TFS 4.0.4 Crashing bugSolved

I can't see a new TurboFloat.dll so I assume the client side is unchanged.

My test setup is running TFS on a Windows 7 physical machine with my product running on an AWS cloud VM.TFS 3.5.6.0 runs fine and the VM requests a lease successfully with the notification: New Connection from IP: 52.51.200.1 (the VM's public IP)

But when I try it with TFS 4.0.4, it activates and the server starts OK, but when the request lease comes in I get:New Connection from IP: ::ffff:52.51.200.1 and TFS stops running..

This is the problem details report produced when the server stops:

Problem signature: Problem Event Name: BEX Application Name: TurboFloatServer.exe Application Version: 4.0.4.0 Application Timestamp: 57dadf3a Fault Module Name: TurboFloatServer.exe Fault Module Version: 4.0.4.0 Fault Module Timestamp: 57dadf3a Exception Offset: 0008e32e Exception Code: c0000417 Exception Data: 00000000 OS Version: 6.1.7601.2.1.0.256.48 Locale ID: 2057 Additional Information 1: d4d6 Additional Information 2: d4d66734b49178e7a704f6ec6c365f18 Additional Information 3: 23f5 Additional Information 4: 23f5af9e0ac741a3636928eeb7f0cb29

It's not working for me at all

When I try to access TFS from a laptop on my local network I get :"Failed to clean the client-supplied activation block" followed immediately by "Failed to process the request: ios_base::failbit set: iostream stream error"

We'll try to reproduce this (TurboFloat Server shouldn't crash under any circumstance, so if it is, that's a big problem that we have to fix).

Regarding your question about the TurboFloat Library, yes there's a new version (everything is currently at 4.0.4): https://wyday.com/limelm/api/#turbofloat

Ahh, I missed the new TurboFloat library package in my haste to get the new version tried out. Sorry.Let me upgrade my client side to that and try it again.

I guess the TFS crashing if you call it from an un-upgraded product is a problem though.

Update

OK, I've rebuilt my app with the latest TurboFloat library, and on my local network everything is working fine. TFS doesn't crash anymore. 😀

But when I do a TF_RequestLease call from my AWS cloud VM I get back error TF_E_ENABLE_NETWORK_ADAPTERS

Now that is the same error I get on the AWS when I try to activate with TurboActivate.That's interesting because I wouldn't expect the TurboFloat client side to check the machine hardware in the same way that activation does..

Does that help your investigation?

Yes, 4.0.5 will be out later today, early tomorrow, that fixes these remaining TF_E_ENABLE_NETWORK_ADAPTERS false-positives.

Both of these bugs (crashing bug and the TF_E_ENABLE_NETWORK_ADAPTERS when the adapters were enabled) are now fixed in TF/TFS 4.0.5 now out.

Sorry to report that I'm still getting code 28 (TA_E_ENABLE_NETWORK_ADAPTERS) when I try to activate or request a lease from TFS with version 4.0.5 on an Amazon Web Services VM.

We haven't been able to reproduce this in our tests. Can you open PowerShell on that instance, run the following command, and then email me the results at wyatt@wyday.com?

Also, what version of Windows are using, and what "instance type" did you choose in Amazon EC2?

Thanks kaj, 4.0.5.1 is now out that fixes this bug.

Thanks for the quick fix. It's all working correctly now with 4.0.5.1 😀

Reply
Discard post