SMF Support > Server Performance and Configuration

What Godaddy's "Expert Hands" team told me...

(1/3) > >>

trekkie2444:
"Dear Sir/Madam,

Thank you for contacting Expert Hands regarding an Apache Optimization on your 'e-p1' Virtual Dedicated Server.

I have optimized Apache as well as MySQL on your server. Though this may help reduce the overall resource consumption, it is not a guaranteed fix for all resource issues. In optimizing MySQL I found the following:

Load on the server is high partly due to unoptimized database queries. During the review of your server I found that the MySQL server had been running for 3 days and 12 hours. During this time it processed 7,589,346 queries. The MySQL runtime information from withing phpMyAdmin shows the following statistics that indicate an issue with queries being run on the server.

/*Snip*/
Handler_read_rnd     17 M     The number of requests to read a row based on a fixed position. This is high if you are doing a lot of queries that require sorting of the result. You probably have a lot of queries that require MySQL to scan whole tables or you have joins that don't use keys properly.
/*Snip*/

/*Snip*/
Handler_read_rnd_next     1,720 M     The number of requests to read the next row in the data file. This is high if you are doing a lot of table scans. Generally this suggests that your tables are not properly indexed or that your queries are not written to take advantage of the indexes you have.
/*Snip*/

This is probably the most telling statistic applicable. This means that a large number of queries are scanning entire database tables to find the value needed (i.e. SELECT *) when a properly written query would use an index to go straight to the value and read only that one row instead of the entire database table. The reason these queries take so long to complete, and therefore site pages load slowly, is that if a database table has 10,000 total rows, which is a conservative estimate for many database designs then 10,000 rows will need to be evaluated instead of a single row for a properly indexed query. This means that the data will have been retrieved approximately 10,000 times quicker with the indexes. Fixing this would require that you verify you have created indexes in your database tables then rewrite the database queries in your website scripts to use those indexes.

/*Snip*/
Select_full_join     477k    The number of joins that do not use indexes. If this value is not 0, you should carefully check the indexes of your tables.
/*Snip*/

This means there are queries in the site which are joining two full table scans together. This can be extremely slow because it results in a multiplication of the total number of rows. As an example, joining two full tables containing 10,000 rows each would result in a total of 100,000,000 values to be scanned in order to retrieve the value that the query is seeking. To ensure optimal performance you should ensure that all database queries use indexes and never join full tables.

As we are not familiar with how you've built and designed your websites to function we are unable to provide assistance with troubleshooting the code in order to optimize the queries. Your website developer would be the most qualified person to contact in regards to optimizing the database queries in your sites and improving the websites' overall performance.

In regards to the CPU consumption of Apache on your server, I monitored which files were being opened by Apache. The most common files that I say being opened during my review were the following:

/home/onerinconhill/public_html/Sources/PortalBlocks.php
/home/onerinconhill/public_html/Sources/QueryString.php
/home/onerinconhill/public_html/SSI.php
/home/onerinconhill/public_html/Sources/Subs.php


A review of all access logs for the website shows that the following 10 pages have been requested most frequently.

   1331 75.69.210.3 "GET /.xml/?type=rss
   1113 213.115.153.50 "GET /Themes/default/scripts/portal.js?233
   1113 213.115.153.50 "GET /Themes/Flagrantly/scripts/theme.js?rc5
   1113 213.115.153.50 "GET /Themes/Flagrantly/css/index.css?rc5
   1112 213.115.153.50 "GET /Themes/default/scripts/script.js?rc5
   1111 213.115.153.50 "GET /index.php?action=dlattach;attach=1581;type=avatar
   1106 98.227.64.63 "GET /Themes/default/scripts/script.js?rc5
   1104 98.227.64.63 "GET /Themes/Flagrantly/scripts/theme.js?rc5
   1096 98.227.64.63 "GET /Themes/default/scripts/portal.js?233
   1093 98.227.64.63 "GET /Themes/Flagrantly/css/index.css?rc5


Please contact us if you have any further questions.

Regards,
Jon W.
Expert Hands"

Total bull or is there something I should actually be doing to fix this?

trekkie2444:
Anyone?

Roph:
Godaddy are experts at massively overselling - a technique of selling more than you can provide, or more than actually exists. You are fighting for server resources with many, many more users than the server should realistically be able to support comfortably, so you will encounter problems.

You're best off looking for another host.

Aleksi "Lex" Kilpinen:
It's not complete bull, but it is very generic info based on mysql statistics - info that generally does not help much, or mean much, with SMF.

trekkie2444:
Is there anything I can do based on what they told me?

Navigation

[0] Message Index

[#] Next page

Go to full version