Engineers are you aware of this?
We need more info on this.
Did you try joining a different server?
Does the player eventually get disconnected? If so, after how long?
Does it affect all players in the server, even the players already playing?
Do you see any activities from the network debug panel? (Shift+F3 on PC)
[quote] We need more info on this.
Did you try joining a different server?
Does the player eventually get disconnected? If so, after how long?
Does it affect all players in the server, even the players already playing?
Do you see any activities from the network debug panel? (Shift+F3 on PC) [/quote]Read the thread and the linked threads, they provide most of this information. Mine, Owenās and TwentyTwoPilots responses.
[quote] We need more info on this.
Did you try joining a different server?
Does the player eventually get disconnected? If so, after how long?
Does it affect all players in the server, even the players already playing?
Do you see any activities from the network debug panel? (Shift+F3 on PC) [/quote]Read the thread and the linked threads, they provide most of this information. Mine, Owenās and TwentyTwoPilots responses.[/quote]
I read this and your thread, and I just read them again. I did not find any answer to my specific questions.
My questions are directed at the server no response after UI loads issue, your issue in the other thread seem to be script related hang as others have pointed out. It would speed up the investigation/fix process if all information is organized at one place.
[quote] My questions are directed at the server no response after UI loads issue, your issue in the other thread seem to be script related hang as others have pointed out. It would speed up the investigation/fix process if all information is organized at one place. [/quote]Itās worth noting that this script does not loop on the point at which it errored and instead, said line fired a remote event for all clients. This is why I assume itās related, since I am assuming the timeout was caused by the server not receiving a confirmation response from any clients and vise-versa. Hopefully thatās more concise.
Iāve witnessed errors at HEX long after the UI has loaded where, as mentioned above, the server and all clients appear to have an intermittent connection before the connection eventually drops and might regain a minute or so later. Unfortunately, I canāt provide any logs for HEX, but I can say there were no client errors for my game, and it was the same for HEX when I checked.
This happens in multiple game servers, but is intermittent as fore mentioned. Iāve seen a few players leave but never disconnect, this is only usually after the server resumes. It affects all players.
[quote] We need more info on this.
Did you try joining a different server?
Does the player eventually get disconnected? If so, after how long?
Does it affect all players in the server, even the players already playing?
Do you see any activities from the network debug panel? (Shift+F3 on PC) [/quote]
Itās hard to answer more specificity as its not a problem I can replicate.
- Some servers work, some servers donāt.
- Cannot say for sure. Most players leave because the game doesnāt work. If I encounter again I will test
- From what Iāve seen it affects all players
Got >30 tweets/PMs about this issue last night.
Someone sent me this:
The ā1ā is when the client calls a remote function. The server is not responding.
People who send screenshots that small should be burned alive.
Weāre trying to figure out what this can be. We do see a decreased average server FPS last night in our metrics, but so far we only have one lead thatās not very promising 
You can repro it on literally an empty place.
That game is exactly as Studio defaults to when you create a new place. It also takes longer to open and publish than it should do. If needed, I can make it public domain to prove that.
EDIT: Actually, that was a lie. It has three LocalScripts in it, running this code:
print("Script in", script.Parent:GetFullName(), "ran!")EDIT 2: It takes really long to join that place and the script in ReplicatedFirst ran at exactly the same time as the ones in the Backpack and PlayerGui did:
EDIT 3: Okay, uncopylocked.
Thatās because you should use
script.Parent:RemoveDefaultLoadingScreen()
[quote] Thatās because you should use
script.Parent:RemoveDefaultLoadingScreen() [/quote]
Even without that, I would think that the code in the replicated first should be printed far in advance before the two in the backpack and playergui. The replicatedfirst local script is supposed to be executed prior to the backpack/playergui even being created. No clue if itās related to the bug in the OP, but something is up with it.
[quote] Thatās because you should use
script.Parent:RemoveDefaultLoadingScreen() [/quote]
Even with it, it occurs:
I guess I donāt see what the problem is then. The script is executed ~1 second before the backpack/player gui, and for me the join process is really quickā¦
Oh, that third set of numbers is seconds. For some reason I thought it was milliseconds. The replicatedfirst code is being executed a full second before the backpack/playergui for me as well ā no issues here.
Iāve told a bunch of people to keep an eye out and screenshot the network diagnostics panel. It is evidently a periodic issue (that can be significantly game-breaking).
This could help:
In effected servers the filter chat function grinds to a halt and takes (from what Iāve heard) >30 seconds to filter chat in some cases. This could be due to very high pings though.
Note: weāre currently experiencing server issues with some API endpoints. This mostly includes an endpoint thatās necessary to join a game (same ācharacter takes 10+ seconds to loadā as last week); chat is also affected.
Could this be related to the server-no response issues? Iām operating entirely on word of mouth here since I havenāt seen this problem in person since April 17th so my descriptions can be a bit vague.
This has been happening in Survival 303! I have stats screenshots from three different people and interesting observations:
#1
#2
#3
#4
#5
#6
#7
#8
#9
The first screenshots were taken at the same time from Davidii2 and MemoryAddress (#1 and #5).
About 5 minutes later, the second images from Davidii2 and MemoryAddress were taken (#2 and #6).
10 minutes later, I got the first screenshot from qoazi. At this point I would like to note that qoazi had been playing in the same server this entire time without any noteworthy lag whatsoever. (#7)
Another 10 minutes later, I got another screenshot from qoazi and took one of my own at the same time and in the same place. (#3 and #8)
5 minutes later, qoazi found my character in-game (nowhere near me on the client) and took a screenshot, as did I from my machine. (#4 and #9)
ā Notes ā
- Usually this has been observed to happen in āwavesā as originally reported, but for some reason seems to be more permanent today.
- While this was going on, I was still able to see characters moving, but apparently they could not see me.
- The playerlist and chat are both run entirely from remotes, so you can see there are still some trickles through chat, but nowhere near what there should have been.
- As was reported, all the while I could move freely on the client and perform client-based actions without issue. This is no surprise but worth noting.
- On a few occasions, ping dropped by ~50000ms and flooded through some chat messages. However, it quickly rose back up and bottle-necked chat messages to nothing.
- Both MemoryAddress and Davidii2 were more-or-less disconnected from the server, but the clients were never actually disconnected.
- Ping reached values over 1,000,000 ms on both MemoryAddress and Davidii2
This happens to a large portion of Survival 303ās players and we have lost almost our entire playerbase due to this issue. Iām not sure if Iāve provided anything helpful, but if thereās anything else I can do PLEASE let me know. I would like to get this resolved.
So does anybody know anything more about this issue? If itās something Iām doing wrong Iād like to get it fixed, or if itās something on Robloxās end it would be nice to know the status on that as well.