rsync - Jun 2006 - rsync-2.6.8 hangs on AIX

We're copying a large number of 256MiB-files from one
AIX-5.3 machine to anoter, using rsync-2.6.8 (with ssh).  In
the morning, the rsync process stopped to copy any more files
(it is still hanging as I write this, and we can leave it alone for
a couple of days).

There is another machine in between that forwards all tcp
packets to the target machine:

SOURCE
(rsync -z --rsh=ssh --remove-sent-files --partial --relative --files-from=- ./
target:/destdir)
-----------
     |
     v
-----------
socat
-----------
     |
     v
-----------
TARGET
-----------


On the "current issues" web page I found this request:
> Q: Rsync appears hung -- what should I do?
> 
> A: When experiencing a hang or freeze please gather the
> following information before killing the rsync process:
>
>     * The state of the send/receive queues shown with netstat on the two
ends.
>     * The system call that each of the 3 processes is stuck in (use truss
on solaris, strace on Linux, etc.).
I'd love to provide that information, but I've no idea how to
gather it on AIX.  Any hints?

-- 


"Feel free" ? 10 GB Mailbox, 100 FreeSMS/Monat ...
Jetzt GMX TopMail testen: http://www.gmx.net/de/go/topmail

On Tue, Jun 20, 2006 at 03:26:40PM +0200, Dominik Vogt
wrote:
> There is another machine in between that forwards all tcp
> packets to the target machine:
I haven't used socat before, but have successfully used nc (netcat), so
you might try switching over to that to see if the problem is socat.  If
that is not possible, make sure that socat isn't doing any character
conversions (e.g. it has options to affect line endings).
> I'd love to provide that information, but I've no idea how to
> gather it on AIX.  Any hints?
Doesn't AIX have the netstat command? Running it will show you the state
of the socket's input/output queues on whatever machine it is run on.
This can provide a clue as to where the blockage is by looking at where
the data is hung up.

..wayne..