New on LowEndTalk? Please Register and read our Community Rules.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
All new Registrations are manually reviewed and approved, so a short delay after registration may occur before your account becomes active.
PCCW Issues
Hi all,
Anyone seeing issues with PCCWBTN now? I am getting 150ms from my house in London to their routers in London. Virgin (my ISP) uses PCCW to reach Psychz in LA. Normally I get 140ms and 4 bar RDP. Right now I have 300ms ping, and 1 bar RDP.
Anyone else experiencing similar issues? Their internal looking glass seems to point to issues:
Comments
Log a case with Psychz and get them to escalate to PCCW. That will be faster.
Also, what program you use for RDP which has the Bar?
I don't have any issues reaching PCCW London from neither of our upstreams, plus my home connection from Zen. This could potentially be Virgin Media related issue or something specific to their link.
Seems to be recovering. Issue seemed to be as soon as you actually tried to go anywhere over the network
That's over Aorta or some direct link between PCCW and VM?
Direct link:
7 14 ms * * tclo-ic-3-ae0-0.network.virginmedia.net [212.250.25.62] 8 * 114 ms 114 ms te7-5.br03.ldn01.pccwbtn.net [195.66.224.167] 9 262 ms 263 ms 260 ms TenGE0-2-0-7.br03.lax05.pccwbtn.net [63.218.72.178] 10 * 259 ms 263 ms ae11.PCCW.lax.us.AS40676.net [23.238.223.9]Interesting, that IP comes out to AS47872 Sofia Connect EOOD, not AS3491 Beyond The Network America, Inc. like the next hop does....
Potentially might be some sort of legacy peering relationship they have with VM, Aorta doesn't route to PCCW directly as per what I see. Either way, it's the first hop of PCCW that already has high latency, which could mean the issue is on either Virgin or PCCW end, or a carrier in between. Best bet would be to report to Virgin NOC and PCCW and see how it goes.
isn't LAX in LA?
The latency in the screenshot looks fine. It shows 140ms at the first hop but that doesn't follow through to subsequent hops. 144ms from London to Los Angeles is fine.
What you want to do is check latency from PCCW London using their LG to your home.
Looking at your second trace from Virgin the issue looks to exist between Virgin and PCCW, potentially a saturated link.
Here is a MTR now: http://pastebin.com/raw/hAzBayQF if it helps but you probably see it's already fine again on your end too although it looks like it's routing via Amsterdam now for me at least.
As per your mtr it seems the outgoing route is from London to Amsterdam PCCW instead of handing over to PCCW in London as in the OP example. Quite possibly they re-routed around the issue (or your area routes that way - not enough data).
It seems to all have gone back to normal, via London. Latencies and RDP are behaving. I also have an rsync going on from Iniz to NorthHosts, and thats using PCCW and hitting 100Mbps.
Glad it's back to normal :-). I find it weird though that they route some via London and some via Amsterdam, but maybe that depends on the region and is a form of load balancing.
I noticed something. When I first started connecting to Psychz LA for hours at a time, it would route over Amsterdam. After a while it switched over to London. It seems once you really use it, they shunt you over to London.
Source based routing is highly unlikely, though they might have some load balancing system in place though where a particular region can be routed differently as needed.