Takeshi Nakatani

I think it is possible to support both FUSE2 and FUSE3 with the same source code(using ifdef etc.). This is a suggestion, but is it possible to separate the packages...

@giulianoc If you expect the cache of s3fs to work effectively as the problem of #1961 is solved, you may be able to judge by the following conditions. If the...

Let me supplement - About Client2 catching file updates etc. for scenario 1. s3fs can cache file stats and its content and complete its access locally. However, when a file...

I still can't reproduce this deadlock problem. Could you give me more information for reproducing? (I think it won't take long to fix if it can be reproduced)

The #2008 and #2012 have been merged and my tests no longer output error reports. @gaul Are the errors still ongoing?

@gaul Thanks for your report. Could you tell me about the situation of this phenomenon and the steps I should reproduce this? (Is it that Semaphore::wait couldn't return ever?)

@gaul Thanks, I will try to reproduce it

@farblos I'm sorry for my late replying. I posted PR #1789 to fix this bug. If you can still try it, please try the code #1789. Thanks in advance for...

Sorry for my late reply. From the logs it looks like s3fs is blocked after calling libfuse main function. Normally, the following line should be output to the log immediately...

This fix is noteworthy. Ultimately, I think it's better to wrap or separate the I/F functions, taking into account the differences between fuse2 and fuse3. (I'm sure that support for...