High disk I/O rate


I am wondering how I can tell what is using up all the disk I/O on my linode. I have been receiving emails about a high I/O rate and can't figure out why.

> has exceeded the notification threshold (1000) for disk io rate by averaging 6776.14 for the last 2 hours

It seems way to high for the time period these occur (overnight during slow usage hours).

Is there any command I could run to tell what is using the current disk I/O? I could easily set that up to run as a cron during these problematic hours and see if I can capture anything useful.

5 Replies

iotop might do the job.

By the way, do you have any disk-intensive cron jobs scheduled in the night? Such as backups, log analysis, etc.?

I haven't configured any cron tasks yet. I am currently pulling backups weekly via a ssh (ssh key) remote machine. My issues are nightly.

iotop looks like it may do the job. I should be able to setup a task to run for an entire hour every second and dump the I/O usage.

As for running crons, it is default for a ubuntu 10.4 server.

> root@server:~# ls /etc/cron.*/


nsd3 php5


apache2 apt aptitude bsdmainutils dpkg locate logrotate man-db ntp popularity-contest standard






Does the io load time match the logrotate time? That can chew up a lot of IO.

Disable the cronjob for locate. That sweeps the file system nightly to build an index of where files are located.

Ahh, locate does make sense since I don't get the emails every night. Now that I think about it, I usually get them after I have been adding, removing or updating software.


Please enter an answer

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

Community Code of Conduct