08-11-2012 12:30 - edited 08-11-2012 12:38
I appear to have very long DNS times since getting the Superhub. Overall speed to a low-ping site on speedtest comes out at about 85Mb DL, 4Mb UL which is OK, assuming I've not had the UL speed doubled yet, but I guess could be better. [retested just now, both direct connection through router & back to modem mode through my own gateway, 105 DL, 4.8 UL, very nice]
The real problem, though, is the huge lag time when doing something as simple as opening a web page, 5 or 10 seconds at worst, then the page flashes in at full speed. This is wired, from any machine in the house.
I usually run my own gateway, with firewall, NAT, DNS & DHCP, & have the superhub in modem mode. Right now, I have a single machine connected, in router mode, to be able to get at the info.
Rebooting the entire network seems to do the trick for a while, but after about 24 hrs it is noticeably worse. The old 50Mb modem I used to reboot about twice a year - every 24h seems a tad less reliable.
I wondered if someone could check these figures look OK... before I start delving into wireshark, which is really beyond my current skill-set ;-)
Locked 56 291000000 Hz QAM256 -1.7 dBmV 39.8 dB 212 0
Locked 57 299000000 Hz QAM256 -1.7 dBmV 40.1 dB 111 0
Locked 58 307000000 Hz QAM256 -1.7 dBmV 39.9 dB 131 0
Locked 59 315000000 Hz QAM256 -2.0 dBmV 39.2 dB 130 0
Locked 60 323000000 Hz QAM256 -1.7 dBmV 39.1 dB 59 0
Locked 50 35800000 Hz ATDMA 35.8 dBmV 16QAM 6400000 20480 Kbits/sec
(darnit - I had that in a nice table, but it didn't seem to like it ;-)
12-11-2012 16:23 - edited 12-11-2012 16:23
I am not able to test your hub as it is currently offline.
Could you please leave it powered on and let us know so we can investigate this for you.
Also, it would help if you could run a traceroute and post the output. This will help to identify any packet loss or high ping times.
As you have this problem when using the hub in modem mode, this does discount any firmware related issues but it might be worth you trying with your pc in safe-mode-with-networking. I am not saying this is a pc fault, but it would be interesting to see the difference (if any) in results.
(Btw, the power levels you posted look fine.)
on 12-11-2012 17:47
Thanks for getting back to me & glad the power levels look OK.
That's odd, though, that you can't see it - the router is always on, unless It's rebooting, which I haven't done in the past 24h. [actually, have now done since starting this reply]
It is in modem mode & the gateway behind it is stealthed, if that would make a difference.
I have since discovered that Mac OS X 10.8 & Safari are having reported problems with slowdowns - but when I'm seeing the issue, it's not only my machine, but any on the network, the rest are Win XP & Win 7, all up to date, no dodgy 'speed boosters' etc. Rebooting this machine does improve it, but not as much as rebooting the entire network, router, gateway, switch, all machines.
Other than the update to 10.8 in the past month, there are essentially no other changes that I am aware of, other than standard Win updates etc.
Running in safe + networking is hard to do without cutting the rest of the machines off for an extended period. I did try with just the Mac & the modem back in full router mode to get the power figures, but I'd be pretty unpopular if I did that for any length of time. The issue never appears when everything is freshly booted.
I took some readings when it got really bad the other day, at that point it was taking over 4 mins just to refresh the forum page. Unfortunately, I didn't keep them, & right now it's pretty much behaving itself, except for that 1st hop outside the building which looks a bit long - I have to admit to not knowing enough about it to figure why that hop has a longer response time than one beyond it.
Pinging that on its own looks jittery, though
PING 220.127.116.11 (18.104.22.168): 56 data bytes
64 bytes from 22.214.171.124: icmp_seq=0 ttl=63 time=7.934 ms
64 bytes from 126.96.36.199: icmp_seq=1 ttl=63 time=9.404 ms
64 bytes from 188.8.131.52: icmp_seq=2 ttl=63 time=17.485 ms
64 bytes from 184.108.40.206: icmp_seq=3 ttl=63 time=46.996 ms
64 bytes from 220.127.116.11: icmp_seq=4 ttl=63 time=9.680 ms
traceroute to virgin.lithium.com (18.104.22.168), 64 hops max, 72 byte packets
1 name removed (192.x.x.x) 0.808 ms 0.335 ms 0.197 ms
2 cpc5-enfi17-2-0-gw.hari.cable.virginmedia.com (22.214.171.124) 12.074 ms 17.890 ms 19.477 ms
3 hari-core-2b-ae7-725.network.virginmedia.net (126.96.36.199) 9.918 ms 10.606 ms 15.921 ms
4 popl-bb-1b-ae8-0.network.virginmedia.net (188.8.131.52) 12.022 ms 7.330 ms 18.025 ms
5 popl-bb-1a-ae0-0.network.virginmedia.net (184.108.40.206) 7.759 ms 7.530 ms 16.254 ms
6 popl-tmr-1-ae4-0.network.virginmedia.net (220.127.116.11) 61.485 ms 9.977 ms 24.510 ms
7 amst-ic-1-as0-0.network.virginmedia.net (18.104.22.168) 15.724 ms 18.886 ms 24.477 ms
8 te0-7-0-10.ccr21.ams04.atlas.cogentco.com (22.214.171.124) 24.253 ms 15.375 ms 33.430 ms
9 te0-2-0-3.mpd22.ams03.atlas.cogentco.com (126.96.36.199) 24.693 ms
te0-0-0-7.mpd21.ams03.atlas.cogentco.com (188.8.131.52) 17.643 ms
te0-0-0-7.ccr22.ams03.atlas.cogentco.com (184.108.40.206) 27.013 ms
10 te1-1.ccr01.ams07.atlas.cogentco.com (220.127.116.11) 20.809 ms
18.104.22.168 (22.214.171.124) 17.684 ms
126.96.36.199 (188.8.131.52) 16.614 ms
11 184.108.40.206 (220.127.116.11) 22.202 ms 21.638 ms 26.411 ms
12 community.virginmedia.com (18.104.22.168) 19.199 ms 19.211 ms 16.466 ms
Let me know if any other info could help.
on 15-11-2012 17:25