daemon not running properly?

[chrisi99]

Code:

0 upgraded, 0 newly installed, 0 to remove and 3 not upgraded.

so that worked!

*waiting at generating pwd.... *

Code:

Setting Engine Permissions: ............done

        Congratulations!

        ispCP '1.0.0 RC5 OMEGA' Setup completed successfully!

the syslog (still) says:

Code:

Jul  1 11:45:05 ubuntu ispcp_daemon[11895]: ispCP daemon v1.1 started!
Jul  1 11:45:05 ubuntu kernel: [1646556.259955] ispcp_daemon[11895]: segfault at 00000046 eip b7e5e1e9 esp bf8fca38 error 4

the daemon.log reads now:

Code:

Jul  1 11:39:47 ubuntu ispcp_daemon[27146]: bind() error: Address already in use ! ^M  Please check for an other running daemon!^M
Jul  1 11:50:40 ubuntu ispcp_daemon[11819]: child 11819 started !
Jul  1 11:50:40 ubuntu ispcp_daemon[11820]: Error in opening stdout: Permission denied
Jul  1 11:50:40 ubuntu ispcp_daemon[11819]: send_line(): socket write error: Connection reset by peer
Jul  1 11:50:41 ubuntu ispcp_daemon[25261]: EINTR was received ! continue;

i am a little bit confused ... what could be the troublemaker? Maybe it has something to do with the strange way of ubuntu handeling the root permissions? I would have gone with debian anyway, but was overruled ...

[sci2tech]

chrisi99 Wrote:Jul 1 11:39:47 ubuntu ispcp_daemon[27146]: bind() error: Address already in use ! ^M Please check for an other running daemon!^M

Kill ispcp-daemon first. You already have an instance running. then check again

[chrisi99]

Code:

notroot@ubuntu:/var/log$ ps aux | grep ispcp_daemo
notroot  12041  0.0  0.2   3004   756 pts/1    R+   11:59   0:00 grep ispcp_daemo

Code:

notroot@ubuntu:/var/log$ /etc/init.d/ispcp_daemon start
* Starting ispCP GUI-Backend communication Daemon ispcp_daemon                                                                        [ OK ]

syslog

Code:

Jul  1 11:59:42 ubuntu ispcp_daemon[12057]: ispCP daemon v1.1 started!
Jul  1 11:59:42 ubuntu kernel: [1647432.670310] ispcp_daemon[12057]: segfault at 00000046 eip b7eb71e9 esp bfdf4738 error 4
Jul  1 12:00:01 ubuntu /USR/SBIN/CRON[12061]: (www-data) CMD ([ -x /usr/lib/cgi-bin/awstats.pl -a -f /etc/awstats/awstats.conf -a -r /var/log$

[sci2tech]

Sorry for delayed answer, but i did install a fresh ubuntu hardy with a fresh new ispcp (first thing i did modify libperl5.10 in libperl5.8). I could not reproduce your problem, but what I did modify /etc/hosts because hostname was truncated

Code:

127.0.0.1       eu.bogus.local  localhost
192.168.1.2     eu.bogus        e
::ffff:192.168.1.2      eu.bogus        e
::1     ip6-localhost ip6-loopback
fe00::0 ip6-localnet
ff00::0 ip6-mcastprefix
ff02::1 ip6-allnodes
ff02::2 ip6-allrouters
ff02::3 ip6-allhosts

instead

Code:

127.0.0.1       eu.bogus.local  localhost
192.168.1.2     eu.bogus        eu
::ffff:192.168.1.2      eu.bogus        eu
::1     ip6-localhost ip6-loopback
fe00::0 ip6-localnet
ff00::0 ip6-mcastprefix
ff02::1 ip6-allnodes
ff02::2 ip6-allrouters
ff02::3 ip6-allhosts

check this (maybe this is triggering segfaults problem)

Quote:eu:/_backup/var/log# cat syslog | grep ispcp
Jul 1 14:23:14 eu ispcp_daemon[30644]: ispCP daemon v1.1 started!
Jul 1 14:28:09 eu ispcp_daemon[30727]: child 30727 started !
Jul 1 14:28:09 eu ispcp_daemon[30727]: Aeee! SIG_PIPE was received ! Will we survive ?
Jul 1 14:28:09 eu ispcp_daemon[30727]: send_line(): socket write error: Broken pipe
Jul 1 14:28:10 eu ispcp_daemon[30644]: EINTR was received ! continue;

it is the normal output when a domain is added (I have 2 in my logs )

[chrisi99]

thanks for your great effort in helping me!

Code:

# 'hosts' file configuration.

127.0.0.1       ubuntu.oellinger.at.local       localhost
192.168.2.34    ubuntu.oellinger.at     ubuntu
::ffff:192.168.2.34     ubuntu.oellinger.at     ubuntu
::1     ip6-localhost ip6-loopback
fe00::0 ip6-localnet
ff00::0 ip6-mcastprefix
ff02::1 ip6-allnodes
ff02::2 ip6-allrouters
ff02::3 ip6-allhosts

I kept my old hosts from my VHCS Server, since the ubuntu server should replace it once operational.

the "system" is an VM Image, fresh from thoughtpolice. I only ran the apt update and installed ispcp with all needed packages...

do you have any ideas left?

Thank you A LOT for your help!

best wishes
Chris

[chrisi99]

interesting:

while manually updating the ispcp daemon a new "error" occurred:

Code:

Status      sys_command() | ERROR: External command '/usr/sbin/groupadd -g 2001 vu2001' returned '9' status !

(in the overview panel)...

[sci2tech]

This new error has nothing to do with daemon. The user that you try to add is already in /etc/passwd. either remove it from /etc/passwd, /etc/shadow, /etc/group and /etc/gshadow either in table auto-num (ispcp database) add this
INSERT INTO `ispcp`.`auto_num` (
`id` ,
`msg`
)
VALUES (
'1', 'auto-num'
);
But please test if this happend again, see if this happend (error message) when a new domain is added or it is not happend again.
Verify than no other domain has same uid/gid (table domain). if it has change uid/gid to 0 and then change the coresponding domain, domain-alias and subdomains status in change and coresponding mail_users staus in toadd, then run ispcp-rqst-mngr. To proper diagnostic must know how many domain exists in your system (i expect 2 and 1 with defective status), haw many corresponding entry do you have in /etc/passwd (DO NOT POST HERE /etc/passwd ).

[chrisi99]

changing the status in the DB to "toadd" and running the mgr resolves the problem. It did not occur a second time (added 3 domains for testing)...

still the original problem stays unsolved. I looked around the internet for about an hour or so, but this problem seems to be quite unique (as you suggested with your testinstallation).

maybe I could get hold of your configuration (because the hardy I use is as mentioned a package from another webpage)... yet that does not sound to promising ^^

[sci2tech]

I`m sorry that with original problem I can`t help further. I could not replicate so i can`t debug to see what is wrong.

[chrisi99]

i do of course understand that! You have done more than most would to help me.

I will see, if help from the developer is possible.

If you are interested from the technical perspektiv I would of course give you root access, since this is a test platform and only a virtual machine that would not be a problem!

best wishes
Chris