You are viewing a plain text version of this content. The canonical link for it is here.
Posted to modperl@perl.apache.org by Ranga Nathan <ka...@cox.net> on 2003/07/03 19:03:05 UTC
Apache config problem .. please help
I made a simple mod_perl change to the config and when restarting Apache
I got this error:
(98)Address already in use: make_sock: could not bind to address
0.0.0.0:2250
no listening sockets available, shutting down
/usr/local/apache/bin/apachectl: line 87: 16512 Segmentation fault
$HTTPD $ARGV
I then backed out the change and retried, got the same error.
Any clues?
Re: Apache config problem .. please help
Posted by Stas Bekman <st...@stason.org>.
> Anyway, all this started from an attempt to access env. vars from legacy
> scripts running under registry. What is the easiest way to get env. var
> access without the accompanying performance penalty that mod_perl
> documentation talks about?
If you need to have your scripts run unmodified, have PerlSetupEnv Off in all
sections that you don't need to access the CGI env vars, and have it On only
in those <Location>s that run legacy scripts.
http://perl.apache.org/search/swish.cgi?query=PerlSetupEnv&sbm=SecE&submit=search
__________________________________________________________________
Stas Bekman JAm_pH ------> Just Another mod_perl Hacker
http://stason.org/ mod_perl Guide ---> http://perl.apache.org
mailto:stas@stason.org http://use.perl.org http://apacheweek.com
http://modperlbook.org http://apache.org http://ticketmaster.com
Re: Apache config problem .. please help
Posted by Ranga Nathan <ka...@cox.net>.
Gedanken wrote:
>I know this is not of much help, but I have had situations where a badly
>terminating process would prevent subsequent processes from using that
>port. on windows, i never found a solution other than to reboot. on
>solaris 7, i never found a solution other than to wait 8 minutes. I did
>some reading and found socket inet options to change how such programs
>bound thge port so as to allow others to use it in more friendly a fashion
>but such a change wasnt possible with the process in question.
>
>
>
The problem is resolved. Unbenownst to me - I need to investigate this -
the Listen IP:port line in Apache had the wrong IP address. I really
dont know how this happened. The number is so different yet
syntactically correct to be a keystroke error.
Thanks everyone for pitching in. For one I can now use ' netstat'
utility to check ports being used,. Thanks Ged!
I keep talking to my corporate IT people about open source , and they
constantly ask me 'who will support it? what do they get out of it?' . I
can not even begin to explain to them the 'connectedness' across the
'net. Should people always measure everything in terms of money?
Anyway, all this started from an attempt to access env. vars from legacy
scripts running under registry. What is the easiest way to get env. var
access without the accompanying performance penalty that mod_perl
documentation talks about?
Regards
Re: Apache config problem .. please help
Posted by Gedanken <ge...@io.com>.
I know this is not of much help, but I have had situations where a badly
terminating process would prevent subsequent processes from using that
port. on windows, i never found a solution other than to reboot. on
solaris 7, i never found a solution other than to wait 8 minutes. I did
some reading and found socket inet options to change how such programs
bound thge port so as to allow others to use it in more friendly a fashion
but such a change wasnt possible with the process in question.
--
gedanken
Re: Apache config problem .. please help
Posted by Ged Haywood <ge...@www2.jubileegroup.co.uk>.
Hi there,
On Thu, 3 Jul 2003, Ranga Nathan wrote:
> >In other words there's an Apache still running
>
> The 'top' output is :
> [snip]
> How can I get a list of ports being used so I can kill the processes?
As I said, there may be more wrong than just what's obvious. It looks
like the obvious explanation that another Apache is running is not the
right one. At the risk of asking questions I wouldn't have to ask if
you'd posted the information I asked for, why port 2205? That's not
what I'd expect to see, but without more information I'm groping about
in the dark. What's trying to use that port? MySQL perhaps? Why not
stop it first, or change the port that Apache is trying to use, or the
port that whatever apachectl is trying to start is trying to use, or...
> As far as I can tell httpd is not running. The IP is 208.179.25.28
c2h5oh:~$ >>> ping 208.179.25.28
PING 208.179.25.28 (208.179.25.28): 56 data bytes
64 bytes from 208.179.25.28: icmp_seq=0 ttl=236 time=389.6 ms
--- 208.179.25.28 ping statistics ---
2 packets transmitted, 1 packets received, 50% packet loss
round-trip min/avg/max = 389.6/389.6/389.6 ms
c2h5oh:~$ >>> telnet 208.179.25.28 80
Trying 208.179.25.28...
telnet: Unable to connect to remote host: Connection refused
Looks like you're right. Have you read the stuff I mentioned?
> I dont see any apparent signs of hacking either.
Don't go tilting at windmills just yet. :)
73,
Ged.
Re: Apache config problem .. please help
Posted by Ranga Nathan <ka...@cox.net>.
Ged Haywood wrote:
>Hi there,
>
>On Thu, 3 Jul 2003, Dennis Stout wrote:
>
>
>
>>>I made a simple mod_perl change to the config and when restarting Apache
>>>I got this error:
>>>(98)Address already in use: make_sock: could not bind to address
>>>0.0.0.0:2250
>>>no listening sockets available, shutting down
>>>/usr/local/apache/bin/apachectl: line 87: 16512 Segmentation fault
>>>$HTTPD $ARGV
>>>
>>>I then backed out the change and retried, got the same error.
>>>
>>>
>>killall httpd
>>
>>then try it again :)
>>
>>
>httpd was not running and I have been running Apache 2x with mod_perl 1.99x for over two months now. It started with a problem I traced to env. var $ENV{HTTP_ACCEPT}. I figured out that I needed PerlOptions +SetupENV. So I added this, stopped and tried to start Apache. That's when I got this shock.
>
>In other words there's an Apache still running when you're trying to
>start a second one which wants to listen on the same socket that the
>first Apache is already listening on. That isn't permitted.
>
>But you shouldn't be getting segmentation faults in that case so
>something else is probably wrong too. Did you build from source?
>Did you follow the instructions in the Guide? Linux? 1.3.27/1.27?
>DSO? Maybe you can post the information requested in the docs?
>
>73,
>Ged.
>
>
>
The 'top' output is : 3:38am up 8 days, 22:58, 1 user, load average:
0.00, 0.00, 0.00
22 processes: 21 sleeping, 1 running, 0 zombie, 0 stopped
CPU states: 0.0% user, 0.3% system, 0.0% nice, 99.6% idle
Mem: 970768K av, 212504K used, 758264K free, 7852K shrd, 98668K
buff
Swap: 530104K av, 3216K used, 526888K free 88104K
cached
PID USER PRI NI SIZE RSS SHARE STAT %CPU %MEM TIME COMMAND
16688 kairanga 10 0 996 996 800 R 0.3 0.1 0:00 top
1 root 0 0 76 64 48 S 0.0 0.0 0:10 init
2 root 0 0 0 0 0 SW 0.0 0.0 0:00 kflushd
3 root 8 0 0 0 0 SW 0.0 0.0 34:03 kupdate
4 root 0 0 0 0 0 SW 0.0 0.0 9:07 kswapd
5 root 0 0 0 0 0 SW 0.0 0.0 0:00 keventd
57 root 0 0 232 232 0 S 0.0 0.0 0:00 rc.M
77 root 2 0 272 272 160 S 0.0 0.0 2:20 syslogd
80 root 0 0 640 640 128 S 0.0 0.0 0:00 klogd
82 root 0 0 368 324 220 S 0.0 0.0 0:04 sshd
86 root 6 0 336 336 240 S 0.0 0.0 0:00 crond
93 root 2 0 692 488 292 S 0.0 0.0 0:03 sendmail
100 root 0 0 2352 1708 936 S 0.0 0.1 0:07 poprelayd
102 root 0 0 2668 1248 292 S 0.0 0.1 0:00 miniserv.pl
104 root 0 0 220 168 148 S 0.0 0.0 0:00 inetd
105 root 0 0 184 184 0 S 0.0 0.0 0:00 safe_mysqld
127 mysql 0 0 2384 2384 1364 S 0.0 0.2 0:00 mysqld
129 mysql 0 0 2384 2384 1364 S 0.0 0.2 0:00 mysqld
130 mysql 0 0 2384 2384 1364 S 0.0 0.2 0:00 mysqld
16316 root 0 0 1284 1268 1124 S 0.0 0.1 0:00 sshd
16318 kairanga 0 0 1364 1352 1208 S 0.0 0.1 0:00 sshd
16319 kairanga 3 0 1240 1240 964 S 0.0 0.1 0:00 bash
There has been no change to Apache httpd.conf other than the addition
of PerlOptions +SetupEnv which I commented out anyway.
How can I get a list of ports being used so I can kill the processes? As
far as I can tell httpd is not running. The IP is 208.179.25.28
I dont see any apparent signs of hacking either.
Re: Apache config problem .. please help
Posted by Ged Haywood <ge...@www2.jubileegroup.co.uk>.
Hi there,
On Thu, 3 Jul 2003, Dennis Stout wrote:
> > I made a simple mod_perl change to the config and when restarting Apache
> > I got this error:
> > (98)Address already in use: make_sock: could not bind to address
> > 0.0.0.0:2250
> > no listening sockets available, shutting down
> > /usr/local/apache/bin/apachectl: line 87: 16512 Segmentation fault
> > $HTTPD $ARGV
> >
> > I then backed out the change and retried, got the same error.
>
>
> killall httpd
>
> then try it again :)
In other words there's an Apache still running when you're trying to
start a second one which wants to listen on the same socket that the
first Apache is already listening on. That isn't permitted.
But you shouldn't be getting segmentation faults in that case so
something else is probably wrong too. Did you build from source?
Did you follow the instructions in the Guide? Linux? 1.3.27/1.27?
DSO? Maybe you can post the information requested in the docs?
73,
Ged.
Re: Apache config problem .. please help
Posted by Dennis Stout <st...@stout.dyndns.org>.
> I made a simple mod_perl change to the config and when restarting Apache
> I got this error:
> (98)Address already in use: make_sock: could not bind to address
> 0.0.0.0:2250
> no listening sockets available, shutting down
> /usr/local/apache/bin/apachectl: line 87: 16512 Segmentation fault
> $HTTPD $ARGV
>
> I then backed out the change and retried, got the same error.
killall httpd
then try it again :)
Dennis