Description of problem: Bucket created via NFS mountpoint on master zone does not sync to other zones. Version-Release number of selected component (if applicable): ceph-radosgw-10.2.5-27.el7cp.x86_64 How reproducible: Always Steps to Reproduce: 1. Configure ganesha on a master zone involved on multisite config 2. Mount the share on a client and create a directory 3. The corresponding bucket is created on the master zone. Check if it syncs across other zones Actual results: Bucket fails to sync to other zones. On master: # swift -A http://rgw1:8080/auth/1.0 -U testuser:swift -K 'tPI9KeitPljlInw2NpxHVD62E97LwGDfVXNflrP4' list buc66 ks-post.log On other zones: # swift -A http://rgw2:8080/auth/1.0 -U testuser:swift -K 'tPI9KeitPljlInw2NpxHVD62E97LwGDfVXNflrP4' list buc66 Container u'buc66' not found Additional info: 2017-02-21 13:12:47.110984 7f65567fc700 0 ERROR: failed to get bucket instance info for buc66(@[]) 2017-02-21 13:12:47.111030 7f64c27fc700 20 cr:s=0x7f64b01d0420:op=0x7f64b03d33e0:26RGWGetBucketInstanceInfoCR: operate() 2017-02-21 13:12:47.111045 7f64c27fc700 20 cr:s=0x7f64b01d0420:op=0x7f64b03d33e0:26RGWGetBucketInstanceInfoCR: operate() returned r=-2 2017-02-21 13:12:48.330488 7f64c27fc700 20 [inc sync] skipping object: buc66:e92e12d2-82be-44a0-892f-f770a2fb8041.154151.1/ks-post.log: squashed operation 2017-02-21 13:12:48.330489 7f64c27fc700 20 parsed entry: id=00000000003.17600.2 iter->object=ks-post.log iter->instance= name=ks-post.log instance= ns= 2017-02-21 13:12:48.330493 7f64c27fc700 20 [inc sync] skipping object: buc66:e92e12d2-82be-44a0-892f-f770a2fb8041.154151.1/ks-post.log: non-complete operation 2017-02-21 13:12:48.963824 7f64997fa700 10 read_permissions on buc66(@[]) ret=-2002 2017-02-21 13:12:48.963829 7f64997fa700 20 op->ERRORHANDLER: err_no=-2002 new_err_no=-2002 2017-02-21 13:12:48.963883 7f64997fa700 2 req 1576355:0.019130:s3:GET /buc66/ks-post.log:get_obj:op status=0 2017-02-21 13:12:48.963889 7f64997fa700 2 req 1576355:0.019136:s3:GET /buc66/ks-post.log:get_obj:http status=404 2017-02-21 13:12:48.963893 7f64997fa700 1 ====== req done req=0x7f64997f4710 op status=0 http_status=404 ====== 2017-02-21 13:12:48.963902 7f64997fa700 20 process_request() returned -2002 It looks like a blocker to me with NFS-RGW being GA.
Casey, could you help me triage this? Thanks, Matt
Correcting my statement in the description. The buckets are failing to sync irrespective of NFS directory. Even plain S3/swift bucket creation fails. The buckets that have failed to sync are: bu66 bu67 buc66 buc67 bu68
*** This bug has been marked as a duplicate of bug 1424725 ***