Search results

For rails 3.0, you need to use RailsRunner in http://www.litespeedtech.com/support/forum/showthread.php?t=4025

This function has been added to our 4.1RC4 release, which has been used in production by a few users with special needs.

mod_security makes big changes from release 1.9 to 2.x, our implementation is based on 1.9, so not all features in 2.x are support. We will add more 2.x features in the future. those features are probably only used in 5% of rules. But if you require 100% compatibility with mod_security, you...

I do not think we will do anything to those "errors", to just make RedBot happy. Nothing is break. You have to take it as is or use other web server if it bothers you.

It has been constantly updated. Just download and compare.

You cannot use standard edition as there are more than 5 vhosts. You have to use VPS license at the minimum.

Have you unchecked "httpd" service from server manager? it should be stay checked.

Are you using suphp with Apache or mod_php? If you do not use suphp, set "PHP suEXEC" to "No". Maybe you can try a reinstall from within WHM plugin.

One of the factor is what "ab" did at the end of the benchmark tests. It can affect one web server more than to the others. To reduce the impact of this, just combine multiple runs into one large run. Another factor is the way web server works, perfork, threaded or event-driven. perfork has poor...

Sorry about that, this issue is addressed in 4.0.19.

yes, httperf is better in this regard. However, httperf does rate limiting, not like "ab" sending request as fast as possible, so, httperf may report lower req/s than "ab" if the rate limit cannot push the server to 100%. Opcode cache will help for all web servers. There are other respects...

It may have something to do with how "ab" works, it send more than 1000 requests during the test, and close all connection once number of responses reaches 1000. It is better to do one test run of ab -k -n 5000 -c 200 ... than running ab 5 times with "-n 1000". In theory, the result should...

litespeed PHP was installed to /usr/bin, should be installed to /usr/local/lsws/lsphp5/bin, Need to remove "--prefix=/usr" from the configuration option.

It is about the same, you can also change "Max connections", "PHP_LSAPI_CHILDREN", to 10 and 100, see which one does better. PHP_LSAPI_MAX_REQUESTS can be increased to a larger number. How many total requests served for each run? I think it should be >1000. There is likely some factor affect...

why you want to use the super slow CGI PHP while super fast LSAPI PHP is the default?

I wonder if you use mod_php or suPHP with Apache? If uses mod_php, make sure "PHP suEXEC" is off to match Apache security model.

add environment LSAPI_ACCEPT_NOTIFY=1 LSAPI_AVOID_FORK=1 to lsphp5 external app should boost Litespeed no-cache results.

smart keepalive will close connection after serving a response with MIME type text/* regardless it is from PHP or static file. So, "Smart Keepalive" should be off when you did the benchmark tests.

Likely a problem with kernel socket limits, there should not be big differences when you run the benchmark test multiple times. Try adding "-k" option to "ab".

duplicate post, see the answer to the other post.