DDraceNetwork - developer

[ddnet/ddnet] Pull request opened: #1955 Cache serverinfo responses

I'm not sure I like it but it works

hey i have an issue with lags, that are probably caused by the ddos protection i already explained them here: (https://discordapp.com/channels/252358080522747904/293493549758939136/611160075201151004) today they are back, after some testing:

they are only presend on DDNet Ger servers (neither non-ddnet nor other regions)
on the Prediction Margin graph (with the keybindings ctrl-shift-d + ctrl-shift-g) the graph will freeze whenever i have lags, while the Gametime Margin continues. the gametime margin will freeze however, if i kill my internet manually.
the Prediction Margin spikes after big lags/resets

(edited)

Discord - Free voice and text chat for gamers

Step up your game with a modern voice & text chat app. Crystal clear voice, multiple server and channel support, mobile apps, and more. Get your free server now!

@Patiga I doubt it's because of the ddos protection as we really don't have filters that would drop actual teeworlds traffic, the hoster might add them, but in that case everyone else should feel them

hm

15:13

true

15:14

tho i dont really know where to point to next

15:15

lets blame telecom and wait till the problem fixes itself after some time again?

15:15

or do you have another idea about what could be blamed?

I'm not really familiar with those graphs, maybe we could find out what they really quantify and work from there?

I noticed the same type of lag on GER servers

15:43

I think that when the graph freezes it means packet loss, while spikes/red color either means jitter/delayed packets or prediction time reset

def-

[ddnet/ddnet] Pull request opened: #1956 Fix uninitialized variable

As reported by valgrind --tool=memcheck: ==344082== Conditional jump or move depends on uninitialised value(s) ==344082== at 0x483BC85: strlen (vg_replace_strmem.c:461) ==344082== by 0x5B9E61D: vfprintf_internal (in /usr/lib/libc-2.30.so) ==344082== by 0x5BB0409: vsnprintf_internal (in /usr/lib/libc-2.30.so) ==344082== by 0x222AE7: str_format (system.c:2350) ==344082== by 0x2196AB: CStorage::GetPath(int, char const, char, unsigned int) (storage.cpp:274) ==344082=...

17:10

def-

[ddnet/ddnet] Pull request opened: #1957 Don't access m_aClients wiht invalid ClientID

@heinrich5991 Do you happen to know how many bytes will the token take in SERVERINFO_EXTENDED?

17:33

@deen could you fix the uninitialized variable we have in the server aswell in #1956?

17:33

It's the members of CPlayer::m_Latency

bors[bot]

[ddnet:master] 2 new commits

f9598c8 Don't access m_aClients wiht invalid ClientID - def- f0245b1 Merge #1957 - bors[bot]

17:53

bors[bot]

[ddnet/ddnet] Pull request closed: #1957 Don't access m_aClients wiht invalid ClientID

@Learath2 m_Latency is a struct, so it is default-initialized

17:53

Info is the one that's not initialized properly apparently

17:54

done

Is there a way to kill a player when enter arena if its not official client ? something other than version control

what u just said makes no sense to me

@mohsen7s no included feature for that, you'd have to write it yourself

1

I'm getting some ridiculous lags on GER :/

same

19:56

packet loss, probably because of the ddos protection

deen: do you have a moment to show me how to set up servers?

20:00

also do we maybe have a sequence number in our packets so we can measure packet loss?

dunno, but the missing packets are very obvious in ctrl-shift-d ctrl-shift-g

20:05

sure, what do you need help with?

bors[bot]

[ddnet/ddnet] Pull request closed: #1956 Fix uninitialized variable

20:09

bors[bot]

[ddnet:master] 3 new commits

258eaf5 Fix uninitialized variable - def- fdaafea Also init m_aClients[*].m_Latency - def- 7d07b31 Merge #1956 - bors[bot]

@Learath2 ?

@deen I couldn't get the server... card blocked

20:20

I guess it'll have to wait until tomorrow

let me guess, ovh?

hetzner

ok

it's the banks fault

20:21

their fraud detection seems waaay too sensitive

I keep getting weird invoices from the russian hoster btw, but without any request to pay something

20:22

not sure if they're trying to rip me off or doing some tax scam

huh, probably some money laundring stuff

hetzner will be another ger2, right?

yeah

20:23

last time I got it I was busy so I couldn't watch how it behaved

do they have some special firewall what's the plan?

20:24

hetzner.de is down for me btw, might be a bad sign

Well I was hoping a 1gbit link would help

bors[bot]

[ddnet:master] 7 new commits

7f0589b Cache serverinfo responses - Learath2 79e4651 Only update serverinfo when needed - Learath2 3d495a9 Initialize variables - Learath2 52cca29 Use a std::list, improve variable naming - Learath2 c4d1811 Use automatic storage for the data - Learath2 c0b0f48 Oops - Learath2 0de0125 Merge #1955 - bors[bot]

20:42

bors[bot]

[ddnet/ddnet] Pull request closed: #1955 Cache serverinfo responses

Everyone else seems to be able to play, but I get a metric shit ton of packet loss

@deen here?