Details
-
Bug
-
Resolution: Cannot Reproduce
-
Critical
-
7.2.0
-
6.5.1-6296 --> 7.2.0-5325
-
Untriaged
-
Centos 64-bit
-
-
0
-
Unknown
Description
Scenario:
- 4 node cluster, 1 coushstore bucket
- Graceful failover nodes + perform offline upgrade one node at a time
- During the final node upgrade (172.23.104.98), seeing rebalance failure with reason,
Rebalance exited with reason {service_rebalance_failed,index,{agent_died,<20089.29558.1>,{linked_process_died,<20089.29559.1>,{'ns_1@172.23.105.123',{no_connection,"index-service_api"}}}}}.Rebalance Operation Id = f08c91e2cdf57b6b931ebe8fb4ba7312
Testrunner test:
./testrunner -i node.ini -p upgrade_version=7.2.0-5325,stop-on-failure=False,skip_cleanup=False,skip_init_check_cbserver=true,groups=simple,gsi_type=memory_optimized,upgrade_test=True,dgm_run=True,skip_log_scan=true,upgrade_version=7.2.0-5325 -t newupgradetests.MultiNodesUpgradeTests.test_offline_upgrade_with_add_new_services,initial_version=6.5.1-6296,nodes_init=4,ddocs_num=3,pre_upgrade=create_replica_index,doc-per-day=1,dataset=default,after_upgrade_buckets_flush=True,groups=simple,initial-services-setting=kv-kv,n1ql,index-kv,index,after_upgrade_services_in=index,num_index_replica=2,disable_HTP=True,offline_failover_upgrade=True,upgrade_test=True,dgm_run=True,init_nodes=False
|
|
