VS 2026 crashing on Save with ReSharper enabled.

Avatar
Robains
Created

I created a support ticket for this (8887611) issue and got a confirmation email but when I click on the link to the support ticket I get a 404 not found?   Anyway ….

Doesn't matter what mode I run ReSharper in Out-of-Process or In-Process it will frequently cause VS to go unresponsive as soon as I click on the “Save All" button in the VS toolbar.   This is ONLY started happening with the ReSharper version 2026.x.x.  I've tried all the recommended solutions except #4 source control (pointless to proceed without source control) and NONE have solved the issue:

Recommended Fixes

Try these steps in order to stabilize your environment:

  1. Clear ReSharper Caches: Corrupted caches often trigger exceptions during auto-save or background compilation.
    • Go to Extensions > ReSharper > Options > Environment > General.
    • Click Clear caches. Restart Visual Studio.
  2. Disable Out-of-Process Mode (if active): If you are using ReSharper’s new out-of-process engine, it may be unstable on certain configurations.
    • Go to Extensions > ReSharper > Options > Environment > Products & Features.
    • Uncheck Run ReSharper in a separate process and click Apply to run in-process
  3. Toggle Solution-Wide Analysis: Constant background analysis on save can overwhelm system memory.
    • Go to Extensions > ReSharper > Options > Code Inspection > Settings.
    • Uncheck Enable solution-wide analysis temporarily to see if the crashes stop.
  4. Disable Visual Studio Git/Source Control Integration: Source control plugins trying to lock files immediately upon a ReSharper-assisted save can cause a hard crash.
    • Go to Tools > Options > Source Control > Plug-in Selection and set it to None.
  5. Exclude Folders From Antivirus: Windows Defender or other antivirus software actively scanning changing cache directories during a save operation can lock files and force Visual Studio to crash.
    • Add the ReSharper cache folder (%LOCALAPPDATA%\JetBrains\Transient) and your project directory to your antivirus ignore list.

I've also tried the standard other options:

Visual Studio 2026 repair

OS - sfc /scannow  and DISM options

The ONLY option that prevents VS crashing is if I disable ReSharper.  I can't continue working like this so I need a solution, if I can't get one then I'll stop using ReSharper and cancel my subscription … not trying to be dramatic just a fact as I got a million other tasks on my plate and debugging ReSharper isn't one of them.

Attached a Zip file of all the logs created for today that cover the VS crashing time period.

Upload id: 2026_06_05_wofCBr43uDRJPpkQ6iE3in (file: JetBrainsLog.JetBrains.2026-06-05T09-09-26.ReSharper_Backend64c#58572.zip)

 

Votes

0

Share

5 comments
Avatar
Alexander Ulitin
Created

Hi, thank you for your time and feedback!

I see a lot of exceptions from Telerik.Blazor.VSPackage.BlazorCompletionCommandHandler in your logs, but they are probably not related to the hang. Unfortunately, the logs do not make it clear why this happened.

Could you please collect a PerfView snapshot as described here: https://youtrack.jetbrains.com/articles/SUPPORT-A-2191  (use the PerfView_until_S_pressed.cmd script).

Thanks

 

0
Avatar
Robains
Created

The link you provided is for “out of process” … as I indicated in my original post, this issue happens with BOTH “in process” and “out of process” … and the issue is STILL happening 2026.1.3 and VS 18.7.1.

The ONLY way I'm able to save my solutions is if I disabled or uninstall JB ReSharper.

Regardless I was able to replicate … I switched to “out of process” and ran the PerfView script outline in your link.  That script did NOT create a .zip file so I created one with 5 files.

It's a large file so here is the link to my Google drive:

PerfView data

Yes, those Blazor entries have nothing to do with this situation because like I indicated, if I uninstall ReSharper I don't have any issues.  So clearly ReSharper is the source of the problem and it's becoming so annoying that I'm considering request for a refund and I'll just uninstall ReSharper for good as “my solution”.  I sure hope you can find something in my perfview files.

Rob.

 

0
Avatar
Alexander Ulitin
Created

Thank you, Rob.

I have updated the instructions for PerfView snapshots. They are suitable for both in-process and out-of-process mode; there was no need to switch.

Based on your snapshot, I can see that the main thread is busy waiting for something, but it is not possible to tell which specific part of the product is responsible for this wait, only kernel stack are visible on my machine. This is probably due to an unmerged PerfView snapshot. The zip/merge phase is critical for viewing snapshots outside your machine.

| Name                                                            Inc %          Inc |
|  |||||||+ ntoskrnl!KiSystemServiceExitPico                        0.2    13 762.521 |
|  ||||||| + win32k.sys!NtUserMsgWaitForMultipleObjectsEx           0.2    13 762.418 |
|  ||||||| |+ win32kfull.sys!NtUserMsgWaitForMultipleObjectsEx      0.2    13 762.418 |
|  ||||||| | + win32kfull.sys!xxxMsgWaitForMultipleObjectsEx        0.2    13 752.692 |
|  ||||||| | |+ ntoskrnl!ObWaitForMultipleObjects                   0.2    13 602.717 |
|  ||||||| | ||+ ntoskrnl!KeWaitForMultipleObjects                  0.2    13 602.717 |

May I ask you for a different type of diagnostics to trace the problem? Sorry, it is not always obvious whether PerfView or a memory dump is required.

Could you please collect a memory dump of the 'devenv' process? It would be better if you were in In Process mode. You can collect the memory dump directly from Task Manager (or Process Hacker or Process Explorer).


You can also use https://uploads.jetbrains.com/ to share large files with us without exposing them to the internet

Thanks!

0
Avatar
Robains
Created

Ok, I removed the perfview zip from my drive and will use JB in the future, it was a large file hence why I didn’t as I was concerned JB might have upload file size limits.

It’ll be this Friday before I can test this again and get a memory dump and upload to JB.  The memory dump could be very large so hope JB will permit upload.

Not sure if this is relevant, but I’m using AMD 9975WX CPU, 256GB ECC DRIMM, and two nVidia 5090s with two Samsung 9100 M.2 on an ASUS Pro WS WRX90E-SAGE motherboard with latest BIOS/EFI and Windows 11.  Drivers are all current.  It may not be relevant, but 32/64 core count does increase threading management so if there are special compiler flags you folks use that aren’t AMD friendly or can’t operate with high core counts (aka threading deadlock) that might be a direction your development team could pursue. 

Also important to note that this issue only started with JB 2026 versions, I never ran into this (on the same hardware) when running JB 2025 versions, so it’s “something” that was introduced in the JB 2026 versions.

0
Avatar
Robains
Created

I was able to replicate and have uploaded two devenv dumps.  UploadID: Upload id: 2026_06_26_s1q9WdGWfw7Gw9fyoo1L8x

Rob.

0

Please sign in to leave a comment.