I have a VPS level 3 plan with Hostgator, providing a server memory of 768MB. I only run one website on it. It's an affiliate wordpress blog and I have optimized its speed as much as I can - mostly by using the W3 total cache plugin. Its still not loading as fast as I would like to be however, sometimes having first render times of up to 4 or 6 seconds, and sometimes under a second. (Overall load time is usually up to 10 seconds or so) So I checked my resource alert logs in the power panel - and It looks like every day I have a few yellow and red alerts. Here's an example list:
Nov 30, 2010 10:15:39 AM Green zone slmmemorylimit
Nov 30, 2010 10:10:39 AM Yellow zone slmmemorylimit
Nov 30, 2010 10:09:39 AM Red zone slmmemorylimit
Nov 30, 2010 06:16:39 AM Green zone slmmemorylimit
Nov 30, 2010 06:14:39 AM Red zone slmmemorylimit
Nov 30, 2010 02:33:39 AM Green zone slmmemorylimit
Nov 30, 2010 02:28:39 AM Red zone slmmemorylimit
Nov 30, 2010 01:30:39 AM Green zone slmmemorylimit
Nov 30, 2010 01:29:39 AM Yellow zone slmmemorylimit
Nov 30, 2010 01:26:39 AM Green zone slmmemorylimit
Nov 30, 2010 01:25:39 AM Yellow zone slmmemorylimit
Nov 29, 2010 12:29:39 PM Green zone slmmemorylimit
Nov 29, 2010 12:24:40 PM Yellow zone slmmemorylimit
Nov 29, 2010 12:23:39 PM Red zone slmmemorylimit
Nov 29, 2010 12:22:39 PM Yellow zone slmmemorylimit
Nov 29, 2010 10:27:39 AM Green zone slmmemorylimit
Nov 29, 2010 10:23:39 AM Yellow zone slmmemorylimit
Nov 29, 2010 10:22:39 AM Green zone slmmemorylimit
Nov 29, 2010 10:20:39 AM Red zone slmmemorylimit
Nov 29, 2010 10:19:40 AM Yellow zone slmmemorylimit
Is this normal for my type of traffic, do I need to be concerned yet? According to awstats my traffic for those two days was:
Day------Visits---Pages---Hits---Bandwidth
28 Nov---2144---5719---74380---1.10 GB
29 Nov---2844---8061---98895---1.49 GB
and according to the traffic log in power panel my traffic usage for those two days were:
Nov 28 601.00 MB Incoming 1,699.34 MB Outgoing
Nov 29 492.12 MB Incoming 1,323.03 MB Outgoing
First of all, I take it that its normal that the bandwidth amount shown in the awstats is different than the overall MB amount shown in the traffic log?
Now - how can I find it out what exactly is causing the resource alerts? E.g. what script or applications? Or which visitors? Maybe there some undesirable referrers that drain unnecessary resources. (In my htaccess, I have hotlink prevention set as well as a couple of referrers already blocked)
Is it possible that some of this comes from too much receiving mail? I do have several email addresses set up and get a fair amount of stuff plus a lot of failed delivery notices on an address that is being spoofed (SPF is set). If I delete some of my email addresses (most of them I dont really use anyway), would that help? Or is it that if people/websites continued to sent email to them that it would still drain resources on my server?
I read on some webpage that modifying the Apache Memory Usage Limit can be a solution. But since Im the only user on my VPS, with usage set to unlimited I dont think this can help me in any way?
Is it possible to do some setting that will tell the kernel what processes to kill first in case of a red alert, before going random?
Last but not least, what is TOP command and can it help me here?
Obviously Im asking all these questions in the hope of tweaking/reducing my server memory as I dont think I can afford an
upgrade to level 4 which would be 30 dollars more (with 1,1GB memory).
Thanks so much for any insight and help,
Dennis
