Re: [fuse-devel] read() commands do not get interrupted

[Miklos S. <mi...@sz...>]

Michael Berlin <mic...@go...> writes:

> Hi,
>
> We are using Fuse to develop our XtreemFS client and it works very well :-)
>
> I'm also using the interruption support of Fuse to allow interrupting 
> the retry of blocked operations. Therefore I did register my signal 
> handler on SIGUSR1 and it works as excepted for all commands except the 
> read() command: If a read() is interrupted, the signal handler does not 
> get called while the program performing the read() itself (for instance 
> a "cat ~/mnt/_Readme.txt") is properly interrupted. As the read() is 
> still running in the background and retrying, the release() on the file 
> is also delayed until the read() finally returned.
>
> I installed Fuse 2.8.5 with enabled debug symbols on my system and tried 
> to narrow down the issue:
>
> find_interrupted() in fuse_lowlevel.c fails to find the active read() 
> request: In particular, the check in line 1005 does not succeed:
>    if (curr->unique == req->u.i.unique) {
> as the value for curr->unique is always off by one (compared to the 
> value in req->u.i.unique), for instance 293 != 294.
>
> Are you aware of any other projects which use the Fuse interruption 
> support? So far I could not find a popular example where I could verify 
> my findings. Therefore, I also included instructions how to reproduce 
> the error for further debugging using our XtreemFS client at the end of 
> this mail.

I'm not aware of any project using interrupts.

> What about interruption of read ahead in Fuse in general? Are interrupt 
> requests sent to the Fuse library for every pending read()? From the 
> current experiences I guess that's not the case.

You are correct, not all interrupt of read(2) will reach the library.
And the cause is indeed read-ahead or read by an unrelated process.

Have you tried the 'direct_io' option?  That should help, but it has
side effects (no caching, no mmap, ...).  It depends on your case
whether you prefer to do 'direct_io' or not.

Thanks,
Miklos