[Bugs] [Bug 1702316] Cannot upgrade 5.x volume to 6.1 because of unused 'crypt' and 'bd' xlators

[bugzilla at redhat.com]

https://bugzilla.redhat.com/show_bug.cgi?id=1702316

Atin Mukherjee <amukherj at redhat.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |amukherj at redhat.com
              Flags|                            |needinfo?(rob.dewit at coosto.
                   |                            |com)



--- Comment #2 from Atin Mukherjee <amukherj at redhat.com> ---
I don't think the upgrade failure or the geo-replication session issue is due
to the missing xlators what you highlighted in the report.

If you notice the following log snippet, the cleanup_and_exit which is a
shutdown trigger of glusterd happened much later than the logs which complaint
about the missing xlators and I can confirm that they are benign.

[2019-04-23 12:38:25.514866] W [MSGID: 101095]
[xlator.c:210:xlator_volopt_dynload] 0-xlator:
/usr/lib64/glusterfs/6.1/xlator/encryption/crypt.so: cannot open shared object
file: No such file or directory
[2019-04-23 12:38:25.522473] E [MSGID: 101097]
[xlator.c:218:xlator_volopt_dynload] 0-xlator: dlsym(xlator_api) missing:
/usr/lib64/glusterfs/6.1/rpc-transport/socket.so: undefined symbol: xlator_api
[2019-04-23 12:38:25.555952] W [MSGID: 101095]
[xlator.c:210:xlator_volopt_dynload] 0-xlator:
/usr/lib64/glusterfs/6.1/xlator/storage/bd.so: cannot open shared object file:
No such file or directory
The message "W [MSGID: 101095] [xlator.c:210:xlator_volopt_dynload] 0-xlator:
/usr/lib64/glusterfs/6.1/xlator/encryption/crypt.so: cannot open shared object
file: No such file or directory" repeated 2 times between [2019-04-23
12:38:25.514866] and [2019-04-23 12:38:25.514931]
The message "E [MSGID: 101097] [xlator.c:218:xlator_volopt_dynload] 0-xlator:
dlsym(xlator_api) missing: /usr/lib64/glusterfs/6.1/rpc-transport/socket.so:
undefined symbol: xlator_api" repeated 7 times between [2019-04-23
12:38:25.522473] and [2019-04-23 12:38:25.522545]

################################# There's a gap of ~14 minutes here
###################################################

[2019-04-23 12:52:00.569988] W [glusterfsd.c:1570:cleanup_and_exit]
(-->/lib64/libpthread.so.0(+0x7504) [0x7fb0f1310504]
-->/usr/sbin/glusterd(glusterfs_sigwaiter+0xd5) [0x409f45]
-->/usr/sbin/glusterd(cleanup_and_exit+0x57) [0x409db7] ) 0-: received signum
(15), shutting down

You'd need to provide us the brick logs along with glusterd logs, gluster
volume status and gluster get-state output from the node where you see this
happening.

Related to geo-rep failures, I'd suggest you to file a different bug once this
stabilises.

-- 
You are receiving this mail because:
You are on the CC list for the bug.
You are the assignee for the bug.