Search results

Yes, you can use lsapi, it works in the similar way as FastCGI, just faster due to the optimization in protocol.

It hard to for us to provision the license is used for physical server or on VPS server. 1 or 2 cpu license should be good choice for VPS users.

It is not possible with current LiteSpeed proxy interface. Will add that extra header for https request automatically in 2.1.17 release.

While Xing tuning the web admin interface, you can try let LSWS using httpd.conf directly instead importing httpd.conf into LSWS configuration format. We have user using LSWS with Cpanel successfully, this way, you can still manage vhosts within Cpanel without importing httpd.conf again and...

Thank you for your guide to fix this issue. You can post it in our Wiki if you want. :-)

The problem is that LSWS cannot found the file used for the error page under the missing subdomain. So it reports "detect loop redirection". You can add a static context /index.html pointing to /home/examplecom/public_html/index.html to fix it.

The ErrorDocument directive bolow the rewrite rule does not work, only the vhost customized error page option works. Can you post some debug logging. one working for example.com, one not working.

I think you probably did not configure the listener to vhost mapping properly, you should use "*" or "*.example.com" as the matching domain.

I misunderstood your first post. It may not as difficult as you think to use a "static" context. :) The extra configuration you need to click between a 'static' and 'authentication' context is only one option "accessible", as you can leave "location" blank.

ts77, The load balancer is a standalone server in front of a web server farm.

There is an option, "Dynamic Requests/second", for this kind of single point DoS attack, maybe it is not exactly what you have been doing in PHP, but should help.

I think you can use the "Apache style configure" by inserting apache directives there when you configure a context. :)

I think .htacess should serve this purpose well, should not be a performance hit itself. Actually, I think it may not be a good idea to do DoS detection in an external application, as one instance may not know the big picture and hard to make it accurate. Our builtin DoS prevention feature is...

lshttpd should consumes little CPU cycles, the rest should be taken by PHP processes. So, you need to focus on PHP tuning, like using opcode caching if it has not been used yet. Also, try finding vBulltin add-on modules which can improve performance. Worst case, add more hardware. :-) Our load...

That's strange. My test shows that it works. Unless there is a high priority context matching errors.log, it should not happen. Please try to adjust the sequence number of /eventun/logs/ to the lowest to give it higher priority. If the problem persist, please send us some debug log for analysis.

The problem with bug @ litespeedtech has been fixed. Please send again. Can you please try the upload with non-SSL connection, we need to isolate the problem. Also, it will help a lot if you can send us the error log with DEBUG logging for analysis. Thanks, George

Just rename the log file in a cron job, litespeed will create a new log file automatically. No special care need to be taken. :)

You need to get rid of --with-apxs2=/usr/sbin/apxs , that's for building PHP as Apache2 module, you can only build one PHP SAPI module at a time.

Thanks for the update. :) We will be very happy to supply a RPM package if they don't mind distributing close sourced packages. :)

Do not use the "php" binary installed under /usr/local/bin, that is the command line version. Just copy php-4.4.2/sapi/litespeed/php over.