On Wed, 23 Jan 2013, Chen, Xiaoxi wrote: > > Hi list, > When first time I start my ceph cluster,it takes more than 15 minutes to get all the pg active&clean. It's fast at first (say 100pg/s) but quite slow when only hundreds of PG left peering. > Is it a common situation? Since there is quite a few disk IO and network IO during this time, what stop osd from faster initialize? > Can multi-monitor do any help with this? How many PGs and OSDs? Do the OSDs come up at the same time, or is there a small number of OSDs that initially get all PGs? sage
