View All Products
Compute
Storage
Networking
Databases
Services
Developer Tools
Industries
Pricing
Community
Engage With Us
Community ▶ Questions ▶ Persisitant Problem
Log in to Ask a Question

Persisitant Problem

development

I run about 10 wordpress sites from my linode. In Total they get 15-20k uniques a month.

Every now and again (around once a week) The site will become unusably slow, the CPU will jump to 100-200% The Network usage will jump from an average of 16KB/s to a max of 4MB/s and my Disk I/O will jump from an average of 254 to a max of 1340

What is going on here? Am I being DDOSed? I've been trying to resolve this problem for months but haven't been able to find any solution.

I've set my MaxClients to 24 and switched 'KeepAlive' off, upgraded my Linode from a Linode512 to a Linode768 and yet the problem persists. I've installed WP Super Cache on all my sites. What else can I do? I even had Linode migrate my server to see if that was the problem.

A list of diagnostic commands I could run during periods of slowness to help me determine the problem would be great.

Thanks

Mark

7 Replies

top (and iotop, and iftop), which are kind of like the windows task manager; you'll get a nice list of processes that you can sort by CPU usage, memory usage, etc. Top is best for CPU/RAM, iotop for disk, iftop for network. Maybe take a look at your web server's log files to see if there is any unusual activity at the time?

You could also install something like munin.

Can someone take a look at my munin logs and tell me whats going on?

http://109.74.204.14/munin/

I'm far from an expert, but the postfix graph jumps out at me. Are you sending an email newsletter, or updates?

Your server is swapping around 3pm each day (see http://109.74.204.14/munin/localdomain/ … -swap.html">http://109.74.204.14/munin/localdomain/localhost.localdomain-swap.html) which is causing your disk io to go nuts and could be causing postfix to queue mails (postfix could be the problem but I doubt it, 24 mails in queue isn't many).

Since it's regular I'd assume a cron, check your crons to see if anything is running around 3pm and check /var/log/syslog for any unusual entries around that time.

@obs:

(postfix could be the problem but I doubt it, 24 mails in queue isn't many).

I read that as 2,400 (2.4k), which sounds like a lot to me.

I could be reading it wrong though. (I'm looking at the weekly stats)

Hrmm you're right, daily says 24, weekly says 2.4-3.2k, monthly shows 3k, yearly shows 950. Not sure why there's a difference but if any but the daily are correct that would be a problem, I think I'd do some manual log grepping to check how many mails have been sent.

Reply

Please enter an answer
Tips:

You can mention users to notify them: @username

You can use Markdown to format your question. For more examples see the Markdown Cheatsheet.

> I’m a blockquote.

I’m a blockquote.

[I'm a link] (https://www.google.com)

I'm a link

**I am bold** I am bold

*I am italicized* I am italicized

Log in to Reply
Community Code of Conduct