Search results

Have you used any kind of PHP opcode cache? eAccelerator or APC? Have you utilize something like memcache? Does the web server connected to backend DB server via a dedicate link? or shared with the web traffic? The trial license does not limit the performance. Please try the PHP opcode cache...

It is likely caused by some DB problem when you see the request processing rate decrease. So, optimize your DB or improve the SQL, if you added new code to your site, you should review it as well.

We will add your site to our case studies page, please send some description about our site.

he can test the server string and make it work with both Apache and litespeed. The change is pretty simple. How is the boss there? :) Yes, it should solve your problem and 20gb should be enough. Maybe next week.

I also wonder if Apache work properly on port 79 with vhosts configured that way. ;)

Yes, you need to remove the listener for port 80. For RoR setup, should use this http://www.litespeedtech.com/support/wiki/doku.php?id=litespeed_wiki:rails_hosting however, it is Enterprise edition only.

We will add some logic to detect this kind of situation and make it work.

How about send the output of the following command to us? cd lsws/bin strace ./lshttpd -d

Yes, there should not be any problem. Actually, LSAPI is the only recommended option for RoR and PHP support.

Are you using the 32bit package on 64bit Linux? Is that possible for our staff member to take a look at it?

With our enterprise edition, the IP to country DB from MaxMind can be used to lookup the country code from IP and ban them with rewrite rules. :)

Those probably are corpse of crashed PHP processes. Might be caused by PHP bugs.

First, it will use tmp swap for file larger than 1MB. Second, we keep the part of a swap file being mapped into process memory space as small as possible to avoid exhausting the address space of a 32bit process. You should continue to pursue the "internal redirect" route by all means though...

You need to look into iptables, or hire an admin to do that for you.

We will try to have your issue addressed in up coming 3.3 release. It may not be the issue of your TMP partition. If you run lshttpd in 32bit system, you may run into virtual memory limit problem if there a few upload/download session for large files. 3.3 release will have it addressed. It is...

Can you please forward the problem configuration file to bug [at] litespeedtech {dot} com. We will take a look. The rewrite rule with non-ASCII characters may not work well as it is not designed to support it. But anyway, we need to address the XML issue first.

I think it is system wide timezone configuration, you can try google "Linux change timezone"

You can change the PHP configuration in lsws/php/php.ini or wherever the php.ini located. You can consult to phpinfo() page for the php.ini being used.

Thanks for the work around, it should benefit others who come across the same problem.

That's why you run out of buffer/swap space. LSWS caches the whole response body produced by PHP, so all download files will be stored in the buffer/swap. You should have your developer to remove the streaming code and replace it with a internal redirect I mentioned earlier. Your site will be...