Zone instance lag spikes

For the past few weeks/monthish (started Atlas expansion later in league), I have been getting really laggy zone instances. I play on TX server.

For instance, I can do a dried lake run with my latency flatlined at 20-30ms, no problems. Go to town, create a new instance of dried lake and my latency starts to jump all over the place, usually hovering around 200ms with spikes up to 500ms.

For instance, here is my latency going from a stable act4 town into a dried lake:

- first spike is me entering town from my hideout
- latency jump and spikes are me walking around in a laggy dried lake



Here is me exiting that dried lake instance (beginning spikes), going into town and creating a stable dried lake (~20-30ms) and walking around



There seems to be no rhyme or reason for certain instances lagging. Not sure what I can do to diagnose the issue.

Here is my MTR report

Yeah this happens for me in texas. I basically recreate instance till one is low ping. Only sucks in lab where you cant do that.

"

Southernhosp wrote:

Yeah this happens for me in texas. I basically recreate instance till one is low ping. Only sucks in lab where you cant do that.

Maps are much worse. Fun ripping a T14 map because the zone is getting 500ms latency spikes.

I am having the same thing. It is making the game unfun. When Mapping some areas are lag free (I stay at 50ms). Some Areas I will just spike around from 50ms to 2000ms. 2 second lag spikes are a fun killer. Not sure what to do. My Hideout is always lag free. Some maps are always lagging. I tested tier 2 maps. Factory always lags. The the other 3 seem to be ok.

"

pantsparty1002 wrote:

For the past few weeks/monthish (started Atlas expansion later in league), I have been getting really laggy zone instances. I play on TX server.

For instance, I can do a dried lake run with my latency flatlined at 20-30ms, no problems. Go to town, create a new instance of dried lake and my latency starts to jump all over the place, usually hovering around 200ms with spikes up to 500ms.

For instance, here is my latency going from a stable act4 town into a dried lake:

- first spike is me entering town from my hideout
- latency jump and spikes are me walking around in a laggy dried lake



Here is me exiting that dried lake instance (beginning spikes), going into town and creating a stable dried lake (~20-30ms) and walking around



There seems to be no rhyme or reason for certain instances lagging. Not sure what I can do to diagnose the issue.

Here is my MTR report

lag spike of over 100ms at the router, try a power cycle and then run the test again

Anyone else chipping in here might want to provide their own WinMTR

|------------------------------------------------------------------------------------------|
| WinMTR statistics |
| Host - % | Sent | Recv | Best | Avrg | Wrst | Last |
|------------------------------------------------|------|------|------|------|------|------|
| 192.168.0.1 - 0 | 126 | 126 | 0 | 0 | 10 | 0 |
| phnx-dsl-gw69.phnx.qwest.net - 0 | 126 | 126 | 20 | 32 | 177 | 21 |
| phnx-agw1.inet.qwest.net - 0 | 126 | 126 | 20 | 20 | 33 | 20 |
| los-brdr-01.inet.qwest.net - 2 | 119 | 117 | 0 | 32 | 154 | 30 |
| 63.146.27.34 - 0 | 126 | 126 | 30 | 30 | 59 | 30 |
| ae-4-90.edge1.LosAngeles6.Level3.net - 2 | 119 | 117 | 37 | 38 | 223 | 37 |
| te2-5.bbr01.cs01.lax01.networklayer.com - 0 | 126 | 126 | 30 | 31 | 54 | 30 |
| ae6.cbs01.cs01.lax01.networklayer.com - 26 | 63 | 47 | 0 | 33 | 53 | 33 |
| ae0.cbs01.eq01.dal03.networklayer.com - 57 | 39 | 17 | 0 | 54 | 55 | 54 |
| ae33.bbr01.eq01.dal03.networklayer.com - 0 | 126 | 126 | 51 | 52 | 68 | 52 |
| ae0.dar01.sr01.dal01.networklayer.com - 0 | 126 | 126 | 51 | 52 | 68 | 52 |
| po1.fcr05.sr06.dal01.networklayer.com - 0 | 126 | 126 | 52 | 52 | 65 | 52 |
| patch101.pathofexile.com - 0 | 126 | 126 | 51 | 52 | 56 | 52 |
|________________________________________________|______|______|______|______|______|______|
WinMTR v0.92 GPL V2 by Appnor MSP - Fully Managed Hosting & Cloud Provider

This was while I was lagging bad in a bog map. Average 52 ms yet i get 500ms to 2000ms spikes ever 1 to 2 seconds.

"

lagwin1980 wrote:

lag spike of over 100ms at the router, try a power cycle and then run the test again

Anyone else chipping in here might want to provide their own WinMTR

Not sure how to read these. Here is the MTR after a fresh router reset. Even with a flaky router, still doesnt make sense to me why certain instances lag and others are completely fine.



Also, thanks for your help.

Has anyone found a solution to this issue. It just started happening for me in random instances. There is definitely no problem with my isp. And it only occurs in about 30% of instances. Recreating the instance almost always immediately corrects it. But this is a huge annoyance, and will likely be very annoying to deal with once I start doing maps.

Same problem to me.

"

pantsparty1002 wrote:

"

lagwin1980 wrote:

lag spike of over 100ms at the router, try a power cycle and then run the test again

Anyone else chipping in here might want to provide their own WinMTR

Not sure how to read these. Here is the MTR after a fresh router reset. Even with a flaky router, still doesnt make sense to me why certain instances lag and others are completely fine.



Also, thanks for your help.

If you want help, make your own thread please. This is Pant's thread to get help in. You all are making it more difficult to help him. Read the stickies and post the information you are requested to. Pay special note of the first paragraph of the first sticky in this forum.

@Pants:

I'll teach you a little bit on how to read it. :)

Loss % = to the amount of packets you are losing. Obviously, zero is a good thing.

Sent/received: <--- Amount of packets. Generally, you want a sample size from 300-500. 500 is better as it gives you a much better data set.

Best/Avg/Worst lines: Best is is fastest that you sent received the packets. Avg is your overall average latency, worst is the longest that it took. The lower the numbers the better.

Last: Simply refers to what time it took the last packet to travel.

So what we are looking at with you here is that third hop. You have a worst of 900+ and that also kills your average, as it raises that to almost 200. Your data set is a little small though, so it may/may not be as terrible as it looks. Also looks like a couple of the networklayer hops were a bit congested as well, which was giving you some higher latency as well. Can you do a trace for 500 seconds for us so we can get a better look see.

Next thing then is your client.txt file. When you get into an instance that is laggy as all hell, open up your client.txt file. It's located in your POE install dir/logs folder. Look for the instance IP of the zone that you are in (should be the last group of 4 lines or so that will have it, as long as you do not portal out/change zones). Take note of that IP addy and run a trace to that specific IP.