Aditya Dhekney wrote: > >So what does the error message say? > > PHP Warning: Unable to load dynamic library './midgard.so' - ./midgard.so: > cannot open shared object file: No such file or directory in Unknown on line > 0 I've done some checking -- 'Unkown' means the extension_dir is not set. If it is set, it may be that php is reading a different php.ini. phpinfo will tell you what file php tries to read. > >If you comment out midgard.so from php.ini, but leave mod_midgard active > >in httpd.conf, does apache work as usual (except midgard requests)? > > NO!!NO!!!!!! IF midgard.so is commented out from php.ini, mod_midgard is > active, apache does not work at all!!! even the httpd stops working! If i > comment out mod_midgard, then apache starts working again...but, if i go and > add mod_midgard again, httpd stops working!! ie. after adding mod_midgard, > httpd stop, httpd start don't work anymore!! The process stays alive even > if i do httpd stop (checked with ./httpd status). Starting/stopping with > /etc/rc.d/init.d/httpd start/stop is hopeless...i just use the binary all > the time now! I'm seeing something like this on a redhat 6.2-ish system. This system is not your everyday reference rh install -- an odd mix of packages, source installs and --forced packages[0]. But apache is segfaulting now when I load mod_migard, and the weird thing is, it keeps segfaulting even if I downgrade mod_midgard (which used to work just fine a few weeks ago). My Debian systems (2xSid, one Potato) are OK (although potatos' php4 install is outdated, and the php-config it installs broken). > Updated extension_dir is being recognized, but again, nothing works if i add > the mod_midgard stuff in the httpd.conf file. That's a different problem that we'll solve separately. Comment out php4 in httpd.conf, and put mod_midgard in. Kill apache, one way or the other, if the regular stop script doesn't work, killall is your friend. Unless you're running Solaris, in which case it's your worst enemy. Start apache by hand (would be /usr/sbin/apache -X for me) and tell me what it does/says in the console, and the error log. Emile [0] don't start. I used to use RH, and I got fed up with the full reinstall that each upgrade required. It's being kept alive(ish) while I set up my Debian replacement server. --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
