This is a continuation of a series of posts about reworking my website into one that is entirely WordPress based. The series starts here.
I decided to take a look at the new WordPress-based server’s response time. Here’s a historical look over the last week:
Oops! The response time went from a few hundred milllseconds to about 2 seconds when I went live with the WOrdPress-based version of the site.
This is going to take some looking in to, but I’ll bet it’s because of the slide show that now runs on the home page of kasson.com. It two seconds OK, considering the slide show? I don’t know.
[Added 10/13. I added monitoring for this site, which doesn’t have the slide show. It showed only a slight improvement:
So maybe this is a penalty imposed by WordPress. ]
In an unrelated performance issue, you may have noticed that this blog loads faster than it used to. In testing various backup plugins, i had several fail on this blog, and work on my other two WordPress instances. One of the tech support people at BackupBuddy pointed out that there was a table in the WordPress database that was taking up 2/3 of the half a gigabyte database. On his advice, I excluded it from the database backup, and the backup went much faster. But most backup programs don’t give you the fine control over what tables are backed up that BackupBuddy does.
That got me thinking. The name of the table, wp_statpress, meant that I must have been written by an obsolete WordPress user statistics analyzer that I stopped using a month or so ago when I found that it was no longer being supported. I guess the uninstall program didn’t clean up the database. So I went to the WordPress Codex to find out how to remove database tables. They recommended a tool called phpMy Admin. I opened up Plesk, and managed to find it (the fact that the online Plesk documentation refers to older versions of the suite didn’t help any). I found the table, checked the entry, went down to the bottom of the page, and picked “Clear”. Then a saw a really scary “are you sure” window. I held my breath, hit OK, and tested the site, heart rate escalated in spite of the fact that I had a handful of backups for it. It seemed to work.
I did a backup without excluding the table. It was the same size as the ones that I’d done with the table excluded. So it seems to have worked.
And the site is more responsive. Sometimes you win one.