Search results

You can apply a free trial of the Enterprise edition. It will work a lot better than the standard edition with cPanel.

It is hard to say which version is the most reliable one, as different user may trigger different bug. My suggestion is to try 5.2.9 and 5.2.10 on your server see which one is the best for you.

That's due to PHP crash in middle of a request. If you use 5.2.10, you should fall back to 5.2.9, there are stability issue with 5.2.10.

4.0.5 should have addressed that. I might be wrong, from the bug report I got, it relates to "Content-Encoding: UTF-8" header set by PHP. LSWS wont double compress response body.

If you using PHP suEXEC, have no effect. Instead, should set "Max Idle Time" to "10" seconds. Initial Request Timeout can be set longer if user has difficulty to process a script takes long time to finish. Connection Keepalive Timeout should be -1 to keep the connection alive as long as PHP...

You can try reset from command line /path/to/lsws/admin/misc/admpass.sh

You need to stop LSWS first. Then you can move the lsws directory to some where else. "grep" you configuration files for "/etc/lsws", if found any, change it to the new location. Then try to start LSWS from the new location.

I think you need to work on the MySQL DB side, the bad performance is usually caused by bad SQL query or missing table index in the DB. When the site is slow, in addition to slow query log, you can check "mysqladmin processlist" output, it will show which query blocks all other requests. Then...

It is bad idea to install lsws under /etc directory, you can edit the configuration file lsws/conf/httpd_config.xml to remove /etc from access denied directory list. recommended location is /usr/local/lsws or /opt/lsws .

You can safely ignore those. Maybe some corrupt rewrite rules in htaccess file.

It was caused by hanging lsphp5 processes. will look into the root cause if it happen again.

We will check what we can do about it.

Maybe those has been cached, try cleaning up the cache. Install FireBug and check the response header for those requests.

Sorry for the late reply. Yes, you can safely remove any log file produced by LSWS. LSWS has built-in log file clean up feature, to remove log file older than certain days, looks like it does not work that well, will investigate.

Should be loop redirection caused by rewrite rule or RedirectXXXX directives in .htaccess for that domain.

Yes, inside virtual host section. if used inside the main httpd.conf, it will be default apply to all vhosts.

checked the server and the conclusion is that PHP 5.2.10 is not stable, give random crash causes 503 errors. LSWS 4.0.5 + PHP 5.2.9 is super reliable and no 503 error at all. So, be cautious upgrading to PHP 5.2.10.

set "Max Connection" to "1", see if it help or not. Maybe the 256M VPS limit has been hit due to multiple "ruby" process running, each process use a lot of memory.

Can you provide a section of debug log for a request result in 500 error?

How about access server A directly with client, does it gzip and give the correct gzip header? What the request headers send by Client? If you use tcpdump to get the whole request and response at server A, and B, it will be helpful.