I'm currently working migrating a WordPress installation (live on the web) from one hosting solution to another. In doing so, I am testing the new hosting solution (shared server) by editing my local host file, so that
www.domain.com
points the IP of hew new solution. On the old host, a page load was averaging ~8 seconds, which is why we are moving hosts. But when I test the new server using my hosts file, the web page loads, and I've verified it's loading from the new server, but it's load time is 30+ seconds.
Is this an issue with the OS having to deal with the hosts file on every request?? The load time on the new server is very even amongst the page requests. It's as if every request time is multiplied by 4.
I need to figure this out before I move my client to a much more awful solution.
I've tested changing the hosts file on OS X 10.10.4 and Windows 7 Enterprise with the same result; pointing www.domain.com to the new server results in 30+ second page loads.
Is there another way to test a domain against a shared server environment, so I can see if this is a local OS issue and not their servers? An IP ping of the shared server stays in the 35ms range so it doesn't seem unhealthy.
Edit Traceroute looks fine, 19 hops.
Edit I was able to get an SSH login. The result of top
looks like this:
top - 12:19:12 up 58 days, 10:27, 2 users, load average: 29.86, 38.15, 35.39
Tasks: 2 total, 1 running, 1 sleeping, 0 stopped, 0 zombie
Cpu(s): 44.0%us, 13.1%sy, 0.0%ni, 35.8%id, 6.1%wa, 0.5%hi, 0.5%si, 0.0%st
Mem: 32953672k total, 31990688k used, 962984k free, 2911692k buffers
Swap: 8388540k total, 1100716k used, 7287824k free, 15149856k cached
PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
11521 lifetiu1 20 0 14952 984 796 R 1.9 0.0 0:00.01 top
18995 lifetiu1 20 0 11528 1628 1292 S 0.0 0.0 0:00.04 bash
Is that Mem use enough to cause this sort of slow down? If so, that's pretty indicative that I should recommend my client go to another host.