Server Speed


#1

Is anyone else experiencing critically slow speeds? I just moved over from Bluehost and my forum ( http://azhforum.co.uk ) is crawling along. I have submitted a support ticket and had no response yet and there is nothing on the announcements site. I just want to check if anyone else is having the same issue or not.

I’m on Punch server.

Also, it has been this way since yesterday so it is more than 24 hours now.


#2

When my sites are slow, I log into the server and do the ‘w’ command which will show me the load average. It should be way below 5.

-Scott


#3

Please could you explain what you mean? I’d like to try that.


#4

I don’t know what server you’re on, but I’m on yoda and the load averages are absurd.

top - 14:50:13 up 7 days, 3:48, 5 users, load average: 96.95, 52.87, 23.51
Tasks: 6 total, 1 running, 5 sleeping, 0 stopped, 0 zombie
Cpu(s): 9.8% user, 52.1% system, 0.0% nice, 38.1% idle
Mem: 3615716k total, 3495832k used, 119884k free, 10156k buffers
Swap: 6313512k total, 69072k used, 6244440k free, 1324304k cached

PID USER PR NI VIRT RES SHR S %CPU %MEM TIME COMMAND
3658 coolgeek 9 0 2084 2040 1828 S 0.0 0.1 0:00.31 sshd
32355 coolgeek 9 0 1260 1256 1028 S 0.0 0.0 0:00.54 ksh
11930 coolgeek 9 0 2340 2340 1584 S 0.0 0.1 0:00.10 mysql
23959 coolgeek 9 0 2108 2064 1832 S 0.0 0.1 0:00.30 sshd
12785 coolgeek 9 0 1300 1296 1064 S 0.0 0.0 0:00.42 ksh
14995 coolgeek 9 0 1036 1036 844 R 0.0 0.0 0:00.09

(accidentally posted before I was done)

Last week I saw a current load average of 158.

I realize this isn’t Dreamhost’s fault (at least not directly), but if they’re not going to provide us with separate dev/prod environments, they’re going to need to monitor this situation much more closely.

I’ve complained about it before, but they don’t seem to understand the significance of the problem, at least at the support level that answers tickets.


#5

Make sure your account is set to be a shell account in the Manage Users section. Then use SSH (PuTTY) or Telnet to connect yourdomain.com to get the command prompt. Then type either “w” or “uptime” without the quotes to get the load average.

-Scott


#6

Scott is talking about logging into your server shell via ssh and using the command line to execute the “w” command to get a report of the state of the server.

Once logged into the shell via ssh, just type “w” (without the quotes) at the prompt and hit the carriage return key. A report will be printed to the terminal screen. Here is an example of such a report:

16:09:58 up 24 days, 13:48, 3 users, load average: 0.64, 0.61, 0.55
USER TTY FROM LOGIN@ IDLE JCPU PCPU WHAT

The load average is what he was refering to when he said it whould be well under “5”.

–rlparker