Description of problem: Cthon Lock test is failing on NFS-Ganesha V3 Version-Release number of selected component (if applicable): glusterfs-ganesha-3.8.4-12.el6rhs.x86_64 How reproducible: Consistently Steps to Reproduce: 1.Setup 4 node ganesha cluster on Rhel6 2.Create 4*2 Distributed-Replicate Volume.Enable ganesha on it. 3.Export the Volume to Client via V3 on Rhel7 Client 4.Run cthon lock test suite # cd /root/cthon04 && ./server -l -o vers=3 -p /replicaVol -m /mnt/rhel6/mountpoint/ -N 1 dhcp37-106.lab.eng.blr.redhat.com sh ./runtests -l -t /mnt/rhel6/mountpoint//dhcp37-192.test Starting LOCKING tests: test directory /mnt/rhel6/mountpoint//dhcp37-192.test (arg: -t) Testing native post-LFS locking Creating parent/child synchronization pipes. Test #1 - Test regions of an unlocked file. Parent: 1.1 - F_TEST [ 0, 1] PASSED. Parent: 1.2 - F_TEST [ 0, ENDING] PASSED. Parent: 1.3 - F_TEST [ 0,7fffffffffffffff] PASSED. Parent: 1.4 - F_TEST [ 1, 1] PASSED. Parent: 1.5 - F_TEST [ 1, ENDING] PASSED. Parent: 1.6 - F_TEST [ 1,7fffffffffffffff] PASSED. Parent: 1.7 - F_TEST [7fffffffffffffff, 1] PASSED. Parent: 1.8 - F_TEST [7fffffffffffffff, ENDING] PASSED. Parent: 1.9 - F_TEST [7fffffffffffffff,7fffffffffffffff] PASSED. Test #2 - Try to lock the whole file. Parent: 2.0 - F_TLOCK [ 0, ENDING] FAILED! Parent: **** Expected success, returned errno=37... Parent: **** Probably implementation error. ** PARENT pass 1 results: 9/9 pass, 0/0 warn, 1/1 fail (pass/total). ** CHILD pass 1 results: 0/0 pass, 0/0 warn, 0/0 fail (pass/total). lock tests failed Tests failed, leaving /mnt/rhel6/mountpoint/ mounted Actual results: Cthon lock test suit is failing with V3 and Passing with V4 Expected results: Cthon Lock test suit should pass with V3 Additional info:
Jan 18 19:35:17 dhcp47-74 rpc.statd[11770]: Failed to insert: creating /var/lib/nfs/statd/sm/dhcp37-192.lab.eng.blr.redhat.com: Permission denied rpc.statd[11770]: STAT_FAIL to dhcp47-74.lab.eng.blr.redhat.com for SM_MON of dhcp37-192.lab.eng.blr.redhat.com It seems to be Day-1 issue. statd process gets started as rpcuser. In the shared_storage nfs-ganesha folder, we need to chown the directories/files used by statd to 'rpcuser'. Will send in a fix upstream for review. The work-around is to either manually chown those files before setting up nfs-ganesha or restart statd process.
patch posted usptream for review - http://review.gluster.org/16433
Verified this bug on glusterfs-3.8.4-14.el7rhgs.x86_64 As Cthon lock test case is passing successfully with nfs-ganesha v3 mount.Hence marking this bug as verified
Hi Laura, Doc text looks good.. but probably there was typo at the beginning. Please find the corrections below - NLM Clients state could not be stored if the rpc.statd process started running prior to the NFS-Ganesha High Availability setup. The directory and its files where the state is maintained are now created with rpcuser as the owning user and group so that the rpc.statd process can access and maintain client state.
The updates look good to me. Thanks!
Since the problem described in this bug report should be resolved in a recent advisory, it has been closed with a resolution of ERRATA. For information on the advisory, and where to find the updated files, follow the link below. If the solution does not work for you, open a new bug report. https://rhn.redhat.com/errata/RHSA-2017-0486.html