Large lag spikes (up to and including 4000ms)

Since the 3.11.1 patch, I have begun to experience large lag spikes. This wasn't an issue before the patch.

I think it is somehow a memory leak issue or some other client side bug. Every time I experience this, I leave the map and restart the client. Upon returning, it is fixed. I've left the map via portal and /exit command - both work. I've hit this bug in delve, maps, and the map-groves so far. I'm still not sure what triggers it.

My screenshots show "lag" spikes hitting 4215ms at worst, but most hovering in the 600-1200ms range. It rarely even attempts to catchup, it just drops the frames and skips ahead (probably by design). I've stayed in the maps when the lag spikes begin and it never gets better. Restarting the client does the trick immediately.

Example image:
https://photos.app.goo.gl/b7gQfknDL85QknSz7

Reference #3,257,160,956
Last bumped on Jul 5, 2020, 11:09:28 PM
Read the stickies in this part of the forum and post a WinMTR so we can see what your trace looks like.
"Programming today is a race between software engineers striving to build bigger and better idiot-proof programs, and the Universe trying to produce bigger and better idiots. So far, the Universe is winning."

"Your mother was a hamster and your father smelt of elderberries!"
Hi,

Thank you for the attention to this bug!

It appears that this was moved to the Technical Support forum. I understand why the title sounds like it belongs in the tech support forum, but in my opinion this is a recently introduced bug, not a technical problem. This has the appearance of a lockstep lag bug. My network, RAM, and processor have all been operating in nominal ranges throughout this issue.

In an attempt to help track down the cause, here is the requested info for this new forum:

https://pastebin.com/pZ1Mi2Ri


Thank you!
The next time I encounter this problem, I will try to run WinMTR as well. The above one can be treated as normal behavior to compare against. Until then, the /bug reference information might be the most relevant data I can provide.
Well, there are only 2 hops in that first trace you posted. There has got to be more than that, yes?

Even me, on my really really short path only has about 5/6...so unless you are in the data center itself, there has got to be a few more. :)

Also, if you are on wireless, try changing to wired if you can. Also, what gateway? If DC, then GGG is having hosting provider issues atm and they are working on correcting them.
"Programming today is a race between software engineers striving to build bigger and better idiot-proof programs, and the Universe trying to produce bigger and better idiots. So far, the Universe is winning."

"Your mother was a hamster and your father smelt of elderberries!"
My friend and I are having the same issue. A lot of the time things are great, but will get instances where my ping spikes massively and I have to portal out and back in to fix it (usually). It was definitely not the case before 3.11.1.
La7me
"
La7me wrote:
My friend and I are having the same issue. A lot of the time things are great, but will get instances where my ping spikes massively and I have to portal out and back in to fix it (usually). It was definitely not the case before 3.11.1.


Make your own thread with the required info like the stickies tell you. The stickies also tell you NOT to hijack someone else's thread. This is Aeb's thread to get help in.
"Programming today is a race between software engineers striving to build bigger and better idiot-proof programs, and the Universe trying to produce bigger and better idiots. So far, the Universe is winning."

"Your mother was a hamster and your father smelt of elderberries!"
"
Well, there are only 2 hops in that first trace you posted. There has got to be more than that, yes?

Even me, on my really really short path only has about 5/6...so unless you are in the data center itself, there has got to be a few more. :)

Also, if you are on wireless, try changing to wired if you can. Also, what gateway? If DC, then GGG is having hosting provider issues atm and they are working on correcting them.


Yeah, you and I both found that odd. Even if I select something like google.com, it only shows one hop. My options haven't been modified from defaults. It says 1 second interval, 64 byte ping size, 128 max hosts in LRU list, resolving names. I couldn't find anything beyond that to modify.

Edit: Washington, DC gateway on wired
Last edited by AebLincoln on Jul 5, 2020, 9:36:30 PM
I wonder if your ISP is somehow obfuscating the rest of the traffic....

You don't run a VPN do you?

I am going off to bed shortly..if I don't catch this again tonight, I'll take a gander at it tomorrow and do some digging in the meantime to try and figure out why we only see 2 hops.
"Programming today is a race between software engineers striving to build bigger and better idiot-proof programs, and the Universe trying to produce bigger and better idiots. So far, the Universe is winning."

"Your mother was a hamster and your father smelt of elderberries!"
No VPN.

WinMTR during most recent lag: https://pastebin.com/uVkXyfzc

New log: https://pastebin.com/hUQn9zj9

There was a disconnection that occurred on this one.

Bug reference #1,948,209,835 never got printed to the screen. I believe it may have been a timeout error.

Report Forum Post

Report Account:

Report Type

Additional Info