Re: (more) epoll troubles

[Andrew Morton <akpm@linux-foundation.org>]

On Sun, 31 Aug 2008 15:38:01 +0200 Michael Noisternig <mnoist@cosy.sbg.ac.at> wrote:

> Hello,
> 
> and sorry again if this is the wrong place to ask (again, please hint to 
> me to an appropriate place to ask in that case).
> 
> After experimenting with epoll edge-triggered mode I am clueless why on 
> a few occassions I seem to not get any input notification despite data 
> is available.
> 
> In detail: I have set up sockets with epoll events 
> EPOLLET|EPOLLRDHUP|EPOLLIN. When I get EPOLLIN for a socket, I read() as 
> long as I get what I asked for, i.e. whenever read() returns either 
> EAGAIN or less data than I asked for I take this as indication that I 
> must wait for another EPOLLIN notification. However, this does not seem 
> to work always.
> 
> Here is some log from my program:
> 
> 0x9e6b8a8: read not avail (1460/2048 read)
> i.e. tried to read 2048 bytes, got 1460 -> assume must wait for EPOLLIN 
> for more data to read
> (note that the fd is always in the epoll set with 
> EPOLLET|EPOLLRDHUP|EPOLLIN)
> 
> 0x9e6b8a8: read not avail (1460/2048 read)
> got EPOLLIN notification, read 1460 bytes again -> wait for another 
> notification
> 
> ...after a few minutes without any notification...
> 
> 0x9e6b8a8: GOT RDHUP (2001:0/80002001:0)
> i.e. got notification for EPOLLRDHUP|EPOLLIN (2001), was waiting for 
> EPOLLET|EPOLLRDHUP|EPOLLIN (80002001)
> 
> 0x9e6b8a8: read not avail (342/2048 read)
> then on trying to read we find data in the buffer (???)
> 
> 0x9e6b8a8: stored 16384 bytes
> this is the amount of data we expected to get /immediately/, i.e. 
> without a minutes delay and rdhup
> 
> Of course you might think that the other side sent the data only just 
> before closing its side, but it is unlikely since we expected to get the 
> 16384 bytes in one go (this is how that L7 protocol works), and I see 
> this odd behaviour of missing input notifications now and then.
> 
> Does anybody know/have an idea why I am missing EPOLLIN notifications?
> 

If you can put together a little test case which others can run then we
should be able to get this resolved super-quick.

Thanks.