New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Creation of a disk file system may fail #88
Comments
Just a minor clarification. It also does not work when mounting to a local folder. It gives the same error as trying to mount as a "disk" drive. |
@hasse69 thanks for the clarification. [The failing |
@hasse69 can you please let me know if this problem also happens with plain MEMFS? From a
This will help us understand whether this is a FUSE/CYGFUSE issue or a general issue on 32-bit Windows. |
Same problem with memfs-x86.exe
|
Thanks. This confirms that the problem is more general than just the FUSE layer. |
I am encountering the same problem but in my case the issue is not restricted to only 32bit machines. It is not happening on all machines but I am unable to find the pattern or similarities on the failing machines (Have seen on Windows 7, Windows 8 and Windows 10). Diagnostics result from one of the failing machines
|
@bonagerivenkatesh can you please confirm that your issue is not related to issue #177? Have a look at the comments towards the end, especially the comments made by @thinkport. |
Also if #177 is not your issue, I would love if I could get debug access (or even just access) to one of the machines that exhibits this problem. |
@billziss-gh unfortunately the fix for #177 didn't fix the problem. I have sent a private email for you to schedule a call to access the machine with the problem. |
@billziss-gh I added some logs to print the file name and number where this error was thrown and it is pointing to
in I used The mount still failed with the error c000002. Please let me know if any of these info helps to narrow down the issue. |
@bonagerivenkatesh thanks for the info, but it was already known. Please see top post of this thread. As said before, I would love to get temporary access to one of the machines exhibiting the problem. |
@billziss-gh Unfortunately its happening only on my client machines and I am still in search of one who is ready to give the debug access to us. |
Understood. It looks like you are able to run custom builds on those machines so it might be worth trying custom FSD builds to see if we can pinpoint the problem. I am currently on an iPad so I can only type so much detail, but here are some pointers:
If we do not see any of these debug prints (and assuming that you have configured things correctly to get debug prints visible on the system) then it must mean that something else intercepts that call and fails it, likely a filter. So can you also let me know what the |
@billziss-gh ya I am able to run custom builds on those machine, I will try your suggestion of |
@bonagerivenkatesh the filterDrivers.txt file does list a couple of third-party filters:
|
@hasse69 if you are still on the thread and have access to the system that exhibited the original problem: could you check whether that system has any of the aforementioned filters? (Wrkrn and stcvsm?) You can use the |
Sorry, I upgraded my system to 64-bit a long time ago. I no longer have access to it :( For what it is worth, here are my filters on my upgraded system
|
@billziss-gh luckily I was able to reproduce this issue by just installing the |
Let me see if I can easily install this software in one of the VM's I use for kernel debugging. |
@billziss-gh to add more info to the above the line 81 failed so it never entered the |
Here is my debugging analysis. In WinDbg I enabled WinFsp debug prints with:
Then got the following logs back:
These logs make no sense: where is the
It appears that Let's confirm our suspicion. I placed a strategic breakpoint in
I may add to this post as I discover additional information. |
@bonagerivenkatesh thanks for additional info. Yes, I have found the same. I am in the process of debugging this. See comments above. |
@bonagerivenkatesh I have updated my debugging analysis, which I posted above. The conclusion:
|
I am trying to find a bug report or support email on their website, but have not been successful so far. EDIT: Found this: support@storagecraft.com. I will write an email to them and see where it takes us. |
@billziss-gh sounds good. I really appreciate your involvement and support. I found this https://support.storagecraft.com/s/contactsupport (requires registration though to post here) But on a side note I heard today that some HP software manager also causing this issue (I could not reproduce the issue though), is there anyway the priority of the driver can be changed so that the request directly goes to WinFsp? If that is possible then it would be great as not sure how many legacy drivers are present that might cause similar issues.. |
WinFsp is a file system driver and has no "priority". Filters are free to intercept its requests and in some cases they may erroneously handle them. Unfortunately I am not aware of any method to get around that. |
Sent a mail to support@storagecraft.com. Will see what happens. EDIT1:
EDIT2: |
StorageCraft resolved this issue a couple of days ago. My thanks to the StorageCraft people if you are reading this! Their newest product SPX 6.8 includes an stcvsm.sys driver with an altitude of 388250. I have confirmed that SPX 6.8 fixes this issue. @bonagerivenkatesh can you also confirm so that we can close this issue? |
@billziss-gh I will not be able to confirm since I am not even certain the original issue I reported has anything to do with SPX. It might have been another filter driver that caused the issue though but I do not think we will ever know for sure. As for my case, I am fine with closing this. |
@billziss-gh I was able to confirm that this is fixed in the latest version of SPX. |
@hasse69 thanks. I knew that you had updated that system and you no longer had the original so you could not confirm. If there was another filter driver it will eventually turn up and we might be able to point to a similar cause as being the problem. @bonagerivenkatesh you are welcome. Closing this. |
There is a report that creation of a disk file system on 32-bit Windows may fail with
STATUS_NOT_IMPLEMENTED
(c000002
) duringFspFsctlCreateVolume
. The issue happens on 32-bit Windows only. See issue hasse69/rar2fs#75.According to @hasse69 this happens when mounting a cygfuse file system as a "disk" drive (
\Device\WinFsp.Disk
), but not a "network" drive (\Device\WinFsp.Net
). After adding debug log instructions to the WinFsp DLL we were able to isolate this problem in the followingDeviceIoControl
call:https://github.com/billziss-gh/winfsp/blob/v1.0/src/dll/fsctl.c#L83-L90
This fails with
ERROR_INVALID_FUNCTION
(which gets translated toSTATUS_NOT_IMPLEMENTED
). The likely reason for theERROR_INVALID_FUNCTION
error return is aSTATUS_INVALID_DEVICE_REQUEST
response from the FSD.I am unable to reproduce this problem on any of my local 32-bit VM's. Complete
diag.bat
output from @hasse69 below:The text was updated successfully, but these errors were encountered: