Granblue Fantasy - Russian Nodes

[Cubix]

Hello.

It has been a reoccuring issue, that some Russian nodes work well (lowering my ping from 280~ down to 160~) for a few days, only for them to start performing poorly (lowering the ping only to 200-210~ at best, mostly worse/not even worth using) for several months.

Specifically, during the time period of 16.03.2024 - 24.03.2024 the following nodes were working great:
RU Europe (Novosibirsk - JustHosting)
RU Europe (Novosibirsk - Adelina)
RU Europe (Novosibirsk - Adelina 2)
Only for them to start performing poorly from the 25th onward again. Before this they had been performing poorly for several months as well.

I understand that there have been several topics on this issue before and that there were probably no changes done on Mudfish’s side, however, would it be possible for you to contact/notify the server providers for these nodes specifically to take a look at the network data in that specific time period and compare it to before/after to perhaps be able to change the configuration to what it was like during the time they were working great? I’d appreciate it!

Thank you very much and have a nice day.

[weongyo]

@Cubix

I understand that there have been several topics on this issue before and that there were probably no changes done on Mudfish’s side, however, would it be possible for you to contact/notify the server providers for these nodes specifically to take a look at the network data in that specific time period and compare it to before/after to perhaps be able to change the configuration to what it was like during the time they were working great?

Yep… At least no changes in Mudfish side. Let me contact to the server provider regarding to this issue.

[Yari]

Can confirm it is a really bad experience for all of my friends using the above mentioned servers. Really appreciate the server provider being contacted, would be really happy to see this ongoing issue resolved. There’s just so much variance in the quality of these servers.

[Yari]

Hi, I was wondering if there is any update after contact with the server provider? Is this something they can look into?

[weongyo]

Unfortunately they didn’t do anything. They said that there are no issues on their end. I’ll try to follow up again.

[Cubix]

To provide some further information on this issue, in case it is of any help:

The nodes had been performing poorly over the weeks since the creation of this post (and before as aforementioned).
Over the last couple days however (like 3-4 days) the nodes were working really well again.
Up until roughly this morning, Monday 22.04.2024 5am CEST~
This is also reflected in the snapshot of the RTT Graph below:

186.34 was the “good” value of the nodes performing well over the last few days (leading to ~162ms ping using Advanced mode, 180-190ms using the Basic Mode with these nodes as a single node) and then there is the spike to 326.46…~ value from this morning, making the nodes once again perform poorly ever since (leading to a ~192ms ping at best using Advanced mode, all the way up to 220-320ms+ at worst, especially when using the Basic Mode with these nodes as a single node)

What’s perhaps useful information:
The best ping (for me personally) of 162ms~ is achieved when the nodes are working well and then using Advanced mode and routing something like:

So my location (Germany) → Germany node → Russian Node → JP Game server

However the best ping of 192ms~ when the nodes are performing poorly is achieved by routing like this:

So my location (Germany) → Russian Node → JP Node → JP Game server
Whereas the before mentioned Advanced mode routing (using Germany Node) would lead to very bad ping 300ms+ (just like using the Russian Node as standalone Basic).

So it would seem that something in the way the routing is handled keeps changing and thus making these nodes very unreliable and have so much variance?
I’d think based on the RTT Graph of the Russian Nodes itself (and especially considering that these Russian nodes go from working “well” as standalone nodes using Basic mode to bascially unusable, performing very poorly as standalone nodes using Basic mode) that they would be at fault but if the server providers have not done any changes I don’t know where this issue is coming from.

I have called and talked to my ISP about this as well and they have also confirmed that they have not done any changes when it comes to the routing and given that other people also have the issue I doubt it’s on our side.

Appreciate your help!

[weongyo]

@Cubix

So it would seem that something in the way the routing is handled keeps changing and thus making these nodes very unreliable and have so much variance?

Yes in my opinion. Normally changing the routing path is to save the traffic cost in the upstream ISPs. In a view of mudfish, it’s at somewhere out of our control.

@Cubix

I have called and talked to my ISP about this as well and they have also confirmed that they have not done any changes when it comes to the routing and given that other people also have the issue I doubt it’s on our side.

If you’re familiar with WinMTR, please try to test your network status from your desktop to 203.104.248.93 (Granblue Fantasy - Asia (Japan)) using How to use WinMTR link. It could be helpful to identify where this issue is from.

[Cubix]

Thank you for your response.
I have left a rather lengthy response in your private messages in case it contains any sensitive (private) IP information.
I’d appreciate it if you could check it out when you have the time to do so!
Feel free to either leave your response to it here in this post or in my private messages.
In any case I will update this post with further information accordingly, for all the others that are interested in this topic.