Search results

Not sure what cause that, the URL must be exactly as the the file name, index file should be set properly. You can turn on debug logging see what exactly is happening.

run "lsws/admin/misc/awstats_install.sh" instead. :-)

I tried this, the my founding is interesting. I made some error in a controler code intentionally, so rails will produce a 500 Error page. When I run it in development mode, both mongrel and litespeed return 500 page, however, when I switch to production mode, rails return a page with 200, for...

For ruby-lsapi, that is because a lsapi ruby worker process only idle for 60 secs by default, you can set environment variable "LSAPI_MAX_IDLE=xxx" to increase it. For PHP FCGI connection, it probably because the TCP connection has been timed out. When number of active connections reaches...

This could happen when a reqest was cancel before it being processed by the backend rails engine. Maybe we could void logging canceled requests.

lsws/admin/misc/uninstall.sh, if anything leftover in lsws directory, just delete them.

Please check this out http://www.litespeedtech.com/support/wiki/doku.php?id=litespeed_wiki:cgi_fcgi_lsapi_php_suexec

And both FCGI PHP and LSAPI PHP with LiteSpeed is faster than Apache's mod_php. LSAPI PHP is the fastest. Apache's mod_fastcgi is not as fast as ours. Please checkout our benchmark page.

"Ruby path" is the path to "ruby" executable, most likely it is /usr/local/bin/ruby on your server. :-)

Yes, it is possible to use Litespeed for all our Rails sites and keep Apache as long as there is no confliction in "IP:PORT" confliction for listener socket. You can test your rails application with litespeed on the same IP but different port without touching Apache configruation. After making...

Security is the top priority here, the risk is too high to do that. Another solution is to set AWStats working directory to /home/test/app/current/awstats/, and have Capistrano backup and restore it automatically. I think it is doable.

/etc/ for password file probably is fine, fixed a bug related to authentication configuration in htaccess, fix will be in 2.2.3 release.

Fix for this bug will be in 2.2.3 release.

Eventually, there will be a native FreeBSD AMD64 release. :)

It is not possible to turn this security feature off with current release. However, you can change your vhroot from "/home/test/app/current/" to "/home/test", and adjust the path to rails application if it is relative to the vhroot.

This will be supported in our dedicated load balancer product, which is session aware.

We don't have this problem in our lab, please try leaving "Keep Alive Timeout" unset if you want it to be keepalived forever.

configuration looks fine to me, maybe you can try increase the concurrent level, LSWS uses the first idle cluster node.

Looks like something wrong with the IP address detection, the 32bit compatible layer of amd64 may not be complete. Can you try the linux binary instead? FreeBSD has a linux ABI compatible layer.

This two should have the same result, but I am not sure why htacess does not work well on your case. If you can avoid using .htaccess, you should avoid it.