Description of problem: ----------------------- While flattening the cloned image from primary rbd host - rbd-mirror crashed on the secondary host. Version-Release number of selected component (if applicable): ------------------------------------------------------------- v10.2.2-1 Steps to Reproduce: 1. Create few images, create snapshots, protect them and create cloned images 2. write some data on the cloned images 3. flatten the cloned images ========== -2> 2016-06-16 17:14:26.402170 7f5b7a7fc700 20 rbd::mirror::image_replayer::ReplayStatusFormatter: 0x7f5ae426d680 handle_update_tag_cache: decoded tag 5: [mirror_uuid=, predecessor_mirro r_uuid=, predecessor_tag_tid=4, predecessor_entry_tid=1] -1> 2016-06-16 17:14:26.402178 7f5b7a7fc700 20 rbd::mirror::image_replayer::ReplayStatusFormatter: 0x7f5ae426d680 send_update_tag_cache: master_tag_tid=4, mirror_tag_tid=2 0> 2016-06-16 17:14:26.428690 7f5b7a7fc700 -1 *** Caught signal (Segmentation fault) ** in thread 7f5b7a7fc700 thread_name:fn_anonymous ceph version 10.2.2-1.el7cp (f1f313912893a3ecab6afbdc5690054dde9789fb) 1: (()+0x3a84da) [0x7f5bfd8304da] 2: (()+0xf100) [0x7f5bf2d5a100] 3: (journal::JournalMetadata::get_tag(unsigned long, cls::journal::Tag*, Context*)+0x3e) [0x7f5bfd7c9f9e] 4: (rbd::mirror::image_replayer::ReplayStatusFormatter<librbd::ImageCtx>::send_update_tag_cache(unsigned long, unsigned long)+0x131) [0x7f5bfd68d731] 5: (rbd::mirror::image_replayer::ReplayStatusFormatter<librbd::ImageCtx>::handle_update_tag_cache(unsigned long, unsigned long, int)+0x1f1) [0x7f5bfd68f8e1] 6: (FunctionContext::finish(int)+0x2a) [0x7f5bfd65b50a] 7: (Context::complete(int)+0x9) [0x7f5bfd659209] 8: (Context::complete(int)+0x9) [0x7f5bfd659209] 9: (()+0x33da17) [0x7f5bfd7c5a17] 10: (()+0x9d65d) [0x7f5bf3b8665d] 11: (()+0x85b49) [0x7f5bf3b6eb49] 12: (()+0x16f646) [0x7f5bf3c58646] 13: (()+0x7dc5) [0x7f5bf2d52dc5] 14: (clone()+0x6d) [0x7f5bf1c38ced] NOTE: a copy of the executable, or `objdump -rdS <executable>` is needed to interpret this.
Upstream master PR: https://github.com/ceph/ceph/pull/9759
Flattened around 50 cloned images and there was no crash seen. Works in 10.2.2-5 build, 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-2016-1755.html