Netchannel: failed reading message net_SetConVar from <IP of the server>

  • Please go to our Discord or SteamRep for trading. Bans and appeals go HERE
PsychoRealm

PsychoRealm

Australian Skial God
Contributor
As I reported here, my attempt to connect freezes somewhere in the middle and when I open the console I get the following error:

Netchannel: failed reading message net_SetConVar from 67.202.71.66:27015.

384CAC97378A62FA7F7F3A452C9CD472937F2760


As you could see from the screenshot this error is obviously related to Mann Vs Machine update as it talks about "bot_eye_glow_soft".

As I read in THIS article:
"Basically in a nutshell when you connect the server sends a massive list of objects that the server is currently using. This includes text messages, player models, map models, cvar settings, etc... Normally it only sends exactly what you need and anything missing is loaded on the fly (this is why some slower PC's can actually see all black player models for a few seconds after connecting and joining a team).

Apparently since the scout update the basic string table list has expanded to a point where burst transmittance from the server containing the entire list is resulting in an outright overflow of the client's internal Netchannel buffer. This results in the buffer being dumped (for sanity) and thus the Netchannel parser throwing up it's hand's with an empty buffer leaving you in an endless connection cycle (wohooo!).

Connecting to 74.63.197.179:27015...
Connected to 74.63.197.179:27015

Team Fortress
Map: ctf_2fort
Players: 24 / 26
Build: 3755
Server Number: 5

Attemped to precache unknown particle system "rockettrail_!"!
Netchannel: failed reading message net_SetConVar from 74.63.197.179:27015.

(Your console log may result in more than a single Netchannel error but the basic problem is the same)"

Am I the only one getting this error??
 
PsychoRealm

PsychoRealm

Australian Skial God
Contributor
THIS thread as well as "Users found this page by searching for:" make me think I'm not the only one who gets this problem.
 
ModSe7en

ModSe7en

Legendary Skial King
Contributor
bot_eye_glow_soft is unrelated, I get that but not the net_SetConVar error. It's kind of like the particle system backblast error, if anyone else gets it.
 
Ninja_DC

Ninja_DC

Face-Melting F2Per
  • Updated client rate value to default to 30000 to improve network performance for most players
    • Rate was previously set by Steam’s Internet connection speed setting
  • Updated client rate value to be stored in the config.cfg file
Team Fortress 2
  • Fixed a crash caused by a string table overflow
Now, stop complaining.
 
Chiefmiest

Chiefmiest

Server-Clearing Cynic
Ninja_DC said:
  • Updated client rate value to default to 30000 to improve network performance for most players
    • Rate was previously set by Steam’s Internet connection speed setting
  • Updated client rate value to be stored in the config.cfg file
Team Fortress 2

  • Fixed a crash caused by a string table overflow
Now, stop complaining.
Click to expand...

shut the fuck up
 
takethepants

takethepants

Australian Skial God
Contributor
Ninja_DC said:
  • Updated client rate value to default to 30000 to improve network performance for most players
    • Rate was previously set by Steam’s Internet connection speed setting
  • Updated client rate value to be stored in the config.cfg file
Team Fortress 2

  • Fixed a crash caused by a string table overflow
Now, stop complaining.
Click to expand...

Too bad they fix the crashing bug. Actually made it worse
 
PsychoRealm

PsychoRealm

Australian Skial God
Contributor
takethepants said:
Too bad they fix the crashing bug. Actually made it worse
Click to expand...
It's...... Valve...
Seriously.. It's ridiculous.. One of the biggest games developing companies keeps releasing patch by patch and fucking it up EVERY single time. I mean, what all those betas and armies of testers are for if the first bunch of bugs fixed in release that comes right after the major update is:

- Fixed a crash caused by a string table overflow
- Fixed a client crash related to the freeze panel
- Fixed a client crash related to ragdolls
- Fixed a client crash related to the Mann vs. Machine upgrade panel
- Fixed a client crash related to Mann vs. Machine stats
- Fixed a crash related to server plug-ins spawning a Mann vs. Machine Tank before the game is ready

What the fuck had they been testing? My dead grandma's vagina???
 
Oli

Oli

Uncharitable Spy
Contributor
Ok, myself and some other guy just had this new problem since the last update. I could join some servers (e.g. Skial NY), but could not join any Chicago or LA servers for the same reason as OP.
 
Roman

Roman

Banned
Contributor
Just got into a Mann Up server that kicked us because it was running an older version of the software.

Valve servers: Out of date.
 
PsychoRealm

PsychoRealm

Australian Skial God
Contributor
Roman said:
Just got into a Mann Up server that kicked us because it was running an older version of the software.

Valve servers: Out of date.
Click to expand...
Fucking awesome. I thought QAD was the worst vendor in terms of SDLC. Apparently, I was wrong. Valve beats all the records. Screwing it up so bad right before the weekends... /sigh.
EDIT: no wonder why my Russian friends are cussing out Valve so bad.
 
Ninja_DC

Ninja_DC

Face-Melting F2Per
I honestly believe Valve doesn't properly debug updates and fixes.

Did you notice how quickly a second update was released yesterday?
Not to mention the numerable client crashes caused by the MvM mode.

Fuck that, valve doesn't debug shit at all.
It shows how much valve only cares about the Mann. co store revenue for TF2.
 
takethepants

takethepants

Australian Skial God
Contributor
Ninja_DC said:
I honestly believe Valve doesn't properly debug updates and fixes.

Did you notice how quickly a second update was released yesterday?
Not to mention the numerable client crashes caused by the MvM mode.

Fuck that, valve doesn't debug shit at all.
It shows how much valve only cares about the Mann. co store revenue for TF2.
Click to expand...
They obviously do, but in a controlled test environment. It's an entire new game mode, there's always going to be bugs. They probably knew about some of them, but even though a lot of people are getting errors, everything works fine 95% of the time for me. They probably have a bank of test servers that are Windows and Linux, but they can't account for the billion different hardware variations. They would take way longer to release if they had to test a bunch of variations. They're thinking is hey, if it works 95% of the time and we can start making boatloads of money NOW rather than 6 months from now, they're going to release now. With a million people playing, there's always going to be problems. I think all of this is going to come down to optimizing the game mode to use less resources, etc.

Working on a software project myself at a large company, we only have 5 production servers and we KNOW ABSOLUTELY the hardware/software configurations and even though our QA environment matches almost exactly, we still release stuff that breaks occasionally. We just had a two day downtime for a bug that didn't show up at all in our QA environment, but caused our production environment to become unusable. Shit happens. We had been testing this stuff for at least a month with no issues and then BAM issue. We only have to deal with about 300 angry people. I can't imagine what it's like to be hammered with thousands, if not tens of thousands of emails complaining about bugs. I think all crash reports get auto sent to them anyways so complaining to them directly doesn't really help much.

For Roman's issue, I bet the server lost connectivity during the update and they didn't catch it over the 1000 servers that updated.
 
  • Like
Reactions: Ninja_DC
Ninja_DC

Ninja_DC

Face-Melting F2Per
Thanks for the info, I completely forgot about different hardware configurations even though it was something i actually knew about.
 
You must log in or register to reply here.
Top Bottom