rsync - Dec 2007 - Rsync rsync: writefd_unbuffered failed to write ?

I am running jungledisk and this is my script:

#!/bin/sh
### Backs up office data to Jungledisk using rsync
LOGFILE=/var/log/backup-jd.log
## Start in rc.local or here
#/usr/local/bin/jungledisk
mount /mnt/s3
echo "`date +"%F %R"`: Start backup-jd" >> $LOGFILE
rsync -r --inplace --size-only --bwlimit=50 /home/shares/allusers/127 /mnt/s3
echo "`date +"%F %R"`: Finish backup-jd" >> $LOGFILE
umount /mnt/s3

Everytime i run i get this error, it driving me nuts! any ideas would be
appreciated.
rsync: writefd_unbuffered failed to write 32768 bytes [sender]: Broken pipe (32)
rsync: connection unexpectedly closed (30 bytes received so far) [sender]
rsync error: error in rsync protocol data stream (code 12) at io.c(454)
[sender=2.6.9]


--
This message was sent on behalf of scott@techeffectcomputers.com at
openSubscriber.com
http://www.opensubscriber.com/messages/rsync@lists.samba.org/topic.html

On Thu, 2007-12-13 at 06:28 -0500, scott@techeffectcomputers.com
wrote:
> I am running jungledisk and this is my script:
> 
> #!/bin/sh
> ### Backs up office data to Jungledisk using rsync
> LOGFILE=/var/log/backup-jd.log
> ## Start in rc.local or here
> #/usr/local/bin/jungledisk
> mount /mnt/s3
> echo "`date +"%F %R"`: Start backup-jd" >>
$LOGFILE
> rsync -r --inplace --size-only --bwlimit=50 /home/shares/allusers/127
/mnt/s3
> echo "`date +"%F %R"`: Finish backup-jd" >>
$LOGFILE
> umount /mnt/s3
> 
> Everytime i run i get this error, it driving me nuts! any ideas would be
appreciated.
> rsync: writefd_unbuffered failed to write 32768 bytes [sender]: Broken pipe
(32)
> rsync: connection unexpectedly closed (30 bytes received so far) [sender]
> rsync error: error in rsync protocol data stream (code 12) at io.c(454)
[sender=2.6.9]
To get more information, please strace rsync (which is especially easy
because all of the processes are local) and enable -vvv verbosity level:

strace -f -o /tmp/rsync.log rsync -vvv -r --inplace --size-only --bwlimit=50
/home/shares/allusers/127 /mnt/s3

If this does not make the problem obvious, post the output and I will
take a look.

Matt

On Thu, 2007-12-13 at 09:48 -0500, Scott Parry wrote:
> I have enclosed the log for you to take a look at. 
> I am running this from the root account.
Ah, this part beginning on line 30371 seems to be the root of the
problem:

22250 write(3,
"\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"..., 262144
<unfinished ...>
22248 <... read resumed>
"\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"..., 262144) =
262144
22250 <... write resumed> )             = -1 EIO (Input/output error)
22248 select(5, NULL, [4], [4], {60, 0}) = 1 (out [4], left {60, 0})
22248 write(4,
"\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0"..., 6400
<unfinished ...>
22250 write(4, "S\0\0\10rsync: write failed on \"/mnt"..., 87
<unfinished ...>
22248 <... write resumed> )             = 6400

Rsync aborted because its write to the JungleDisk virtual filesystem
failed with an I/O error.  You should look at the JungleDisk logs to
find out why JungleDisk returned the I/O error.
> its says unfinished after every file?
In strace output, "unfinished" after a system call just means that
something happened in another process before the call finished.  The
result of the call is given in a subsequent line with "CALL-NAME
resumed".  This isn't anything bad.

Matt