I've been testing rsync with today's build.
I've got some problems with bricks that have content from older versions, so
I jumped
ahead and created new clean directories and restarted some tests. Shouldn't
the "cleanup"
or "self healing" feature fix anything broken?
I'm seeing some new problems which I haven't seen before.
This is a 3 brick setup using ALU. Config attached.
The Transport endpoint is not connected sporadically appears, but all
bricks are up. I assumed that if a brick was down, that ALU would
not choose that brick and allocate the file to another brick, so I don't
expect to see this type of error message
SOFTWARE.NOT/X3/EUROPE.VS_
149732 100% 709.46kB/s 0:00:00
SOFTWARE.NOT/X3/EXE.DE_
189 100% 0.00kB/s 0:00:00
rsync: mkstemp
"/glusterfs/SOFTWARE.NOT/HTTPD/NETSCAPE/I386/.BLANK.GI_.NS9xyV" failed:
Transport endpoint is not connected (107)
SOFTWARE.NOT/X3/FILTERS/
rsync: mkstemp
"/glusterfs/SOFTWARE.NOT/HTTPD/NETSCAPE/I386/.BUFFER.H_.qDnlA3" failed:
Transport endpoint is not connected (107)
SOFTWARE.NOT/X3/FILTERS/EBBMP9.FLT
44832 100% 214.77kB/s 0:00:00
SOFTWARE.NOT/X3/FILTERS/EBBMP9.INI
375 100% 0.00kB/s 0:00:00
SOFTWARE.NOT/X3/FILTERS/EBPCX9.FLT
43088 100% 4.46MB/s 0:00:00
_________________________________________________________________
http://liveearth.msn.com
