Lighttpd + squirrelmail = no go? [solved]

[OneInchMen]

Hi there,

This weekend I decided to switch from Apache2 to Lighttpd. I run a little P3/256MB server at home, so Lighty seems the better choice when it comes to memory and speed... I also like the the 'simple vhosts' option...

Compiling the latest keyword-masked version went fine. Configuration was really simple and I got it running perfectly. All my own sites (mysql backed php) work perfectly (and faster?) except for squirrelmail... I also tried the stable version, to no avail...

I can log in to Squirrelmail, but when showing the folder list & contents, only the frames page is loaded an nothing happens after that. (I only get the frame border and 2 empty panes)

Does anybody have squirrelmail running under Lighty?

-thx!-

Simple vhost config:

Code: Select all

$HTTP["host"] !~ "^(mail\.domain\.ext)$" {
  simple-vhost.server-root   = "/var/www/domain.ext"
  simple-vhost.default-host  = "default"
  simple-vhost.document-root = "public"
}

$HTTP["host"] == "mail.domain.ext" {
  server.document-root = "/var/www/localhost/htdocs/squirrelmail/"
}

PS: src/configtest.php outputs: Congratulations, your SquirrelMail setup looks fine to me!

[hegga]

i can't imagine that there should be any problems to setup squirrelmail with lighttpd.
i will try this on my server with lighttpd, when i get the time. does lighttpd say anything
in the logs? or is it possible to run squirrelmail with debug?

[OneInchMen]

Aha! Found something in error.log:

Code: Select all

2006-05-01 14:37:08: (log.c.75) server started
2006-05-01 14:39:53: (mod_fastcgi.c.2723) establishing connection failed: Connection refused socket: tcp:127.0.0.1:1029
2006-05-01 14:39:53: (mod_fastcgi.c.2723) establishing connection failed: Connection refused socket: tcp:127.0.0.1:1028
2006-05-01 14:39:53: (mod_fastcgi.c.2723) establishing connection failed: Connection refused socket: tcp:127.0.0.1:1027
2006-05-01 14:39:53: (mod_fastcgi.c.2723) establishing connection failed: Connection refused socket: tcp:127.0.0.1:1026
2006-05-01 14:39:55: (mod_fastcgi.c.3457) all handlers for  /src/webmail.php on .php are down.
2006-05-01 14:39:59: (mod_fastcgi.c.2669) fcgi-server re-enabled: tcp:127.0.0.1:1029
2006-05-01 14:39:59: (mod_fastcgi.c.2669) fcgi-server re-enabled: tcp:127.0.0.1:1028
2006-05-01 14:39:59: (mod_fastcgi.c.2669) fcgi-server re-enabled: tcp:127.0.0.1:1027
2006-05-01 14:39:59: (mod_fastcgi.c.2669) fcgi-server re-enabled: tcp:127.0.0.1:1026

fastcgi use-flag is disabled:

Code: Select all

www-servers/lighttpd-1.4.11-r1  +bzip2 -doc -fam -fastcgi -gdbm -ipv6 -ldap -lfs -lua +memcache -minimal +mysql +pcre +php -rrdtool +ssl -test -webdav +xattr 0 kB [1]

[hegga]

do you have fastcgi enabled in lighttpd.conf?

[OneInchMen]

lighttpd.conf:

Code: Select all

 cat /etc/lighttpd/lighttpd.conf |grep -i fast
# uncomment for php/fastcgi support
include "mod_fastcgi.conf"

mod_fastcgi.conf:

Code: Select all

server.modules += ("mod_fastcgi")
fastcgi.server = ( ".php" =>
                            ( "localhost" =>
                                    ( "host" => "127.0.0.1",
                                            "port" => 1026,
                                            "bin-path" => "/usr/bin/php-cgi"
                                    )
                            )
                    )

/usr/bin/php-cgi

Code: Select all

la /usr/bin/php-cgi
lrwxrwxrwx 1 root root 25 Feb 16 12:55 /usr/bin/php-cgi -> /usr/lib/php5/bin/php-cgi

emerge -pv dev-lang/php

Code: Select all

[ebuild   R   ] dev-lang/php-5.1.2-r1  -adabas -apache -apache2 -bcmath -berkdb -birdstep +bzip2 -calendar -cdb +cgi -cjk -cli +crypt -ctype -curl -curlwrappers -db2 -dbase -dbmaker -debug -discard-path -doc -empress -empress-bcs -esoob -exif -fastbuild -fdftk -filepro -firebird -flatfile +force-cgi-redirect -frontbase +ftp +gd -gd-external -gdbm -gmp -hardenedphp -hash -hyperwave-api -iconv +imap -informix -inifile -interbase -iodbc -ipv6 -java-external -kerberos -ldap -libedit -mcve -memlimit -mhash -ming -msql -mssql +mysql -mysqli -ncurses -nls -oci8 -oci8-instant-client -odbc -pcntl +pcre -pdo -pdo-external -pic -posix -postgres -qdbm +readline -recode -reflection -sapdb -sasl +session -sharedext -sharedmem -simplexml -snmp -soap -sockets -solid -spell -spl -sqlite +ssl -sybase -sybase-ct -sysvipc -threads -tidy -tokenizer +truetype +unicode -vm-goto -vm-switch -wddx -xml -xmlreader -xmlrpc -xmlwriter -xpm -xsl -yaz +zip +zlib 0 kB

ps ax|grep php

Code: Select all

13671 ?        Ss     0:00 /usr/bin/php-cgi
13673 ?        Ss     0:00 /usr/bin/php-cgi
13674 ?        Ss     0:00 /usr/bin/php-cgi
13675 ?        S      0:00 /usr/bin/php-cgi
13676 ?        S      0:00 /usr/bin/php-cgi
13677 ?        S      0:00 /usr/bin/php-cgi
13678 ?        Ss     0:00 /usr/bin/php-cgi
13679 ?        S      0:00 /usr/bin/php-cgi
13701 pts/0    S+     0:00 grep php

[hegga]

i see, but should not lighttpd be compiled with the fastcgi use flag then?

[OneInchMen]

Been there, done that... I didn't see the errors posted above again, but nothing changed in squirrelmail though...

So: no errors anymore, just squirrelmail dying after login...

[hegga]

can this help?

http://lists.debian.org/debian-isp/2003 ... 00309.html

[OneInchMen]

Thanx! That did help! Although the proposed solution is completely different. I did have a corrupt config.php which worked on apache2, but failed on Lighttpd (lighty).

When re-emerging squirrelmail, I noticed 7 of these lines flashing by:

Code: Select all

CONFIG_PROTECT="/var/www/localhost/htdocs/squirrelmail//plugins/secure_login" etc-update

After doing so, squirrelmail started working again!!!

Why didn't this show up at etc-update? It seems that etc/config file management is somewhat crappy on gentoo (loving the rest though)

I'm going to strip all the use-flags and packages I've added in the process one-by-one and see if anything changes...

Thank you all!!!