Are there any recommendations for improving the performance of migrating data off of an OSD - i.e. marking it OUT and then waiting for it to be "safe-to-destroy"? I have a Luminous cluster with 113 filestore OSDs spread fairly evenly across 6 storage servers, I marked 1 of the OSDs "out" over 24 hours ago and it is still a long way from being safe (52 pgs currently mapped, it started with about 160 yesterday). It seems like it is taking far longer than should be expected. Perhaps I need to tune some parameters to improve the performance but Im not sure where to look. Any suggestions? Thanks Wyllys Ingersoll Keeper Technology, LLC -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
