Hi
I'm using s3ql to backup several servers and roughly every other night
one of the backup fails during the rsync with:
rsync: write failed on "/example.file.txt": Transport endpoint is not
connected (107)
Is there anything I can do to reduce these errors? Is there any further
information about these failures that I could provide that could help
solve this?
I have been trying to work around this issue but I haven't found a
reliable way to do this as things seem to happen in this order during
the backup:
1. The s3ql filesystem is mounted and the rsync starts
1. The s3ql filesystem no longer appears to be mounted with a `df`
2. The rsync continues, apparently without a problem
3. The rsync fails, around 5 to 10 mins after 1. with "Transport
endpoint is not connected" error
4. `df` doesn't list the s3ql filesystem
5. `df` starts to report the s3ql filesystem with the "Transport
endpoint is not connected" error
Is there a better way than using `df` to check for this problem -- the
delay makes it unreliable -- this is a script I have been using to try
to work around this problem, running it before and after each backup:
#!/bin/bash
#
# check if a Transport endpoint is not connected
S3QL_DISCONNECTED_ENDPOINTS=$(df 2>&1 | grep "Transport endpoint is not
connected" | awk '{ print $2 }' | tr -d '‘' | tr -d '’:')
if [[ "$S3QL_DISCONNECTED_ENDPOINTS" ]]; then
for s3qlfs in $S3QL_DISCONNECTED_ENDPOINTS; do
if fusermount -u -z "${s3qlfs}" ; then
echo "Success unmounting ${s3qlfs}"
exit 0
else
echo "Problem unmounting ${s3qlfs}, consider using: killall -9
mount.s3ql"
exit 1
fi
done
fi
Any pointers for helping reduce these errors would be appreciated.
All the best
Chris
--
Webarchitects Co-operative
http://webarchitects.coop/
+44 114 276 9709
@webarchcoop
--
You received this message because you are subscribed to the Google Groups
"s3ql" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
For more options, visit https://groups.google.com/d/optout.
