Bug #59735
openfs/ceph: cross check passed in fsid during mount with cluster fsid
0%
Description
Right now, basic checks are done for the passed in fsid - just that it "looks" like a fsid. However, the kclient can cross check this fsid with the cluster fsid and fail the mount (maybe?) on mismatch.
Updated by Venky Shankar 11 months ago
- Subject changed from fs/ceph: cross check passed in fsid with cluster fsid to fs/ceph: cross check passed in fsid during mount with cluster fsid
Updated by Xiubo Li 11 months ago
Hu Weiwen has one patch series maybe fixing this issue: https://patchwork.kernel.org/project/ceph-devel/list/?series=745634, I haven't check it in detail yet.
Updated by Xiubo Li 11 months ago
Xiubo Li wrote:
Hu Weiwen has one patch series maybe fixing this issue: https://patchwork.kernel.org/project/ceph-devel/list/?series=745634, I haven't check it in detail yet.
Checked Weiwen's patches, they just save the 'fsid' from the mount options and do the sanity check, but didn't check whether is the 'fsid' validate or not by comparing it with the one from ceph cluster.
So this fix should be easy based Weiwen's patches, which have been in the 'testing' branch now.