Game suddently doesn't launch

Technical Support
1

Hello,

I previously played D4 without any problems. Then took a hiatus of several month and now tied to play again. Things worked fine at first but then suddenly the following happened: when I’m in the Battle.net client and press play the button first changes to “launching” and then “now playing”, but then all that happens is a sound rings and then it goes back to “play” without launching the game. I have tried the following to no avail:

  • Uninstall and reinstall: After this, I was able to play once, but then one the next launch the error reappeared
  • Disable Windows Defender (my only antirvirus)
  • Make sure that Battle.net and D4 are added to firewall
  • Renamed dstorage.dll
  • Run Scan and Repair in Battle.net client
  • Updated Graphics Driver (Running an NVidia GeForce RTX 3090/Intel i7-12700KF)

Despite all of this, the error remains. Is there anything else I can try? My latest FenrisDebug.txt is here: I 2026.07.27 05:48:18.730065 0 [Crash] User defined symbols path: C:\Program Fil - Pastebin.com

2

Hey Kazander,

Took a look at your FenrisDebug.txt. The log cuts off right at Initializing graphics subsystem… dpi = 144 144 — that’s the very first line of the D3D/renderer setup, and nothing logs after it. Combined with the “sound plays, then snaps back to Play” behavior you’re seeing, that points to the crash happening during GPU device/context creation rather than asset loading, TACT/CASC, or networking — all of which complete cleanly earlier in your log.

A couple things stand out:

DirectStorage: Enabled is set in the log. Since you already renamed dstorage.dll and it didn’t help, either the reinstall restored the original DLL, or the crash isn’t actually DirectStorage-related and just happens to occur right after that init step.

Your hardware (RTX 3090 + i7-12700KF) and driver update rule out most of the obvious culprits, so this is looking more like a GPU context/exception issue than a config problem.

The paste you posted stops right before the actual failure would show up. Can you post the rest of the FenrisDebug.txt (the lines right after the graphics init line), or better yet, check if there’s a .dmp crash dump in your D4 install folder (usually under a Crashes or similar subfolder)? That tail end / dump file should show the actual exception and give a much clearer read on what’s failing.

3

Hi Rocaholic,

Thanks a lot for having a look at the FenrisDebug.txt. The paste included the whole file I am afraid. Here is the most recent FenrisDebug.txt: I 2026.07.28 02:23:15.949871 0 [Crash] User defined symbols path: C:\Program Fil - Pastebin.com

I checked but there are no .dmp files in my D4 folder. Is there anywhere else I could look?

Furthermore I should say that this is really quite a strange error because yesterday suddenly D4 worked again. I could play for a day but now it’s the same error again. I don’t know what is exactly happening.

In any case, any pointers as to what I could try would be helpful.

4

Hi Kazander,

Thanks for the update and the new log — I wasn’t able to pull it up on my end just now (Pastebin seems to be timing out), so if you’re able to repaste the tail end of it directly in a reply, that’d help me confirm whether it’s failing at the same point as before.

On the .dmp files: they won’t be in the D4 install folder. Try checking:

%LOCALAPPDATA%\Diablo IV\ (look for a Crashes or ErrorReports subfolder)

%LOCALAPPDATA%\CrashDumps (Windows’ own crash dump location)

You can paste either path directly into File Explorer’s address bar to jump there.

The intermittent nature (working fine for a full day, then failing again) actually shifts things a bit. A corrupted install or a driver that’s flat-out incompatible would usually fail consistently — the fact that it comes and goes points more toward something environmental:

GPU thermal throttling or instability — with a 3090, if it’s overheating or losing power stability under load, the driver can silently reset right as the game tries to create its graphics context, which matches where your log cuts off

Overlay software conflicts — GeForce Experience overlay, MSI Afterburner/RTSS, Discord overlay, or RGB control software can all hook into that same graphics init step and cause sporadic failures like this

A security/Defender rule re-applying itself — even with it “disabled,” definition updates or scheduled tasks can silently re-enable protections between sessions

A couple things worth trying in the meantime:

Temporarily close/disable any overlay software (GeForce Experience, Afterburner, Discord) before launching

Check GPU temps during a play session if you can (afterburner/HWInfo) to rule out thermal issues

Check Event Viewer (Windows Logs > Application and System) around the timestamp of a failed launch for any related errors

If you can get me the new log content directly, I can compare it to the first one and see if it’s dying at the same spot.

5

Having the same issues and have also done the steps above.

6

Great, then you should have no problem providing the details of those Event Logs, FenrisDebug.txt, and/or information from Reliability Monitor. I will wait for more information from you.

7

Hi Kazander,

I went back to the Pastebin link, and it worked for me this time. Good news — this new log actually points somewhere different than I expected, and I think we’re onto something.

This time, the game doesn’t even make it to graphics initialization. It’s dying earlier, during [tact] Initializing cdn config — right after the CDN server list gets built. That’s a different failure point than your first log (which got as far as graphics init), which explains the intermittent behavior: you’re hitting a problem at a different stage each time, pointing to something environmental rather than a driver or install issue.

Here’s the part that stood out: every single CDN node in your log is a Korea/APAC server — blizzard.gcdn.cloudn.co.kr, kr.cdn.blizzard.com, level3.blizzard.com. For a US-based connection, you’d normally expect US/NA CDN nodes. Getting routed entirely to Korean servers is a strong sign of a DNS issue — either your ISP is returning bad records for Blizzard’s CDN, or something is redirecting your DNS resolution. If those distant nodes are then slow to respond or time out, the game can hang/fail during that CDN setup step — which lines up with what you’re seeing (launch sound plays, then it silently bounces back).

A few things worth trying:

  1. Flush your DNS cache: open Command Prompt and run ipconfig /flushdns

  2. Switch your DNS servers to a public resolver — Cloudflare (1.1.1.1) or Google (8.8.8.8) — in your network adapter settings, then try launching again

  3. Check C:\Windows\System32\drivers\etc\hosts for any unexpected entries (should just be default/commented lines)

  4. Make sure no VPN, proxy, or “smart DNS” service is running in the background, even briefly at startup

  5. If you’re running a router-level DNS filter or ad-blocker (Pi-hole, AdGuard Home, etc.), try temporarily bypassing it to test

Let me know if switching DNS servers changes anything — if it does, that confirms the routing/DNS theory, and we can look at making the fix permanent.

8

Hi Rocaholic,

Thanks so much for all your help! Being routed to Asian servers is normal I think as I am based in Asia.

I can report though that, for some reason, things are currently working again, i.e. when I randomly tried again this morning things suddently worked and that has been the status today. I hope it stays this way but will report back if there is a problem again. Once again, thanks for all your help!

9

Got ya, I should have asked that earlier on when I noticed the Korean-based servers instead of assuming wrong routing.

Glad to hear you are back in action again!