Hi,
I am testing AFR in Gluster Storage Platform 3.0.5. I noticed the
recommended way to do self heal is by execting "ls -alR" (
http://www.gluster.com/community/documentation/index.php/Understanding_AFR_Translator).
I did this but it seems that not all files are synced. I then used another
command "find /mnt -type f -exec head -c 1 {} \; > /dev/null"
(/mnt is the
gluster mount point) and looks like files are synced now. The test I am
doing is relatively small scale with several hundred files and less than 1GB
in total.
"ls -alR" is taking much less time than "find /mnt/ -type f -exec
head -c 1
{} \; > /dev/null" and should be the best way for self-healing, if it
work.
My question is, is it a known bug that is fixed in newer version like 3.1.0?
or am I doing something wrong?
C. William