PingPlotter to deadmen

[delanvital]

WW got lagged. Reiken and I had 30 second lags. Others had on occasion some as well. I only know about boomer. Benny suggested PP, so here it is:

AFAIK pacet loss in-game can cause these long hick-ups, where you will either end up with "there is a connection problem" and get dropped or game will continue after a long lag.

Amazing how consistent those packet losses are? It is like clockwork.



What I really like is my improved ping with BS.A. though. Some evenings I can get consistent pings of 16ms!

Edit: I can't seem to find the logic in it, but I think it is a problem on my side. I can't recreate this, without getting packet loss all over. Don't know why it was only the dMw jumps for 6 minutes earlier. I guess I need to work more on this, so feel free to delete post. Unless ofc you think I am on to something but I doubt it... Unless 1 sec trace intervals causes more packet loss, due to too quick repetition or something?

[Anonymous]

Try using 5 sec intervals.

Additionally try in the morning and again in the evening so you can see if there is any "time based" loading on certain routers.

[OldBloke]

Might be a coincidence but we are seeing non-UK based racers suffering sporadic discons from our LFS server. Might be a problem on a link into the UK.

[KKND]

Might be a coincidence but we are seeing non-UK based racers suffering sporadic discons from our LFS server. Might be a problem on a link into the UK.

Always the UK :devil:

[Benny]

knock the interval down as BB says, I see the same last three hops from here with zero loss.

When heading your way however I see loss on the hop as it hits the transit to  .dk, interestingly I see loss thorugh BT aswell, but that's not related here.

Looks like running across the sea ain't to pretty.

Run it to another UK server or host....say 194.72.20.249, and post what you get.

[delanvital]

First of all, it seems I get packet loss all over in periods, from the wireless connection I am on. Must be the gear here. However, there is one Tiscali stop here that is consistently bad, and that is the one prior to hitting DK servers. I will try the stuff you suggest Benny, from a cable connection, and post again.

Any other Danes on TDC here who can do the same for comparison?

[M_a_r_k_o]

Hi All

I know this thread is a bit old, but, I have only just noticed it.

A note about ping and ICMP packets through our network.

ICMP Packets are not Express Packet Switched, they are in fact Fast Switched.

This means that the packets are handled in software as opposed to hardware within our routers. And this means that they are treated with the lowest priority.

Packet loss of ICMP packets can be artifically induced by simply having too many people running ping plotter at the same time or having the ping rate too high as seen in the plot at the top of this post.

Regards

Mark

[delanvital]

Hi All

I know this thread is a bit old, but, I have only just noticed it.

A note about ping and ICMP packets through our network.

ICMP Packets are not Express Packet Switched, they are in fact Fast Switched.

This means that the packets are handled in software as opposed to hardware within our routers. And this means that they are treated with the lowest priority.

Packet loss of ICMP packets can be artifically induced by simply having too many people running ping plotter at the same time or having the ping rate too high as seen in the plot at the top of this post.

Regards

Mark

Hi Mark. Nice to hear you say that, because I noticed that but had no explanation for it.

In this case I got packet loss all over the place, even LAN wise, and after changing an old wireless adapter it was fixed.

Cheers for the explanation!

[OldBloke]

Many thanks for taking the time to post, Mark. Much appreciated :thumb:

[Dr Sadako]

cs.deadmen.co.uk.pp2
 
Pings still terrible. :sad:

[kregoron]

those telehouse nodes are spiking alot for me too.. no idea why..

[OldBloke]

We had quite a few discos on LFS tonight - me included. Settled down just after 9 pm.

[Whitey]

tracert cs.deadmen.co.uk

Tracing route to cs.deadmen.co.uk [77.108.129.46]
over a maximum of 30 hops:

  1     1 ms    <1 ms    <1 ms  www.routerlogin.com [192.168.0.1]
  2    33 ms    34 ms    34 ms  217.32.0.186
  3    34 ms    33 ms    32 ms  217.32.0.161
  4    32 ms    33 ms    32 ms  217.41.174.21
  5    34 ms    32 ms    45 ms  217.41.218.74
  6    35 ms    33 ms    33 ms  217.41.218.54
  7    33 ms    33 ms    32 ms  217.47.186.241
  8    34 ms    33 ms    35 ms  core2-pos12-1.edinburgh.ukcore.bt.net [217.32.16
9.213]
  9    41 ms    42 ms    42 ms  core2-pos6-1.birmingham.ukcore.bt.net [62.6.201.
29]
 10    44 ms    44 ms    43 ms  core2-pos0-1-4-0.ealing.ukcore.bt.net [62.6.201.
85]
 11    44 ms    45 ms    45 ms  core2-pos1-0-0.telehouse.ukcore.bt.net [62.6.201
.86]
 12    43 ms    44 ms    43 ms  194.74.65.58
 13    45 ms    44 ms    44 ms  bgp.brightstar.ltd.uk.enta.net [87.127.231.114]

 14    45 ms    45 ms    44 ms  router.brightstar.ltd.uk [77.108.128.2]
 15    47 ms    49 ms     *     ba3nic1.deadmen.co.uk [77.108.129.46]
 16    47 ms    46 ms    46 ms  ba3nic1.deadmen.co.uk [77.108.129.46]

Trace complete.

It's been really bad so far tonight.  I've tried a restart of the CS Servers but that has had no affect.  It's making it virtually unplayable.

[kregoron]

Yeah was really bad last night, didnt even bother doing a plotter.. but happened several times everyone getting 130+ pings..

[Whitey]

I've got pingplotter installed now, the traceroute above doesn't really show much. The problem was that although the pings looked OK during the spell on MH, there were lots of lag spikes, so people were stuttering around.