The first couple of days on Xen, everything was bliss. It was fast. But now it's so slow... It takes a frustratingly long time just to get a password prompt via ssh, and the once I do log in, the motd displays, but again it takes forever for the command line to display. I'm not even doing anything that heavy, just lighttpd + php.

If I reboot, it becomes more responsive for a short period of time, but 20-30min in it gets down right glacial again.

I don't know if the box is overburned with too many users, or if it's Xen, or some configuration issue. But it's been unreliable and pretty much unusable for the past week.

Mind helping me out a little? Can you paste the output of "cat /proc/swaps" and "vmstat 1 10"?

-Chris

/proc/swaps is empty

root@webmages:~ # cat /proc/swaps 
root@webmages:~ # 

root@webmages:~ # vmstat 1 10
procs -----------memory---------- ---swap-- -----io---- --system-- ----cpu----
 r  b   swpd   free   buff  cache   si   so    bi    bo   in    cs us sy id wa
 0  0      0   3864   8376  89680    0    0     0     0    1     2  0  0 100  0
 0  0      0   3864   8376  89680    0    0     0     0    6     2  0  0 100  0
 0  0      0   3864   8376  89680    0    0     0     0    9    10  0  0 100  0
 0  0      0   3864   8376  89680    0    0     0     0    5     4  0  0 100  0
 0  0      0   3864   8376  89680    0    0     0     0    5    10  0  0 100  0
 0  1      0   3864   8384  89740    0    0     0    28    5     7  0  0 100  0
 0  0      0   3864   8384  89740    0    0     0     0    7    10  0  0 100  0
 0  0      0   3864   8384  89740    0    0     0     0    3     4  0  0 100  0
 0  0      0   3864   8384  89740    0    0     0     0    6    10  0  0 100  0
 0  0      0   3864   8384  89740    0    0     0     0    6     8  0  0 100  0

Top reports:

top - 21:07:40 up 2 days, 11:23,  1 user,  load average: 0.00, 0.00, 0.00
Tasks:  61 total,   1 running,  60 sleeping,   0 stopped,   0 zombie
Cpu(s):  0.0% us,  0.0% sy,  0.0% ni, 100.0% id,  0.0% wa,  0.0% hi,  0.0% si
Mem:    123072k total,   119292k used,     3780k free,     8412k buffers
Swap:        0k total,        0k used,        0k free,    89712k cached

I did just notice that my swap wasn't actually enabled -- I created a new swap partition when I moved over, and it's on a different device (hda4 instead of hda2). It would make sense if that was part of the problem..

That was it! No swap was the problem. It couldn't page out what wasn't actively being used.

Chalk it up to another case of user error.

hrm... actually it fixed it momentarily.
I'm stuck after the motd again with not command prompt. Even though it uses more RAM I think I might try switching back from dropbear to openssh

It would help if you could catch me in the #linode-xenbeta channel next time you experience slowness. That way, I can try a few things and see if they affect the interactivity you're experiencing.

http://www.linode.com/irc/

Thanks,
-Chris