FIO 3.12 (2ce58465415fc4d900c4dd89b86acbcaa51d9dfb) here. I'm trying to get FIO w/ RBD engine running on a working CEPH cluster. The error I'm getting is: [cephuser@sm10 ~]$ sudo /usr/local/bin/fio ./rbd.fio rbd_iodepth32: (g=0): rw=randwrite, bs=(R) 4096B-4096B, (W) 4096B-4096B, (T) 4096B-4096B, ioengine=rbd, iodepth=32 fio-3.12 Starting 1 process rados_connect failed. fio_rbd_connect failed. Run status group 0 (all jobs): The job file is a slightly modified example/rbd.fio ----- [global] invalidate=0 ioengine=rbd clientname=cephuser clustername=ceph pool=rbd rbdname=sm10_v0 rw=randwrite bs=4k [rbd_iodepth32] iodepth=32 ------- I'm able to run 'ceph status' from the client node (sm10) where FIO fails either as root or cephuser. The pool rbd does exist: [cephuser@sm10 ~]$ ceph osd lspools 4 rbd The RBD image does exist in that pool: [cephuser@sm10 ~]$ rbd ls -p rbd | grep sm10_v0 sm10_v0 fio --debug=all isn't helpful. Google turned up some results where people changed client name from username to client.username which didn't help in my case. Any pointers how to troubleshoot? -- Michael Green mishagreen@xxxxxxxxx
