postgrey not starting after release upgrade

[Shinji]

Hi!

After upgrading my server from ubuntu 13.04 to 13.10 the "postgrey" daemon does not start.

If I call

===
root@mail /home/shinji # /etc/init.d/postgrey start
 * Starting postfix greylisting daemon postgrey [ OK]
===

everything seems to be fine. But I get this in my logfiles:

===
Nov 18 20:42:09 mail postfix/smtpd[4031]: warning: connect to 127.0.0.1:10023: Connection refused
Nov 18 20:42:09 mail postfix/smtpd[4031]: warning: problem talking to server 127.0.0.1:10023: Connection refused
===

So postgrey is not running :(

===
root@mail /home/shinji # sudo netstat -pln | grep postgrey
root@mail /home/shinji #
===

My "/etc/default/postgrey" file looks like this

===
POSTGREY_OPTS="--inet=10023 --delay=120 --max-age=90"
===

And if I try to start it as 'root' with

===
postgrey --inet 10023
==

i'll get this:

===
2013/11/18-20:45:59 postgrey (type Net::Server::Multiplex) starting! pid(4063)
Resolved [localhost]:10023 to [127.0.0.1]:10023, IPv4
Binding to TCP port 10023 on host 127.0.0.1 with IPv4
Insecure dependency in bind while running with -T switch at /usr/lib/perl/5.14/IO/Socket.pm line 202.
===

[Shinji (thorsten-reichelt)]

I did an

===
 apt-get remove --purge postgrey
===

After that I rebooted the server and entered

===
 apt-get install postgrey
===

But this is the result:

===
root@mail /home/shinji # service postgrey start
 * Starting postfix greylisting daemon postgrey [ OK ]
root@mail /home/shinji # lsof -i TCP:10023
root@mail /home/shinji #
===

:(

[Shinji (thorsten-reichelt)]

I can start postgrey by calling

====

root@mail /home/shinji # postgrey -v --inet=127.0.0.1:10023
2013/11/18-22:15:33 postgrey (type Net::Server::Multiplex) starting! pid(3239)
Binding to TCP port 10023 on host 127.0.0.1 with IPv4
Setting gid to "116 116"
Setting uid to "109"

====

But it WILL NOT start by calling "/etc/init.d/postgrey start" or "service postgrey start"!

[Marius Gedminas (mgedmin)]

Given that I use postgrey on my servers, and that I will upgrade to 14.04 LTS when it comes out, I decided to try some preventative maintenance and see if postgrey is really broken on 13.10.

I was unable to reproduce your problem:

    $ sudo apt-get install postgrey

    $ ps $(cat /var/run/postgrey.pid)
      PID TTY STAT TIME COMMAND
    17720 ? Ss 0:00 /usr/sbin/postgrey --pidfile=/var/run/postgrey.pid --daemonize --inet=10023

    $ sudo lsof -i TCP:10023
    COMMAND PID USER FD TYPE DEVICE SIZE/OFF NODE NAME
    /usr/sbin 17720 postgrey 6u IPv4 2248799 0t0 TCP localhost:10023 (LISTEN)

    $ less /var/log/syslog
    Nov 19 09:56:41 platonas postgrey[17720]: Process Backgrounded
    Nov 19 09:56:41 platonas postgrey[17720]: 2013/11/19-09:56:41 postgrey (type Net::Server::Multiplex) starting! pid(17720)
    Nov 19 09:56:41 platonas postgrey[17720]: Resolved [localhost]:10023 to [127.0.0.1]:10023, IPv4
    Nov 19 09:56:41 platonas postgrey[17720]: Binding to TCP port 10023 on host 127.0.0.1 with IPv4
    Nov 19 09:56:41 platonas postgrey[17720]: Setting gid to "139 139"
    Nov 19 09:56:41 platonas postgrey[17720]: Setting uid to "127"
    Nov 19 09:59:45 platonas postgrey[17720]: 2013/11/19-09:59:45 Server closing!

Hm.

[Shinji (thorsten-reichelt)]

I can start postgrey with

===
service postgrey start
===

If I remove the "-T" parameter from "/usr/sbin/postgrey"! So instead of

===
#!/usr/bin/perl -T -w
===

I have to write

===
#!/usr/bin/perl -w
===

Now postgrey starts again. It seems that disabling the perl tainted mode solves the problem. But I guess this opens some security holes. :(

[Shinji (thorsten-reichelt)]

Ok, my final solution is now to edit the postgrey pearl file as described here:

https://github.com/yasuhirokimura/postgrey/commit/9673b54064691a5b9c295ffea340d8a1f9ee1cb8

But I am running Perl 5.14 and not 4.18!

[Rolf Leggewie (r0lf)]

Is this still an issue in xenial or later?

[Rolf Leggewie (r0lf)]

ping shinji