rsync - Jan 2010 - Problems with rsync between NAS mounted filesystems

Hi All,

Here is my setup.  I've to two NAS filesystems mounted on a SunFire  
V490.  They're mounted on /rsync/ieeprodhome/ECF and  
/rsync/ieeprodhome2/ECF.  The ieeprodhome filesystem is 1.008 TB wiht  
812GB used and the ieeprodhome2 filesystem is 1.2TB with 813GB used.   
One other thing to add that might be of importance is that the source  
filesystem has over 100k files in it.

Here is the command that I'm trying to use:

/opt/csw/bin/rsync -avvvv --delete /rsync/ieeprodhome/ECF/  
/rsync/ieeprodhome2/ECF

This will seem to run just fine for a couple of days, but then it will  
error out with the following:

delete_in_dir(70/36/35/70363504)
[generator] make_file(70/36/35/70363504/1000B,*,2)
[generator] make_file(70/36/35/70363504/GWS,*,2)
[receiver] receiving flist for dir 39230724
[generator] receiving flist for dir 39230724
received 1 names
[generator] flist start=137943897, used=1, low=0, high=0
[generator] i=137943897 6 70/36/39/70363977/1000B/02020924.GZ  
mode=0100750 len=1867 uid=5300 gid=802 flags=0
recv_file_list done
[receiver] receiving flist for dir 39230725
[generator] receiving flist for dir 39230725
received 0 names
[generator] flist start=137943899, used=0, low=0, high=-1
rsync: connection unexpectedly closed (338935497 bytes received so  
far) [sender]
_exit_cleanup(code=12, file=io.c, line=600): entered
rsync error: error in rsync protocol data stream (code 12) at  
io.c(600) [sender=3.0.6]
_exit_cleanup(code=12, file=io.c, line=600): about to call exit(12)

Any clues on why it might be losing connection?  I don't have rsync  
running in daemon mode, so what would it be losing a connection to?

Ed Coates

On Wed, Jan 13, 2010 at 7:23 AM, Ed Coates <ed at coatesfam.us> wrote:
> rsync: connection unexpectedly closed (338935497 bytes received so far)
> [sender]
> [...] Any clues on why it might be losing connection?

In a local connection the sender is the client side, so this is indicating
that the receiving side went away.  Because you got no terminating message,
it is probable that the receiver was the process that died first, and its
message did not make it through the generator before things closed down.
 You might try the latest "nightly" tar file rsync release of
3.1.0dev,
which should do better at getting the errors to you when an abnormal
shutdown is happening.  Or, to diagnose what you're running now, you can
turn off some of the verbosity and attach an strace to each of the receiving
rsyncs after the transfer starts up.  That should reveal why the process
died.

..wayne..
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<http://lists.samba.org/pipermail/rsync/attachments/20100115/f55efd31/attachment.html>