On 16 Mar 2002, Martin Pool <mbp@samba.org> wrote:
> While testing a frontend to rsync, we have been bitten by rsync eating
> all the memory of the host. We have found that it was due to rsync
> trying to write an error and then failing to write the error and
> trying to report that it fails to write the error and so on. I have
> attached a patch which breaks the loop.
Thankyou for the patch.
I wonder if you can give more detailed instructions on how to
reproduce the bug? I suspect it happens when the network connection
is abruptly closed. I'd like to write a regression test for it if we
can.
--
Martin