slowness on Host 45

Anyone using Host 45 notice lots of slowness this last week? Every once and a while my linode starts swapping like crazy. I'll be typing away then something like 'ls' while hang for 10+ seconds.

I'll immediately check vmstat, top, and my iotokens and all is good. My CPU usage is 98% idle, I have 399900+ iotokens available… but si and so from vmstat are going nuts. Also I notice the load will reach upwards of 7 before calming down again.

Also, is the load computed for the entire host or just my linode?

3 Replies

I am on Host45. I do not have any empirical data to share, but I have noticed my Linode having "delays". High load readings with low CPU usage, I/O rate and tokens look fine.

Rebooting my Linode does not seem to have any effect. Today it took 22 minutes for Linode to reboot. Also I am unable to connect via LISH at host45.linode.com. I get an immediate close connection response.

Not sure if it is related, but I have noticed the following errors in my logs:

Jun 3 06:30:15 www kernel: _allocpages: 0-order allocation failed (gfp=0xf0/0)

May 26 10:15:12 www kernel: _allocpages: 0-order allocation failed (gfp=0x1d2/0)

May 26 10:45:13 www kernel: _allocpages: 0-order allocation failed (gfp=0x1d2/0)

May 26 10:45:13 www kernel: _allocpages: 0-order allocation failed (gfp=0x1d2/0)

This morning my linode on Host45 became unresponsive. I was hitting it pretty hard (lots of tasks) and it wasn't swapping too bad. But it just slowed to a crawl.

I tried to get to lish but kept getting the following error message:

sshexchangeidentification: Connection closed by remote host

I then went to the web and executed a Shutdown job. It took over a half hour for that job to finish. Then the Boot job took over a half hour to complete. The jobs kept in the "waiting for host" state.

After my linode booted it still remained very slow and the host load was 5+.

Note that I still can't get to lish even after rebooting my linode. Now I get a password prompt but then it just hangs.

@drware:

Not sure if it is related, but I have noticed the following errors in my logs:

Jun 3 06:30:15 www kernel: _allocpages: 0-order allocation failed (gfp=0xf0/0)

May 26 10:15:12 www kernel: _allocpages: 0-order allocation failed (gfp=0x1d2/0)

May 26 10:45:13 www kernel: _allocpages: 0-order allocation failed (gfp=0x1d2/0)

May 26 10:45:13 www kernel: _allocpages: 0-order allocation failed (gfp=0x1d2/0)

This means your Linode ran out of both RAM and swap and further attempts to allocate memory failed.

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

Community Code of Conduct