Omega Owners Forum
Chat Area => General Discussion Area => Topic started by: aaronjb on 16 December 2010, 19:26:33
-
I'm having a lot of trouble with pages stalling and so on, and it usually seems to be waiting for images.omegaowners.com to return a response..
It could be Virgin being chocolate-teapot-y, or it could be OOF.. anyone else seeing it?
FWIW, for those that'll know what it means:
trinity:~ aaronjb$ ping images.omegaowners.com
PING images.omegaowners.com (69.163.225.236): 56 data bytes
64 bytes from 69.163.225.236: icmp_seq=0 ttl=51 time=184.019 ms
64 bytes from 69.163.225.236: icmp_seq=1 ttl=51 time=183.853 ms
64 bytes from 69.163.225.236: icmp_seq=2 ttl=51 time=183.282 ms
^C
--- images.omegaowners.com ping statistics ---
3 packets transmitted, 3 packets received, 0.0% packet loss
round-trip min/avg/max/stddev = 183.282/183.718/184.019/0.316 ms
trinity:~ aaronjb$ ping www.omegaowners.com
PING www.omegaowners.com (217.155.56.220): 56 data bytes
64 bytes from 217.155.56.220: icmp_seq=0 ttl=53 time=40.220 ms
64 bytes from 217.155.56.220: icmp_seq=1 ttl=53 time=49.661 ms
^C
--- www.omegaowners.com ping statistics ---
2 packets transmitted, 2 packets received, 0.0% packet loss
round-trip min/avg/max/stddev = 40.220/44.941/49.661/4.720 ms
images.oof.com seems to be quite a bit more sluggish than www and it looks like the problem is somewhere in comcasts network as it transits the atlantic:
trinity:~ aaronjb$ traceroute images.omegaowners.com
traceroute to images.omegaowners.com (69.163.225.236), 64 hops max, 52 byte packets
1 dd-wrt (192.168.1.1) 3.351 ms 0.767 ms 1.366 ms
2 10.159.44.1 (10.159.44.1) 9.137 ms 7.070 ms 7.526 ms
3 winn-core-1b-ge-018-950.network.virginmedia.net (62.253.121.189) 9.054 ms 8.612 ms 8.421 ms
4 winn-bb-1b-so-130-0.network.virginmedia.net (213.105.175.33) 7.461 ms 8.015 ms 8.437 ms
5 winn-bb-1a-ae0-0.network.virginmedia.net (213.105.172.161) 8.030 ms 8.067 ms 8.394 ms
6 manc-bb-1b-as7-0.network.virginmedia.net (212.43.163.189) 13.495 ms 13.321 ms 14.517 ms
7 tele-ic-3-ae0-0.network.virginmedia.net (212.43.163.70) 22.116 ms 19.851 ms 20.339 ms
8 45-236-228-213.packetexchange.net (213.228.236.45) 111.433 ms 113.278 ms 110.187 ms
9 te-2-10-0-2-cr01.ashburn.va.ibone.comcast.net (68.86.84.237) 112.031 ms
te-2-10-0-4-cr01.ashburn.va.ibone.comcast.net (68.86.84.241) 111.197 ms 111.167 ms
10 pos-1-8-0-0-cr01.atlanta.ga.ibone.comcast.net (68.86.87.90) 126.394 ms 125.894 ms 125.825 ms
11 pos-1-11-0-0-cr01.dallas.tx.ibone.comcast.net (68.86.85.221) 148.734 ms 147.125 ms 148.178 ms
12 pos-0-11-0-0-cr01.losangeles.ca.ibone.comcast.net (68.86.87.37) 181.677 ms 181.722 ms 182.209 ms
13 as26347.losangeles.ca.ibone.comcast.net (75.149.228.206) 185.258 ms 188.922 ms 181.771 ms
14 ip-66-33-201-114.dreamhost.com (66.33.201.114) 183.691 ms 183.892 ms 188.830 ms
Whereas OOF is hosted somewhere decent (like TBs house?) via Zen and it's nice and fast:
trinity:~ aaronjb$ traceroute www.omegaowners.com
traceroute to www.omegaowners.com (217.155.56.220), 64 hops max, 52 byte packets
1 dd-wrt (192.168.1.1) 1.102 ms 1.907 ms 0.765 ms
2 10.159.44.1 (10.159.44.1) 8.362 ms 36.792 ms 7.798 ms
3 winn-core-1b-ge-019-951.network.virginmedia.net (62.253.121.193) 8.538 ms 10.060 ms 8.027 ms
4 winn-bb-1b-so-130-0.network.virginmedia.net (213.105.175.33) 7.878 ms 7.508 ms 8.345 ms
5 manc-bb-1a-ae6-0.network.virginmedia.net (212.43.163.185) 159.930 ms 14.536 ms 14.139 ms
6 tcma-ic-1-ae0-0.network.virginmedia.net (212.43.163.226) 14.570 ms 14.409 ms 12.852 ms
7 zeninternet.edge-ix.net (212.121.34.40) 15.497 ms 18.549 ms 19.890 ms
8 ge-2-0-0-0.cr2.wh-man.zen.net.uk (62.3.80.50) 14.348 ms 14.838 ms 13.129 ms
9 ae2-0.cr2.mbr-roch.zen.net.uk (62.3.80.54) 14.744 ms 14.087 ms 13.549 ms
10 galileo-fe-3-1-172.hq.zen.net.uk (62.3.80.174) 21.040 ms 13.427 ms 16.010 ms
-
Any transatlantic link will have latency. Remember, ping is a awful tool for speed checking, as every transit and border router out there will give it minimal priority.
150ms - 200ms is about right for many transatlanic links, and the images server is the other side of the USA as well....
-
mines ok
-
Any transatlantic link will have latency. Remember, ping is a awful tool for speed checking, as every transit and border router out there will give it minimal priority.
150ms - 200ms is about right for many transatlanic links, and the images server is the other side of the USA as well....
True, actually, for some reason I didn't read that back and think about it before posting :D
Hmm.. I'm definitely seeing it fail to return a response in a timely manner quite often, though (and I'm counting in minutes & seconds rather than ms there ;)) ..
Maybe it's just Virgin being chocolate-teapot-y
-
One of my ISPs uses Level3 to get across the water, the other use Above.
Both returning an increase of 120-150ms between the UK hop and the East coast US hop
-
averaging 185ms here, which is about right for this time of night :y
-
Any transatlantic link will have latency. Remember, ping is a awful tool for speed checking, as every transit and border router out there will give it minimal priority.
150ms - 200ms is about right for many transatlanic links, and the images server is the other side of the USA as well....
True, actually, for some reason I didn't read that back and think about it before posting :D
Hmm.. I'm definitely seeing it fail to return a response in a timely manner quite often, though (and I'm counting in minutes & seconds rather than ms there ;)) ..
Maybe it's just Virgin being chocolate-teapot-y
I think others have seen similar.
For me, on both Zen and BT Business lines, its working as expected for shared hosting solutions (which is why we only use that server for static elements).
I have a theory about DNS for it, but need to run more intense tests... ...after Christmas ;D
-
;D Drink first, investigations later.. sounds like a plan!
I've not been affected so far.. though I did just notice Virgin have upgraded my upstream to 5Mbit - before they did that in this area they were already reporting a congested upstream link, so if they've upgraded everyone without fixing that and now all the kiddywinkies are merrily torrenting away the backhaul will be so oversubscribed that the rest of us are suffering :(
*mutters*
-
It was hanging here as well, Must be a Virgin issue
TB we need more virgins to sacrifice to the server gods
-
Bit slow occasionally. Seems random, but can tell it's images as it's smileys etc that hold it up. Other occasions it's fine. :-?
-
Has it only really been since the 9th?
Does it happen only on non-IE browsers?
-
Possibly since the 9th. Safari iPad gayness here.
-
its worse on IE here, tho cant see how browser would affect ping times?
-
Has it only really been since the 9th?
Does it happen only on non-IE browsers?
Getting the sluggish page loads with IE9 and firefocks.
-
Hang on, let me get rid of those crappy, gay google optimisations... >:(
-
google optimsations removed from images, but I have kept them on http://theboy.omegaowners.com if any of those affected people can have a play
-
Test
-
omegaowners.com pings at 90ms
images.omegaowners.com 185ms
theboy.omegaowners.com 185ms
browser hangs while posting this, loads theboy.omegaowners instantly, but variable delay in loading forum and functions.
-
Again
-
This time images appeared instantly on reply page, previos two posts where much slower.
But it's all usable, wouldn't worry too much tbh.
-
Doesn't seem to be hanging as much as before.
-
Might take a while for it all to drop from caches etc
-
hanging here as well and i am on with sky :-?
-
Hmm, it does seem to have got rather more 'snappy' now..
-
Working better here too. Not hanging before "busy" blocks of icons, for a start.
Was there a DNS hiccup this yesterday morning, by the way? Nothing at omegaowners.com was resolving for me for a while. Forum appeared to be working with a local bodge in place.
Kevin
-
Working better here too. Not hanging before "busy" blocks of icons, for a start.
Was there a DNS hiccup this yesterday morning, by the way? Nothing at omegaowners.com was resolving for me for a while. Forum appeared to be working with a local bodge in place.
Kevin
Yes - we had 2 senior 3rd line Linux peeps looking into the problem, before one of those engineers decided to check the primary DNS server (we had ruled it out because neither authorititive DNS server was resolving). Some silly idiot* hadn't reset the TTL on the domain after the last migration.
* Me :-[
-
Snigger, thanks TB.... You love all this stuff really. ;)