More deployment issues

Add new comment
Your name - will be displayed with your comment. Required
Your email address - will not be displayed. Required
Your homepage URL - your name will link to this. Optional
Your comment - textile syntax, 1000 chars max. Required

I have migrated the server to a new host, a XEN VPS [1] with 96 MB RAM, hosted on an Athlon 64 3000 (1.8 Ghz).

The software setup is the same, I only updated ningx to version 0.4.14 and the distribution is 64bit.

Here's the output of free(1) after the site has been running for a couple of minutes:

~$ free
             total       used       free     shared    buffers     cached
Mem:         93000      90904       2096          0      17632      24528
-/+ buffers/cache:      48744      44256
Swap:        98296         20      98276

This is a list of the processes running (kernel daemons excluded):

  • init
  • syslogd
  • klogd
  • getty (6x)
  • sshd
  • nginx
  • python (CherryPy+TG)

and a bash shell.

I'll have to watch this over the next few days, but it seems, that the amount of memory is enough. top(1) shows ~180MB virtual memory usage for python though, but it doesn't seem to affect performance.

To make transition easier, I also set up a new dynamic domain name, paddyland.serveblog.net, but the old name paddyland.no-ip.org now points to the new server as well until further notice.

If I have the time, I plan to do a few perfomance measurements with httperf or a similar tool to get a proper confirmation that the server specs are sufficient for my needs.

[1]Thanks for the folks at contextshift.eu for providing the hosting.

Reader comments

There are no comments on this article yet. Add a comment now...