Description
Build : 5.5.0-2928
A 2-node 5.0.1 cluster with some rbac users was upgrade to 5.5.0-2928 using swap rebalance method. Post upgrade, the users were intact, but seeing the following error in the error.log and diag logs:
[user:error,2018-06-22T10:49:11.859Z,ns_1@10.112.180.103:<0.17374.0>:menelaus_users:upgrade_to_55:695]Unsuccessful user storage upgrade.
{exit,{shutdown,{gen_server,call,[users_storage,
,infinity]}},
[{gen_server,call,3,[
,
{line,188}]},
{menelaus_users,do_upgrade_to_55,1,
[
{menelaus_users,upgrade_to_55,2,
[{file,"src/menelaus_users.erl"}
,
{line,692}]},
{cluster_compat_mode,do_upgrades,5,
[
{cluster_compat_mode,do_consider_switching_compat_mode,2,
[{file,"src/cluster_compat_mode.erl"}
,
{line,215}]},
{ns_orchestrator,consider_switching_compat_mode_dont_exit,0,
[
{ns_orchestrator,handle_rebalance_completion,2,
[{file,"src/ns_orchestrator.erl"}
,
{line,1069}]},
{gen_fsm,handle_msg,7,[
,
{line,505}]}]}
[ns_server:error,2018-06-22T10:49:11.862Z,ns_1@10.112.180.103:<0.17374.0>:cluster_compat_mode:do_consider_switching_compat_mode:220]Refusing to upgrade the compat version from [5,0] to [5,5] due to failure of rbac upgrade
Steps:
1. Create a 2-node cluster with 5.0.1
2. Install all sample buckets
3. Create 2 rbac users
4. Upgrade the cluster using swap rebalance method to 5.5.0
5. Post full upgrade, check if the users are intact
6. Check the diag logs.
The above error will be visible in the diag logs.