Unstable Server Singapore/Japan for Path of Exile 1

Came back to use mudfish for Path of Exile after not using it for a while and connection to Singapore and Japan servers are very unstable during the evenings around 5pm - 11pm. This issue never happened before since I was always using mudfish to avoid this specific bad routing issue, ping spiking 150-250 for several hours making it near unplayable even on predictive mode.

Another issue is with constant disconnects on certain mudfish nodes so I would have to constantly go back and forth switching nodes to find one that won’t kick me out or is incredibly unstable, its very tedious since I always have to reenter credentials and get the code via email to login.

Japan seems to only be the stable server on mudfish for me too, I used to get 39-50ms on Singapore servers when using mudfish, but now I’m getting over 80-100 in Singapore and Japan is my only option giving me around 60ms which still feels sluggish. This issue only came up this year.

On top of that after topping up my balance, around 10-15 GB of data was used really fast for the first few days while only playing Path of Exile, but now for some reason when I leave my PC on overnight with Path of Exile on, its only using ~300mb, where did the 10-15 GB from the first few days go if I was only playing Path of Exile? I had no other updates or background apps running besides Discord and the trade website.

First sorry for this inconvenience. Unfortunately, I think your issue is related to Amazon EC2, Twitch and Mudfish and CDN and Mudfish links. Please read these documents.

However, I’d reset your mudfish credits a little bit.

Could you please confirm that it’s also same with Full VPN mode enabled?

If it works fine with Full VPN mode enabled, it means there’s something a problem in the game item of Mudfish.


image
image

These are the results when testing with Full VPN Mode, still the same result as without using full vpn, around 60-80ms with spikes up to 100 in game

I also want to add that I also get high ping in certain times in the day even when using mudfish on both Japan and Singapore nodes.


Umm… When I checked this issue, this issue might be related to your ISP or international routing. It seems it’s changing a lot up to the time.

Please try to test your network status from your desktop to 172.104.63.216 (the public IP of SG Asia (Singapore - Linode 2) mudfish node) using How to use WinMTR link. I guess that this result shows some insights about where this issue is from.

image
Here’s the WinMTR results with mudfish turned off

Okay. As you can see, here are my observations:

  • At no 2, its IP is 100.77.0.1 (a private network) for your ISP. So this router is a closest router from your home router. But as you can see RTT avg is 14 ms though the worst is 98 ms.
  • At no 3, I’m not sure that pldt.net is still your ISP. But it’s a second closest router. As you can see, already at this point, your RTT avg is over 101 ms and worst is 169 ms.

An important thing is wheterh you can avoid or bypass this pldt.net 122.2.206.106… If you can’t bypass, I don’t think we can solve this issue.

To bypass this routing point, you might need to change a mudfish node to another

Thanks for looking into this.

Unfortunately pldt.net is my current ISP and they seem to be doing some poor routing still. Is there a way to bypass this or is the best option to swap ISP? Trying a bunch of mudfish nodes and still getting very mixed results.

As you might know, we need to avoid this 122.2.206.106 IP block because most of the latency comes from this area.

So one thing I can suggest is:

  • From https://mudfish.net/api/staticnodes link, you can find all public IPs of mudfish nodes.
  • Pick a public IP of mudfish nodes which locates at near your physical location.
  • Perform WinMTR test against this public IP and monitor the routing path.

If there’s a mudfish node which does’t go through a bad routing path, I think you need to use it if possible.