT from mail" via citadel being sent as binary attachments @ Dog Pound BBS II

Subject: "cat <text file>

[Reply] [ReplyQuoted] [Headers] [Print]

I'm running into something that I find odd.

I'm running "Lynis" on my servers, at home.

At the end of the daily job (run via a script in /etc/cron.daily), it runs:

(/usr/local/lynis/lynis -c --cronjob > /var/log/lynis.log 2>&1
cat /var/log/lynis.log | mail -s "Daily Lynis Scan Report" <someone@somewhere)

The only mailer daemon on this system is Citadel

The lynis.log is a text file, but when it is emailed from my Citadel system to my other server, it comes across as a binary attachment. I've tested this manually, as well. Meanwhile emailing myself some other text file like, say, the hosts file, comes through as text when I run the same 'cat <file> | mail -s "some subject" <someone@somewhere>' command.

Needless to say, I'm a bit perplexed...is this known/expected behavior?



[#] Thu Apr 23 2015 10:36:24 EDT from paliasp @ Uncensored

[Reply] [ReplyQuoted] [Headers] [Print]

I enabled it. http://pastebin.com/raw.php?i=kLSH9LNJ ps: What does that mean?

[#] Thu Apr 23 2015 11:04:23 EDT from s3cr3to @ Uncensored

Subject: Old citadel/easyinstall with stuck mails ¿how to runqueue?

[Reply] [ReplyQuoted] [Headers] [Print]

I have an old Citadel version installed with the easyinstall in debian squeeze

WebCit 7.86, server build 33e4f20515571323671401780956558c414094f8

 

I have a load of mails that where not sent because the target/server was down (a virtual citadel repository) now is up again but if I try:

sendcommand SMTP runqueue
-bash: sendcommand: command not found

I wonder why sendcommand doesn't exists here... ¿is an external tool?

Exists some way to make the queue to run? maybe restarting citadel service (but I prefer to I ask first)

Regards



[#] Thu Apr 23 2015 12:16:09 EDT from s3cr3to @ Uncensored

Subject: Re: Old citadel/easyinstall with stuck mails ¿how to runqueue?

[Reply] [ReplyQuoted] [Headers] [Print]

Oh I see that citadel is attempting by itself... WOW!

Now I need to be patient for the queued mails to be send.

Thu Apr 23 2015 11:04:23 AM EDT from s3cr3to @ Uncensored Subject: Old citadel/easyinstall with stuck mails ¿how to runqueue?

I have an old Citadel version installed with the easyinstall in debian squeeze

WebCit 7.86, server build 33e4f20515571323671401780956558c414094f8

 

I have a load of mails that where not sent because the target/server was down (a virtual citadel repository) now is up again but if I try:

sendcommand SMTP runqueue
-bash: sendcommand: command not found

I wonder why sendcommand doesn't exists here... ¿is an external tool?

Exists some way to make the queue to run? maybe restarting citadel service (but I prefer to I ask first)

Regards



 



[#] Thu Apr 23 2015 15:53:03 EDT from vishalanm @ Uncensored

Subject: Re: citserver crashes

[Reply] [ReplyQuoted] [Headers] [Print]

Hi all ,

 

any feedback or options on the error

 



[#] Fri Apr 24 2015 14:30:20 EDT from jacksgt @ Uncensored

Subject: Webcit + Nginx Proxy

[Reply] [ReplyQuoted] [Headers] [Print]

Hi everybody!

 

I'm having an issue with Citadels Web Interface behind a Nginx Proxy.

My plan is the following:

- run the web interface internally (e.g. localhost:5050)

- proxy_pass with nginx domain.tld/webcit to localhost:5050

 

I have used the proposed configuration from the Knowledge Base:

http://www.citadel.org/doku.php/faq:installation:apacheproxy#how.can.i.install.webcit.so.it.runs.alongside.apache.nginx.lighthttpd.on.port.80.443

 

When I go to domain.tld/webcit it manages to load the base file (the index.html is there), but everything else (CSS, Images, etc.) fails with a 404.

 

Anyone using Webcit + Nginx here?



[#] Fri Apr 24 2015 16:25:45 EDT from s3cr3to @ Uncensored

Subject: Re: Webcit + Nginx Proxy

[Reply] [ReplyQuoted] [Headers] [Print]

This is from a test server that I setup some time ago that works, I hope stills works for you.

Is for secured connection with SSL I believe you can change it if you don't need ssl.

I take the full idea from some post here in Citadel Support, if you use the search function you'll see it.

The only issues that I found are:

  • I can't upload attachments using Firefox (sometimes works, many times no)
  • With newer versions of FF sometimes when sending mails it shows a message like "can't send empty mails" but not always. Maybe is some congestion on my external connection.
  • haven't tested with chrome and IE

Regards

I test in debian and this works:
Your ports may be different

*** /etc/default/webcit
export WEBCIT_APACHEFLAG=" -f "
export WEBCIT_LISTEN_IP=127.0.0.1
export WEBCIT_HTTP_PORT=33001 #60800
export WEBCIT_HTTPS_PORT=33443
export WEBCIT_LANG=UNLIMITED
export WEBCIT_CITADEL_PORT=504
export WEBCIT_CITADEL_IP=127.0.0.1

*** /etc/nginx/sites-available/default

server {
        ssl_protocols TLSv1.2 TLSv1.1 TLSv1;
        listen   80 default; ## listen for ipv4
        listen   [::]:80 default ipv6only=on; ## listen for ipv6
        listen   443 ssl;

        #http ssl_session_cache   shared:SSL:2m; #10m
        #http ssl_session_timeout 10m;
        keepalive_timeout   70;

        server_name localhost;
        server_name yourcitadel.domain.com;
        #ssl on;

        ssl_certificate /etc/ssl/webcit/citadel.cer;
        ssl_certificate_key /etc/ssl/webcit/citadel.key;

        access_log /var/log/nginx/citadel.access_log main;
        error_log /var/log/nginx/citadel.error_log info;
        #access_log /var/log/nginx/citadel.ssl_access_log main;
        #error_log /var/log/nginx/citadel.ssl_error_log info;


        location / {
                proxy_pass https://localhost:33443; #20001

                # allow anyone in
                allow   192.168.22.0/24; # our internal network
                allow   111.222.333.32/28; # our external network
                # me
                allow 1.2.3.4; # my home IP

                # drop rest of the world
                deny    all;

error_page 404 /e404.html;
location /404.html {
        internal;
}
error_page 403 /e403.html;
location /403.html {
        internal;
}

                valid_referers none blocked
                        yourcitadel.server.com 111.222.333.33 192.168.22.33 yourcitadel.server.lan
                        ;
                if ($invalid_referer) {
                        return 403;
                }
                proxy_set_header Host $host;
                proxy_set_header X-Real-IP $remote_addr;
                proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;

                proxy_redirect off;
        }

        location /static {
                # allow anyone in
                allow   192.168.22.0/24; # our internal network
                allow   111.222.333.32/28; # our external network
               
                allow 1.2.3.4; # my home IP

                # drop rest of the world
                deny    all;
               
               
                #root /usr/local/webcit;

                valid_referers none blocked
                        yourcitadel.server.com 111.222.333.33  192.168.22.33 yourcitadel.server.lan
                        ;
                        root /usr/share/citadel-webcit;
                }

                error_page 403 /403.html;
                location = /403.html {
                        proxy_pass https://localhost:33443; #20001
                        #root /var/www/html/403.html;
                        internal;
                        proxy_redirect off;
                }

                error_page 404 /403.html;
                #location = /403.html {
                #       proxy_pass http://localhost:33001;
                #       #root /var/www/html/403.html;
                #       internal;
                #       proxy_redirect off;
                #}

}

 

Fri Apr 24 2015 02:30:20 PM EDT from jacksgt @ Uncensored Subject: Webcit + Nginx Proxy

Hi everybody!

 

I'm having an issue with Citadels Web Interface behind a Nginx Proxy.

My plan is the following:

- run the web interface internally (e.g. localhost:5050)

- proxy_pass with nginx domain.tld/webcit to localhost:5050

 

I have used the proposed configuration from the Knowledge Base:

http://www.citadel.org/doku.php/faq:installation:apacheproxy#how.can.i.install.webcit.so.it.runs.alongside.apache.nginx.lighthttpd.on.port.80.443

 

When I go to domain.tld/webcit it manages to load the base file (the index.html is there), but everything else (CSS, Images, etc.) fails with a 404.

 

Anyone using Webcit + Nginx here?



 



[#] Fri Apr 24 2015 18:03:01 EDT from jacksgt @ Uncensored

Subject: Re: Webcit + Nginx Proxy

[Reply] [ReplyQuoted] [Headers] [Print]

Apologies for that weird double post, I have no idea how or why that happened.

 

@s3cr3to:

Thanks for your suggestion. I tried it and it works with webcit.domain.tld, however not with domain.tld/webcit (because it is fetching the static content from the wrong URL). I just get the plain HTML file.

Is it possible to tell Webcit that these files are located under a different sub-directory (in my case domain.tld/webcit/static ... )?

 

Note: It is not possible for me to locate Webcit at its own subdomain (even though I would love to do that), because my SSL Cert does not have wildcard.



[#] Fri Apr 24 2015 21:00:58 EDT from danda @ Uncensored

Subject: Citserver not running

[Reply] [ReplyQuoted] [Headers] [Print]

I installed Citadel on a Raspberry Pi about a year ago. It has been running great until a week or so ago. Now Citserver is not running. (at least it does not show up on a ps -eaf). No changes were made to the system at the time. It is a dedicated mail server. Could not find any error messages in any log files but I may not be looking in the right place.

When I try running citserver from a terminal I get: symbol lookup error: /usr/lib/arm-linux-gnueabihf/libdb-5.1.so: undefined symbol: en, version DB5_0

 

Any ideas?

I'm not a UNIX guy. Just playing with a home email server.

Thank You



[#] Fri Apr 24 2015 21:24:14 EDT from TaMeR @ Uncensored to citadel_support@citadel.org

Subject: text/pain turns in to Content-Type: multipart/mixed

[Reply] [ReplyQuoted] [Headers] [Print]

Hi,

I working on some code that will import in to citadel, markdown files
with toml or yaml headers (the kind people use to generate static html
pages)

I use imap to import, not port 504 on this project.
I am setting a header to "Content-Type: text/plain; charset=UTF-8"

Following is the source of the imported email.
As you can see it is turned in to "Content-Type: multipart/mixed;"

What could be the source of this behavior?
Why is it being tuned in to multipart?



Return-Path: Dennis T Kaplan
To: room_import@example.com
Date: Sat, 25 Apr 2015 00:57:19 -0000
Subject: Git Howto
Message-ID: <553AE66F-00001DB6@mail.riky.net>
From: "Dennis T Kaplan" <tamer@example.com>
Mime-Version: 1.0
Content-Type: multipart/mixed;
boundary=9b41f432623a90f2cfe46018fdd064553ce1b6cc8f70fdafd721e8b24647


--9b41f432623a90f2cfe46018fdd064553ce1b6cc8f70fdafd721e8b24647
Content-Type: multipart/alternative;
boundary=e0ac7962fb3da80d08c428e71edd15b066271d9b678f1fb4af0b372db28f

--e0ac7962fb3da80d08c428e71edd15b066271d9b678f1fb4af0b372db28f
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable


#### Stop tracking a file but keep the file:

git rm --cached filename

.... Cutting ...


--e0ac7962fb3da80d08c428e71edd15b066271d9b678f1fb4af0b372db28f--

--9b41f432623a90f2cfe46018fdd064553ce1b6cc8f70fdafd721e8b24647--

[#] Sat Apr 25 2015 16:33:19 EDT from TaMeR @ Uncensored to citadel_support@citadel.org

Subject: Re: text/pain turns in to Content-Type: multipart/mixed

[Reply] [ReplyQuoted] [Headers] [Print]

Never mind, that is a Go library that just formats everything multipart.
I am a minimalist stuff like that just bugs me.

[#] Mon Apr 27 2015 01:53:03 EDT from sparky973 @ Uncensored

Subject: Changelogs??

[Reply] [ReplyQuoted] [Headers] [Print]

**oops posted this in the lobby also :(..**

Hey guys, I just upgraded my system from 8.24 to 9.01 without 1 hiccup! (very nice!!). I was looking for a changelog listing the changes from versions, can anyone point me to one?

Thanks!



[#] Mon Apr 27 2015 01:54:02 EDT from sparky973 @ Uncensored

Subject: Re: Changelogs??

[Reply] [ReplyQuoted] [Headers] [Print]

apologies, please delete...



[#] Mon Apr 27 2015 09:47:04 EDT from Freakdog @ Uncensored

Subject: Re: text/pain turns in to Content-Type: multipart/mixed

[Reply] [ReplyQuoted] [Headers] [Print]

I wonder if what you're experiencing, here, is anything like the issue I reported, wherein 'cat <textfile> | mail -x "some subject" <someuser@some.server>' comes through as a MIME ".bin" attachment.

[#] Mon Apr 27 2015 13:44:31 EDT from s3cr3to @ Uncensored

Subject: Report strange behaviour in webcit

[Reply] [ReplyQuoted] [Headers] [Print]

OK.. maybe this occurs only in my install, because I'm using nginx as a proxy, maybe I have something wrong.. I will disable Nginx to use Webcit and post what I found.

Using Citadel 8.17 with WebCit 8.17, server build 8.17, debian squeeze. I'm using a self signed certificate, maybe FF & Chrome have something new that creates this odd effects.

 

(Paranoid mode activated) BUT, I wonder if this is an unexpected behaviour that is a sign of a eminent failure.

Thanks for reading:

I notice that after some browser updates I have strange behaviour in webcit. This issues occurs in ubuntu, arch, windows8, all with Firefox updated and chrome

  • If I try to login it shows something like "user not found", I need to press "Esc" and it logs
  • If I edit an account the form "display_edituser" is empty or with another user account.
  • If I edit the details it shows the "display_edituser" form.
  • I need to retry many times, but I feel that it doesn't edit the right data because sometimes It shows data that not correspond to the edit account.

I wonder If I can add/edit/add alias from the command line (I need to read the manual), because using webcit is not and option right now. Other some odd thing that happens:

  • FF/Webcit don't post the mail, it shows something like "can't sent empty messages"
  • Sometimes allows to add attachments, and sometimes not, and when it fails with empty message + attachment, after creating a new mail It shows the attachment from the previous attempt.
  • If I try to edit an user it shows empty fields, I need to re-select the user, a quickly click the need button and a few times works
  • If I try to login with whatever user, a message says "not found"... I need to press "Escape" and I'm logged (sometimes)

If by luck it allows me to edit the details of the user. If I clic "Cancel" or "Save changes" it always show this form ".../submit_vcard", the changes are not saved and display some data like:

{ "nummsgs": 3,
"startmsg": 0,
"roomname": "blah",
"msgs": [ [341695,"user","name",1234567890,"Mar 18  8:22AM",true ],
 ]
}

I tried with:

  • Ubuntu (3.13.0-49-generic #81~precise1-Ubuntu SMP Wed Mar 25 16:32:15 UTC 2015 x86_64 x86_64 x86_64 GNU/Linux) in chrome and Firefox 37.0.2
  • Manjaro (updated last week) in Firefox
  • Same effect in windows7-8

Regards



[#] Mon Apr 27 2015 14:44:18 EDT from s3cr3to @ Uncensored

Subject: Re: Report strange behaviour in webcit

[Reply] [ReplyQuoted] [Headers] [Print]

OK, I disable the apache flag, restart webcit and connecting with FF to webcit and all works again...

"fixed"? maybe, but why? When I re-enable the apache flag, the issues are no more in FF+Nginx.

I start to believe that webcit needs to be restarted periodically (today it have and uptime of +63 days)

Next time this occurs I will restart webcit first and check if the bugs are washed out.

 

Mon Apr 27 2015 01:44:31 PM EDT from s3cr3to @ Uncensored Subject: Report strange behaviour in webcit

OK.. maybe this occurs only in my install, because I'm using nginx as a proxy, maybe I have something wrong.. I will disable Nginx to use Webcit and post what I found.

Using Citadel 8.17 with WebCit 8.17, server build 8.17, debian squeeze. I'm using a self signed certificate, maybe FF & Chrome have something new that creates this odd effects.

 

(Paranoid mode activated) BUT, I wonder if this is an unexpected behaviour that is a sign of a eminent failure.

Thanks for reading:

I notice that after some browser updates I have strange behaviour in webcit. This issues occurs in ubuntu, arch, windows8, all with Firefox updated and chrome

  • If I try to login it shows something like "user not found", I need to press "Esc" and it logs
  • If I edit an account the form "display_edituser" is empty or with another user account.
  • If I edit the details it shows the "display_edituser" form.
  • I need to retry many times, but I feel that it doesn't edit the right data because sometimes It shows data that not correspond to the edit account.

I wonder If I can add/edit/add alias from the command line (I need to read the manual), because using webcit is not and option right now. Other some odd thing that happens:

  • FF/Webcit don't post the mail, it shows something like "can't sent empty messages"
  • Sometimes allows to add attachments, and sometimes not, and when it fails with empty message + attachment, after creating a new mail It shows the attachment from the previous attempt.
  • If I try to edit an user it shows empty fields, I need to re-select the user, a quickly click the need button and a few times works
  • If I try to login with whatever user, a message says "not found"... I need to press "Escape" and I'm logged (sometimes)

If by luck it allows me to edit the details of the user. If I clic "Cancel" or "Save changes" it always show this form ".../submit_vcard", the changes are not saved and display some data like:

{ "nummsgs": 3,
"startmsg": 0,
"roomname": "blah",
"msgs": [ [341695,"user","name",1234567890,"Mar 18  8:22AM",true ],
 ]
}

I tried with:

  • Ubuntu (3.13.0-49-generic #81~precise1-Ubuntu SMP Wed Mar 25 16:32:15 UTC 2015 x86_64 x86_64 x86_64 GNU/Linux) in chrome and Firefox 37.0.2
  • Manjaro (updated last week) in Firefox
  • Same effect in windows7-8

Regards



 



[#] Mon Apr 27 2015 15:18:17 EDT from dothebart @ Uncensored

Subject: Re: Citserver not running

[Reply] [ReplyQuoted] [Headers] [Print]

 

Fri Apr 24 2015 21:00:58 EDT from danda @ Uncensored Subject: Citserver not running

I installed Citadel on a Raspberry Pi about a year ago. It has been running great until a week or so ago. Now Citserver is not running. (at least it does not show up on a ps -eaf). No changes were made to the system at the time. It is a dedicated mail server. Could not find any error messages in any log files but I may not be looking in the right place.

When I try running citserver from a terminal I get: symbol lookup error: /usr/lib/arm-linux-gnueabihf/libdb-5.1.so: undefined symbol: en, version DB5_0

 

Any ideas?

I'm not a UNIX guy. Just playing with a home email server.

Thank You



Sounds like your system replaced the berkeley db library with something non compatible.

ldd should help you find out whats going on.

http://citadel.org/doku.php/faq:mastering_your_os:libs

you can use ln to replace the symlink to the old version of the library if its still there, and try ldconfig to refresh the cache, then try to restart citserver.



[#] Tue Apr 28 2015 11:39:46 EDT from TaMeR @ Uncensored to citadel_support@citadel.org

Subject: Re: text/pain turns in to Content-Type: multipart/mixed

[Reply] [ReplyQuoted] [Headers] [Print]

On 04/27/2015 04:47 PM, Freakdog wrote:
I wonder if what you're experiencing, here, is anything like the issue I reported,
wherein 'cat <textfile> | mail -x "some subject" <someuser@some.server>' comes
through as a MIME ".bin" attachment.



No, I am sure, it was a problem in the external library I used.
I fixed the code. works like a charm now.

[#] Tue Apr 28 2015 16:47:12 EDT from danda @ Uncensored

Subject: Re: Citserver not running

[Reply] [ReplyQuoted] [Headers] [Print]

Thanks for the tip.

I looked at the date and size of libdb-5.1.so on a virgin O/S install and they were the same. I saw no differences in ldd results. I'm rebuilding citadel from data backups and a clean O/S. It will be easier than debugging.

Thank You.



[#] Wed Apr 29 2015 03:04:18 EDT from smittyhead @ Uncensored

Subject: Server keeps stopping after crash from hours of excess cpu usage

[Reply] [ReplyQuoted] [Headers] [Print]

Hi, I've had a citadel server running successfully for about a year now and had a few things go on with the server running at 90% for long periods of time recently.  Now the server cannot go for more than a few minutes without crashing during high traffic times.  I have to constantly restart the service!

I have reinstalled the server completely 3 times now, restoring the data but it still does the same thing.  Can anyone help me please. 



Go to page: First ... 43 44 45 46 [47] 48 49 50 51 ... Last