Red Hat Bugzilla – Bug 1244415
Enabling management SSL on a gluster cluster already configured can crash glusterd
Last modified: 2016-09-17 10:39:23 EDT
Description of problem: On a gluster storage which has already nodes configured with volumes and mounted on clients, enabling management SSL crashes glusterd. Backtrace of the core is same as 1243722 Version-Release number of selected component (if applicable): glusterfs-3.7.1-10.el7rhgs.x86_64 How reproducible: Always Steps to Reproduce: 1. Have few gluster volumes already on the system 2. stop glusterd service 3. enable management ssl. touch /var/lib/glusterd/secure-access 4. start glusterd service Actual results: glusterd crashes. Expected results: glusterd shouldn't crash. Management SSL should be enabled. Additional info: Backtrace from the core: #0 list_del (old=0x7f7bc00024c8) at ../../../../libglusterfs/src/list.h:76 #1 glusterd_rpcsvc_notify (rpc=<optimized out>, xl=0x7f7be5ee40f0, event=<optimized out>, data=0x7f7bc0001a10) at glusterd.c:347 #2 0x00007f7be428b35f in rpcsvc_handle_disconnect (svc=0x7f7be5eef110, trans=trans@entry=0x7f7bc0001a10) at rpcsvc.c:754 #3 0x00007f7be428d718 in rpcsvc_notify (trans=0x7f7bc0001a10, mydata=<optimized out>, event=<optimized out>, data=0x7f7bc0001a10) at rpcsvc.c:792 #4 0x00007f7be428f873 in rpc_transport_notify (this=this@entry=0x7f7bc0001a10, event=event@entry=RPC_TRANSPORT_DISCONNECT, data=data@entry=0x7f7bc0001a10) at rpc-transport.c:543 #5 0x00007f7bd6d5de64 in socket_poller (ctx=0x7f7bc0001a10) at socket.c:2582 #6 0x00007f7be332ddf5 in start_thread () from /lib64/libpthread.so.0 #7 0x00007f7be2c741ad in clone () from /lib64/libc.so.6
(In reply to krishnaram Karthick from comment #0) > Description of problem: > On a gluster storage which has already nodes configured with volumes and > mounted on clients, enabling management SSL crashes glusterd. Backtrace of > the core is same as 1243722 > > Version-Release number of selected component (if applicable): > glusterfs-3.7.1-10.el7rhgs.x86_64 > > How reproducible: > Always > > Steps to Reproduce: > 1. Have few gluster volumes already on the system > 2. stop glusterd service The documentation [1] mentions that all Gluster processes need to be stopped. Not just GlusterD. Stopping just GlusterD is not enough in this case. Though the crash is a bug, if the steps were correctly followed it wouldn't have happened. I'd recommend closing this bug as it is invalid. [1]: http://jenkinscat.gsslab.pnq.redhat.com:8080/view/Gluster/job/doc-Red_Hat_Gluster_Storage-3.1-Administration_Guide%20(html-single)/136/artifact/tmp/en-US/html-single/index.html#idm140449326918528 > 3. enable management ssl. touch /var/lib/glusterd/secure-access > 4. start glusterd service > > > Actual results: > glusterd crashes. > > Expected results: > glusterd shouldn't crash. Management SSL should be enabled. > > Additional info: > > Backtrace from the core: > > #0 list_del (old=0x7f7bc00024c8) at ../../../../libglusterfs/src/list.h:76 > #1 glusterd_rpcsvc_notify (rpc=<optimized out>, xl=0x7f7be5ee40f0, > event=<optimized out>, data=0x7f7bc0001a10) > at glusterd.c:347 > #2 0x00007f7be428b35f in rpcsvc_handle_disconnect (svc=0x7f7be5eef110, > trans=trans@entry=0x7f7bc0001a10) > at rpcsvc.c:754 > #3 0x00007f7be428d718 in rpcsvc_notify (trans=0x7f7bc0001a10, > mydata=<optimized out>, event=<optimized out>, > data=0x7f7bc0001a10) at rpcsvc.c:792 > #4 0x00007f7be428f873 in rpc_transport_notify > (this=this@entry=0x7f7bc0001a10, > event=event@entry=RPC_TRANSPORT_DISCONNECT, > data=data@entry=0x7f7bc0001a10) at rpc-transport.c:543 > #5 0x00007f7bd6d5de64 in socket_poller (ctx=0x7f7bc0001a10) at socket.c:2582 > #6 0x00007f7be332ddf5 in start_thread () from /lib64/libpthread.so.0 > #7 0x00007f7be2c741ad in clone () from /lib64/libc.so.6
downstream patch https://code.engineering.redhat.com/gerrit/#/c/53141/ merged now.
Crash is no more seen when enabling management SSL on an existing setup. Steps followed to verify are, 1. Have few gluster volumes already on the system 2. stop glusterd service 3. enable management ssl. touch /var/lib/glusterd/secure-access 4. start glusterd service No gluster crash is seen. Mount succeeds when proper steps are followed as per documentation. Marking the bug as verified. Build used to verify: glusterfs-3.7.1-14.el7rhgs.x86_64
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-2015-1845.html