Hello,
Vijay Bellur wrote:
I considered this option too, but there are the following issues: We assume that I install one new machine with CentOS7 and start copying the data from CentOS5 1. Gluster 3.0.4 just does not work on CentOS7, it just crashes, and I need it in order to copy the data 2. This is damn slow and I doubt it will copy all the files, because I found some weird behavior in 3.0.4 which for some reason does not show all the files despite the fact that you can access them, ls or find just doest not show them.
The volumes in 3.0 are created manually by me. I am pretty familiar with them. When I created the volumes with the newer version, which used the same test bricks as 3.0 I used the following command: gluster volume create testvolume 10.15.236.249:/exportdir1 10.15.236.249:/exportdir2 10.15.236.249:/exportdir3 10.15.236.249:/exportdir4 10.15.236.249:/exportdir5 10.15.236.249:/exportdir6 10.15.236.249:/exportdir7 force I took a look at the new volume config files created from gluster 3.7 and they are quite similar to the ones manually created by me for 3.0.4. It also looks like working when trying to access 100k test files created under 3.0.4. However I can not be 100% sure that this will be ok with production data, that is why I am asking for suggestions. Thank you. Regards, Deyan |
_______________________________________________ Gluster-users mailing list Gluster-users@xxxxxxxxxxx http://lists.gluster.org/mailman/listinfo/gluster-users