Slow OpenVZ servers.


New Member
Hey Guys

I've got two OpenVZ servers from UrPad with 768MB RAM - 256MB SWAP, lately I've been experiencing very slow login speeds to SSH and anything to do with resolving hostnames..

Before you mention it, yes I did submit a support ticket

Could you please have a check now, we have tweaked the settings and suspended abuser on this node.

Dennis Ross
Support Supervisor - A FTNHosting Company
The issue is still persisting.. UrPad's support has gotten very very sloppy since the move to another company..

~# ping
PING ( 56(84) bytes of data.
64 bytes from ( icmp_req=1 ttl=57 time=0.392 ms
64 bytes from ( icmp_req=2 ttl=57 time=0.369 ms
64 bytes from ( icmp_req=3 ttl=57 time=0.379 ms
64 bytes from ( icmp_req=4 ttl=57 time=0.396 ms
64 bytes from ( icmp_req=5 ttl=57 time=0.267 ms
64 bytes from ( icmp_req=6 ttl=57 time=0.288 ms
64 bytes from ( icmp_req=7 ttl=57 time=0.408 ms
64 bytes from ( icmp_req=8 ttl=57 time=0.311 ms
64 bytes from ( icmp_req=9 ttl=57 time=0.354 ms
64 bytes from ( icmp_req=10 ttl=57 time=0.406 ms
64 bytes from ( icmp_req=11 ttl=57 time=0.413 ms
64 bytes from ( icmp_req=12 ttl=57 time=0.382 ms
64 bytes from ( icmp_req=13 ttl=57 time=0.301 ms
64 bytes from ( icmp_req=14 ttl=57 time=0.483 ms
64 bytes from ( icmp_req=15 ttl=57 time=0.309 ms

Whilst initiating traceroute and ping it pauses for a good minute or two before doing anything.

:~# traceroute
traceroute to (, 30 hops max, 60 byte packets
1 ( 0.043 ms 0.011 ms 0.010 ms
2 ( 0.289 ms 0.321 ms 0.397 ms
3 colo-lax6 ( 0.284 ms 0.359 ms 0.363 ms
4 core02 ( 0.327 ms 0.387 ms 0.433 ms
5 ( 0.244 ms 0.248 ms 0.229 ms
6 ( 0.478 ms 0.651 ms 0.604 ms
7 ( 0.911 ms 1.008 ms 0.826 ms
8 ( 0.304 ms 0.300 ms 0.300 ms

Not a disk issue

~# dd if=/dev/zero of=test bs=64k count=16k conv=fdatasync
16384+0 records in
16384+0 records out
1073741824 bytes (1.1 GB) copied, 2.4224 s, 443 MB/s

Wget's are pain full even github takes a few minutes to resolve.... following guide -

python --share
Retrieving configuration...
Retrieving server list...
Testing from ColoCrossing (
Selecting best server based on ping..
Waited 5 minutes and gave up on that test.

Cachefly time

--2013-06-28 04:01:55--
Connecting to||:80... connected.
HTTP request sent, awaiting response... 200 OK
Length: 104857600 (100M) [application/octet-stream]
Saving to: `100mb.test'

100%[=================================================>] 104,857,600 59.4M/s in 1.7s

2013-06-28 04:04:12 (59.4 MB/s) - `100mb.test' saved [104857600/104857600]

Again resolving took ages.. speed is good though..

I have reported this twice, it has now spread to another server, any ideas on what I can suggest to them?

D. Strout

Resident IPv6 Proponent
Well, your actual Internet speed is fine, as evidenced by wget and ping. It must be your DNS resolving. You could switch the server to use Google's DNS.


New Member
You could switch the server to use Google's DNS.
Apparently this was done the first time on the first server, not the second one.


I have updated google nameserver's and its working fine now. Please do verify from your end.

/# ping

PING ( 56(84) bytes of data.

64 bytes from ( icmp_req=1 ttl=47 time=23.1 ms

64 bytes from ( icmp_req=2 ttl=47 time=22.6 ms

--- ping statistics ---

2 packets transmitted, 2 received, 0% packet loss, time 1001ms

rtt min/avg/max/mdev = 22.648/22.903/23.159/0.296 ms


Dylan Shelton

URPad Technical Support Operator
Last edited by a moderator: