Description of problem: ----------------------- While deleting all the directories created in the nfs share from one of the Client, IO errors were seen on clients and nfs daemon on rgw node had crashed. Version-Release number of selected component (if applicable): ------------------------- ceph-radosgw-10.2.5-22.el7cp.x86_64 nfs-ganesha-2.4.2-4.el7cp.x86_64 nfs-ganesha-rgw-2.4.2-4.el7cp.x86_64 Steps to Reproduce: ------------------- Had few directories created with huge data , upgraded to the latest builds by running yum update on all the cluster nodes and rgw clients, restarted the daemons running and initiated delete of all the directories created. nfs daemon crashed on rgw node.. Client :- ---------- # ls bucket1 bucket2 bucket3 bucketlist copy copynew dir1 dir2 dir3 dir4 dir5 dir6 dir7 folder1 folder3 [root@magna048 hell]# rm -rf * rm: cannot remove ‘dir1’: Directory not empty rm: cannot remove ‘dir6’: Input/output error rm: cannot remove ‘dir7’: Input/output error rm: cannot remove ‘folder1’: Input/output error rm: cannot remove ‘folder3’: Input/output error # ls ^C^C^C^C^Cls: cannot open directory .: Input/output error ----------------------------------------------------------------------- rgw Log -------- 0> 2017-02-07 12:43:52.810428 7f024cf31700 -1 *** Caught signal (Segmentation fault) ** in thread 7f024cf31700 thread_name:ganesha.nfsd ceph version 10.2.5-22.el7cp (5cec6848b914e87dd6178e559dedae8a37cc08a3) 1: (()+0x57322a) [0x7f04129d122a] 2: (()+0xf370) [0x7f041f226370] 3: (rgw::RGWFileHandle::reclaim()+0x1d7) [0x7f041296b477] 4: (cohort::lru::LRU<std::mutex>::evict_block()+0x102) [0x7f0412979092] 5: (rgw::RGWLibFS::lookup_fh(rgw::RGWFileHandle*, char const*, unsigned int)+0x393) [0x7f04129801e3] 6: (rgw::RGWLibFS::stat_leaf(rgw::RGWFileHandle*, char const*, unsigned int)+0xf51) [0x7f041296e891] 7: (rgw_lookup()+0xba) [0x7f041296fd6a] 8: (()+0x4bb7) [0x7f041c0bfbb7] 9: (()+0x6c16) [0x7f041c0c1c16] 10: (rgw::RGWReaddirRequest::send_response()+0x453) [0x7f041297da43] 11: (rgw::RGWLibProcess::process_request(rgw::RGWLibRequest*, rgw::RGWLibIO*)+0x637) [0x7f0412982ad7] 12: (rgw::RGWLibProcess::process_request(rgw::RGWLibRequest*)+0x29c) [0x7f0412983c8c] 13: (rgw::RGWFileHandle::readdir(bool (*)(char const*, void*, unsigned long), void*, unsigned long*, bool*, unsigned int)+0xb31) [0x7f041296ab21] 14: (()+0x4ab9) [0x7f041c0bfab9] 15: (mdcache_dirent_populate()+0x112) [0x7f0420d6b242] 16: (()+0x105621) [0x7f0420d61621] 17: (fsal_readdir()+0x15d) [0x7f0420c9a61d] 18: (nfs4_op_readdir()+0x26b) [0x7f0420cd535b] 19: (nfs4_Compound()+0x63d) [0x7f0420cc1ded] 20: (nfs_rpc_execute()+0x5bc) [0x7f0420cb2f9c] 21: (()+0x585fa) [0x7f0420cb45fa] 22: (()+0xe2289) [0x7f0420d3e289] 23: (()+0x7dc5) [0x7f041f21edc5] 24: (clone()+0x6d) [0x7f041e8ed73d] NOTE: a copy of the executable, or `objdump -rdS <executable>` is needed to interpret this.
No Crashed seen in latest builds while deleting the directories. Moving to Verified State.
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/RHBA-2017-0514.html