FM7 Advanced & FX.PHP

[carl_vrijsen]

I'm hosting 20 dynamic websites width about 20.000 pageviews a day. Since my old Windows Server was to slow I installed a brand new XSERVE width FM7SA. I had to change some parameters, but everything works just fine. I use FX.PHP and PHP5 for my websites.

Except: Sometimes the link between the datasource and Apache breaks. This without any reason. Apache & Filemaker Server don't stop working, it's just the dynamic generated part of my websites that stops. The only option I have is to completely reboot the server. Very nasty, because now I have to check the site every minute. I just don't trust the system anymore. And every night I do a complete reboot, to be sure that he keeps working the next day....

I found some errors in watchdog:

2005-03-06 22:42:51 CET Started child "/Library/JBoss/3.2/bin/run.sh" as pid 349.

2005-03-06 22:42:51 CET Automatic reboot timer enabled.

2005-03-06 22:42:51 CET Reaped child process 349 ("/Library/JBoss/3.2/bin/run.sh"); quit with exit status 127.

2005-03-06 22:42:51 CET Process "/Library/JBoss/3.2/bin/run.sh" respawning too rapidly!

2005-03-07 17:45:27 CET Terminating due to signal.

2005-03-07 17:45:27 CET Automatic reboot timer disabled.

2005-03-07 17:45:27 CET Reaped child process 345 ("/usr/sbin/hwmond"); quit due to signal 15.

It seems that Tomcat and JBoss are causing this. And watchdog isn't able to restart these processes.

Of course I'm not sure that this is the reason, but where can I really find the reason for this?

Many thanks for someone who can help me! I'm sure I'm not the only one width this problem, because my configuration is almost standard...

[carl_vrijsen]

I'm hosting 20 dynamic websites width about 20.000 pageviews a day. Since my old Windows Server was to slow I installed a brand new XSERVE width FM7SA. I had to change some parameters, but everything works just fine. I use FX.PHP and PHP5 for my websites.

Except: Sometimes the link between the datasource and Apache breaks. This without any reason. Apache & Filemaker Server don't stop working, it's just the dynamic generated part of my websites that stops. The only option I have is to completely reboot the server. Very nasty, because now I have to check the site every minute. I just don't trust the system anymore. And every night I do a complete reboot, to be sure that he keeps working the next day....

I found some errors in watchdog:

2005-03-06 22:42:51 CET Started child "/Library/JBoss/3.2/bin/run.sh" as pid 349.

2005-03-06 22:42:51 CET Automatic reboot timer enabled.

2005-03-06 22:42:51 CET Reaped child process 349 ("/Library/JBoss/3.2/bin/run.sh"); quit with exit status 127.

2005-03-06 22:42:51 CET Process "/Library/JBoss/3.2/bin/run.sh" respawning too rapidly!

2005-03-07 17:45:27 CET Terminating due to signal.

2005-03-07 17:45:27 CET Automatic reboot timer disabled.

2005-03-07 17:45:27 CET Reaped child process 345 ("/usr/sbin/hwmond"); quit due to signal 15.

It seems that Tomcat and JBoss are causing this. And watchdog isn't able to restart these processes.

Of course I'm not sure that this is the reason, but where can I really find the reason for this?

Many thanks for someone who can help me! I'm sure I'm not the only one width this problem, because my configuration is almost standard...

[carl_vrijsen]

I'm hosting 20 dynamic websites width about 20.000 pageviews a day. Since my old Windows Server was to slow I installed a brand new XSERVE width FM7SA. I had to change some parameters, but everything works just fine. I use FX.PHP and PHP5 for my websites.

Except: Sometimes the link between the datasource and Apache breaks. This without any reason. Apache & Filemaker Server don't stop working, it's just the dynamic generated part of my websites that stops. The only option I have is to completely reboot the server. Very nasty, because now I have to check the site every minute. I just don't trust the system anymore. And every night I do a complete reboot, to be sure that he keeps working the next day....

I found some errors in watchdog:

2005-03-06 22:42:51 CET Started child "/Library/JBoss/3.2/bin/run.sh" as pid 349.

2005-03-06 22:42:51 CET Automatic reboot timer enabled.

2005-03-06 22:42:51 CET Reaped child process 349 ("/Library/JBoss/3.2/bin/run.sh"); quit with exit status 127.

2005-03-06 22:42:51 CET Process "/Library/JBoss/3.2/bin/run.sh" respawning too rapidly!

2005-03-07 17:45:27 CET Terminating due to signal.

2005-03-07 17:45:27 CET Automatic reboot timer disabled.

2005-03-07 17:45:27 CET Reaped child process 345 ("/usr/sbin/hwmond"); quit due to signal 15.

It seems that Tomcat and JBoss are causing this. And watchdog isn't able to restart these processes.

Of course I'm not sure that this is the reason, but where can I really find the reason for this?

Many thanks for someone who can help me! I'm sure I'm not the only one width this problem, because my configuration is almost standard...

[jfrancis]

We are also experiencing this. We are running filemaker server 7 advanced on mac os x, connecting through fx.php and php5. Our web sites run fine for like 3 days then filemaker just stops responding. The server is still running, since the terminals can still connect via the remote host, it's just web requests. It appears that tomcat craps out, here's the error messages that I see in the web_server_module_log.txt log:

[jk_ajp_common.c (681)]: Error: can't receive the response message from tomcat, network problem or tomcat is down.

[jk_ajp_common.c (1050)]: Error reading reply from tomcat. Tomcat is down or network problems.

[jk_ajp_common.c (1187)]: ERROR: Receiving from tomcat failed, recoverable operation. err=0

[jk_ajp_common.c (1198)]: Error connecting to tomcat. Tomcat is probably not started or is listenning on the wrong port. Failed errno = 61

I also see this error a lot, when the server IS responding to the fx.php:

[jk_ajp_common.c (926)]: Error sending request body

Anyone have any ideas why tomcat is crashing or how to solve this?

[jfrancis]

We are also experiencing this. We are running filemaker server 7 advanced on mac os x, connecting through fx.php and php5. Our web sites run fine for like 3 days then filemaker just stops responding. The server is still running, since the terminals can still connect via the remote host, it's just web requests. It appears that tomcat craps out, here's the error messages that I see in the web_server_module_log.txt log:

[jk_ajp_common.c (681)]: Error: can't receive the response message from tomcat, network problem or tomcat is down.

[jk_ajp_common.c (1050)]: Error reading reply from tomcat. Tomcat is down or network problems.

[jk_ajp_common.c (1187)]: ERROR: Receiving from tomcat failed, recoverable operation. err=0

[jk_ajp_common.c (1198)]: Error connecting to tomcat. Tomcat is probably not started or is listenning on the wrong port. Failed errno = 61

I also see this error a lot, when the server IS responding to the fx.php:

[jk_ajp_common.c (926)]: Error sending request body

Anyone have any ideas why tomcat is crashing or how to solve this?

[jfrancis]

We are also experiencing this. We are running filemaker server 7 advanced on mac os x, connecting through fx.php and php5. Our web sites run fine for like 3 days then filemaker just stops responding. The server is still running, since the terminals can still connect via the remote host, it's just web requests. It appears that tomcat craps out, here's the error messages that I see in the web_server_module_log.txt log:

[jk_ajp_common.c (681)]: Error: can't receive the response message from tomcat, network problem or tomcat is down.

[jk_ajp_common.c (1050)]: Error reading reply from tomcat. Tomcat is down or network problems.

[jk_ajp_common.c (1187)]: ERROR: Receiving from tomcat failed, recoverable operation. err=0

[jk_ajp_common.c (1198)]: Error connecting to tomcat. Tomcat is probably not started or is listenning on the wrong port. Failed errno = 61

I also see this error a lot, when the server IS responding to the fx.php:

[jk_ajp_common.c (926)]: Error sending request body

Anyone have any ideas why tomcat is crashing or how to solve this?

[carl_vrijsen]

I've looked at my httpd.conf file, and saw that the installation of PHP5 made an include in this file. So in the beginning of the file, PHP4 gets loaded, and in the end PHP5 gets loaded by this include.

That's why my server admin didn't show PHP5. It showed PHP4, but the server was running PHP5.

So I've deleted the include, and changed the lines like it should. Now PHP4 is not loaded anymore, and my server admin works now like it should.

Strange thing: my server works now for 4 days without any problem. So perhaps this is the solution? Let's hope so. I'll let you know!