Unusual use for a BBQM Graph - VM Superhub

I have been using my BQQM graph to help me identify a problem with my VM Superhub.
I noticed that I was getting network issues and that the graph was looking messy with higher than normal latency Further investigation showed the pings were increasing over time.

Ramping ping

I have isolated the cause to be the SH itself rather than the connection and was wondering if any other users have experienced similar graphs.

Tim

Edited by deleted (Mon 25-Mar-13 13:59:20)

http://www.thinkbroadband.com/ping/share/3a7879176d7...

The link was missing the l at the end.

Can't see I've seen that 24 hour based ramp before.

The weird thing is that I only get it with the SH wireless turned on.

I also have evidence to show it's not caused by the VM network in general as if I turn the Wireless off I get a flat graph. Also pinging the SH local address 192.168.0.1 shows rising pings as well.

Finally I have an IPv6 Graph running as well which is terminating on my desktop PC via a tunnelbroker.net tunnel. Although the graph shows some of the same major features the base latency is stable throughout.

Ravenstar68

here is good example of bad incorrect high ping from an superhub (i rebooted the hub about 3:30am ago you can see the ping is now normal)

the red bit is when it was powered off and back on at 3:30am not sure about the other longer red one(seems BQM/VM global related as f8lure was affected as well) but it is the router that is messing the pings up not the connection seems the router starts to slow down after some time

http://www.thinkbroadband.com/ping/share/69e4ac6c9fc...

you can look at the history on the one below, press "View Historic Graphs" on the firebrick one) the high packet loss was due to high utilization and single upstream channel, once dual upstream was enabled ping went to normal(no packet loss at peak time any more), the last one is when i restart the router to get the pings working correctly
http://f8lure.mouselike.org/index.asp?DoAction=ViewH...

i guessing if i put it into modem mode an used the asus WL-520GC i have, it will probably not do this odd ping (does make it annoying to monitor superhub routers but as long as there is no packet loss as it seems ok but it could be messing up your avg for virgin media customers when you avg it out)

Edited by leexgx (Thu 28-Mar-13 04:52:47)

Seems to be an issue

http://community.virginmedia.com/t5/Wireless-Network...

but not all super hubs do it

mine below before and after

http://www.thinkbroadband.com/ping/share/36dccd5e5c5...
http://www.thinkbroadband.com/ping/share/681b6cb4b17...
(can't seem to get day 28 to work)

before and after reboot (these will expire after 30 days)
http://f8lure.mouselike.org/archived_graphs/86.27.17...
http://f8lure.mouselike.org/archived_graphs/86.27.17...
below should allways work as long as the IP has not changed
or http://f8lure.mouselike.org/index.asp?DoAction=ViewH...

just mite be usfull to add it to the FAQ if they see this type of ping on BQM with an Vm superhub or some other routers, it is the router that glitches on the time it takes to respond to pings so you get high pings when they are not (rebooting the router corrects the incorrect BQM pings)
good example is below

this tends to happen other an week of the super hub been on (if wifi is off ping tends to stay normal)

http://www.thinkbroadband.com/ping/share/52134718ee8...

this was after an reboot/crash, long red bit router had stopped until the my friend got back in and noticed Internet was dead, but has nothing to do with Ramping pings that was going on before the reboot happened the pings are very normal after that

as he has Dual upstream channels now there is no longer any packet loss at peak time any more, look at the f8lure on my last post to see the affect it had when dual channel was enabled, no packet loss any more at the, bottom is the newer ones, top is oldest
http://www.thinkbroadband.com/ping/share/3df529b9e04...

Edited by leexgx (Mon 08-Apr-13 02:13:03)