You are most likely seeing this issue [1] where hung threads associated with the unreachable remote cluster are preventing the rbd-mirror daemon from releasing the image's lock on the local cluster. It has hopefully been resolved as part of Luminous but it isn't a good candidate for backport given all the associated changes it depends upon. [1] http://tracker.ceph.com/issues/18963 On Fri, Jul 28, 2017 at 3:36 AM, YuShengzuo <yu.shengzuo@xxxxxxxxxxx> wrote: > HI all, > > > > env: CentOS 7.2 > ceph: jewel-10.2.9 > > > > A image created in primary cluster, and replicated in secondary cluster by > rbd-mirror > > When the primary cluster crash or shutdown, I do promote image in secondary > with '--force' > the image still can't read or write. > > > > So I have to restart rbd-mirror daemon in secondary, it can work. > > > > Does anyone have any better method to solved it ? > > > > Yu Shengzuo > > -- Jason -- 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
