perl execution failed

17 posts / 0 new
Last post
#1 Thu, 07/21/2011 - 10:57

perl execution failed

Hello, I'm a newbie I have webmin 1.550 and virtualmin instaled on ubuntu server 10.04.2. Today all clients were logged out and didn't log in anymore. So I 've opened webmin and found the folowing message: HTTP/1.0 500 Perl execution failed Server: MiniServ/1.550 Date: Thu, 21 Jul 2011 15:53:05 GMT Content-type: text/html Connection: close Error - Perl execution failed postfix::is_postfix_running failed : Echec de la requête sur la configuration postfix pour obtenir la valeur actuelle du paramètre queue_directory: /usr/sbin/postconf: fatal: open /etc/postfix/ No such file or directory at /usr/share/webmin/ line 1360.

Can someone help me? Thanks

Thu, 07/21/2011 - 11:07


Hmm, it looks like your Postfix configuration file is missing. That's an unusual issue :-)

Is it possible someone could have accidentally removed it?

Anyhow, to get that back, you'd need to restore it from a backup. You may also want to verify that in /etc/postfix, a file exists there as well -- that's also needed for Postfix to function.


Thu, 07/21/2011 - 15:42

Hello Eric, it's possible that someone have removed it accidentally, but I really don't know how. And I also really don't know how to restore it. Is there a topic anywhere how to do this? Thanks Vincent

Thu, 07/21/2011 - 15:45


Well, you can only restore it if you made a backup of your system somewhere along the way. Do you have a system image, or a backup of /etc?


Thu, 07/21/2011 - 15:51 (Reply to #4)


I usually make backups with partimage of windows installations, but I didn't for a linux. I didn't have any problems like this before with a linux installation...

Is there another way, or will I have to make a complete reinstall?


Thu, 07/21/2011 - 23:04

Well, it's pretty unusual for files to go missing -- however, anything is capable of happening (including disk failure!), so it's always good to have a full backup, no matter what operating system you're working with.

Let's start here though, so that we know the extent of the problem... what is the output of this command:

ls -l /etc/postfix/

Fri, 07/22/2011 - 10:13

Hello Eric, I've pick up the server at the office and bring it home. So I can give you the answer:

total 72 -ru-r--r-- 1 root root 318 2011-05-04 09:59 -ru-r--r-- 1 root root 5301 2011-05-04 09:59 -ru-r--r-- 1 root root 18992 2010-02-18 07:38 postfix-files -rwxr-xr-x 1 root root 8729 2010-02-18 07:38 postfix-script -rwxr-xr-x 1 root root 24256 2010-02-18 07:38 post-install druxr-xr-x 2 root root 4096 2010-02-18 07:38 sasl

By the way, I've shutted down the server with webmin, but at home I can't access to the server with webmin??? IP address for eth0 didn't change. Under System information, I've noticed: => There are 2 zombie processes.

Thanks for the help

Fri, 07/22/2011 - 12:52

Okay, so it just looks like you're missing the file.

The following is what a default looks like on Ubuntu 10.04. What you could do is, as root, take the following lines, and put them into /etc/postfix/, and then restart Postfix using "/etc/init.d/postfix restart".

Also, on the line beginning with "mydestination", you'll want to make sure your system's hostname is listed at the end there in place of where it says "hostname.domain.tld".

alias_database = hash:/etc/aliases
alias_maps = hash:/etc/aliases
allow_percent_hack = no
append_dot_mydomain = no
biff = no
broken_sasl_auth_clients = yes
config_directory = /etc/postfix
home_mailbox = Maildir/
inet_interfaces = all
mailbox_command = /usr/bin/procmail-wrapper -o -a $DOMAIN -d $LOGNAME
mailbox_size_limit = 0
mydestination = domain.tld, localhost.localdomain, localhost, hostname.domain.tld
myhostname = ubuntu.domain.tld
mynetworks = [::ffff:]/104 [::1]/128
readme_directory = no
recipient_delimiter = +
relayhost =
sender_bcc_maps = hash:/etc/postfix/bcc
smtp_tls_session_cache_database = btree:${data_directory}/smtp_scache
smtpd_banner = $myhostname ESMTP $mail_name (Ubuntu)
smtpd_recipient_restrictions = permit_mynetworks permit_sasl_authenticated reject_unauth_destination
smtpd_sasl_auth_enable = yes
smtpd_sasl_security_options = noanonymous
smtpd_tls_cert_file = /etc/ssl/certs/ssl-cert-snakeoil.pem
smtpd_tls_key_file = /etc/ssl/private/ssl-cert-snakeoil.key
smtpd_tls_session_cache_database = btree:${data_directory}/smtpd_scache
smtpd_use_tls = yes
virtual_alias_maps = hash:/etc/postfix/virtual
Fri, 07/22/2011 - 21:28

Hello Eric

thanks for the information but

I don't know how to put those lines into /etc/postfix/

I know this is basic, I've done something similar when I've first installed the server, but can't remember how...


Fri, 07/22/2011 - 22:36


Well, you need to create a new file containing the above contents.

You could use a text editor on your server such as nano, and copy and paste the contents in there... or you could create the file locally, and copy it up to your server using SFTP, FTP, or the like.

Personally, I'd use a text editor on the server, but there's other ways to do it :-)


Sun, 07/24/2011 - 05:04

Hello Eric,

I've found some time to do all this and got the answer:

  • Stopping Postfix Mail Transport Agent postfix

  • Starting Postfix Mail Transport Agent postfix

mkdir: cannot create directory 'etc/ssl': permission denied

So what do I have to do now? By the way, I still can't access to webmin: in my browser I put the right address : and receive the answer : the server at the address takes too long time to answer!

Thanks a lot for all you've already done for me



Sun, 07/24/2011 - 11:37

Hmm, so what output do you receive if you type this command:

ls -ld /etc/ssl

Also, what output do you receive if you type this:

dmesg | tail -10

The first command will show if there's a directory named "/etc/ssl" -- if not, that's really odd, as it appears to be missing too!

The second command will show recent kernel output... if files just start missing, it's possible someone deleted them. But it's also possible that there's a problem with the hard disk. If there's hard disk issues, those might show up in the kernel output.


Mon, 07/25/2011 - 09:41

Hello Eric,

back into the office and have some news.

Webmin is here working very well. It seems the server works as DHCP server too (I didn't know that before, so my collegues were without internet this week-end!) When I'm looking into status of the server, there aren't any failings anymore. Just notice that the NFS server isn't working.

On the command ls -ld /etc/ssl I've got the answer:

drwxr-xr-x 4 root root 4096 2011-05-04 09:58 /etc/ssl

On the command dmesg | tail - 10 , I've got the answer:

[ 9.977778] Bridge firewalling registered [ 9.987509] virbr0: starting userspace STP failed, starting kernel STP [ 9.998433] ip_tables: (C) 2000-2006 Netfilter Core Team [ 10.065846] nf_comtrack version 0.5.0 (16384 buckets, 65536 max) [ 10.066758] CONFIG_NF_CT_ACCT is deprecated and will be removed soon. Please use [ 10.066763] nf_conntrack.acct=1 kernel parameter, acct=1 nf_conntrack module option or [ 10.066764] sysctl net.netfilter.nf_conntrack_acct=1 to enable it. [ 10.286137] lo: Disabled Privacy Extensions [ 19.088508] eth0: no IPv6 routers present [ 20.608504] virbr0: no IPv6 routers present

Farther I also can tell you that windows clients still can't access the server: does Samba works properly?



Mon, 07/25/2011 - 10:26

Hmm, it sounds like you're having a few problems there!

I don't see any unusual messages in your dmesg output.

Webmin doesn't act as a DHCP server... so Webmin being down shouldn't prevent anyone from getting an IP address. If DHCP isn't working, that means that the DHCP daemon isn't running, or is otherwise malfunctioning.

Samba isn't setup automatically, that's something you would need to configure. Was that once working?

However, Samba isn't needed if your users are just using FTP to access your server.


Mon, 07/25/2011 - 10:41

Hello Eric,

In webmin, I see that the DHCP server is activated. So when the server was down, no-one could get a IP addess.

Samba was configured earlier. Windows clients (xp-pro) were connected to the domain and so had access to the files. Some Windows clients (xp-home) used an executive called "kix" to connect to the server.

From ubuntu clients, I was connecting using "connect to server, partage windows share, username, domain" connect, and then password.


Mon, 07/25/2011 - 10:56


If you're looking at Webmin -> Servers -> DHCP, that's Webmin's way of configuring the system DHCP daemon that exists on your server.

I don't believe Webmin contains a built-in DHCP server -- it wouldn't need one since all major distributions include a DHCP service in their software. That said, if you're seeing something I'm not -- I've been wrong on more than one occasion about what features Webmin comes with :-)

It sounds like a number of things all stopped working all at the same time though... Postfix, Samba, NFS... do you know of anything else that happened around the same time that might have affected all of that? It's rare that any one of those problems would come up, so having 3 services all stop functioning at the same time suggests a problem of some sort :-)


Mon, 07/25/2011 - 11:11


I really don't know what happened. I've heard a few minutes ago from a colleague that one other knows the password and was creating an account for himself and then delete it. Meanwhile, i already have changed the password. So at this time I'm sure I am the only one to access Webmin.

About DHCP and webmin, I really don't know myself. I have to recognize that it is a relative who came to program us the server. And as he is a professor, he went on holiday to the end of August. So I can't ask him any help for still one month.


Topic locked