both namenode started but now as active stand by

[#1]

hi all

we have ambari cluster , HDP version 2.6.5

since we have problem with the namenodes

we decided to copy folder - /var/hadoop/hdfs/namenode/current from the good namenode

to the bad name node

and then restart the journal node and the namenode

after restart both namenode started bu not as active/standby

what is the additional configuration that we need to perform in order to get actiive/standby namenode?

from both namenode logs , we see the following

2020-01-28 06:16:14,137 INFO namenode.FSEditLogLoader (FSEditLogLoader.java:loadEditRecords(266)) - replaying edit log: 868173/2048342 transactions completed. (42%)
2020-01-28 06:16:15,138 INFO namenode.FSEditLogLoader (FSEditLogLoader.java:loadEditRecords(266)) - replaying edit log: 1002581/2048342 transactions completed. (49%)
2020-01-28 06:16:16,139 INFO namenode.FSEditLogLoader (FSEditLogLoader.java:loadEditRecords(266)) - replaying edit log: 1142328/2048342 transactions completed. (56%)
2020-01-28 06:16:17,140 INFO namenode.FSEditLogLoader (FSEditLogLoader.java:loadEditRecords(266)) - replaying edit log: 1301059/2048342 transactions completed. (64%)
2020-01-28 06:16:18,141 INFO namenode.FSEditLogLoader (FSEditLogLoader.java:loadEditRecords(266)) - replaying edit log: 1463751/2048342 transactions completed. (71%)
2020-01-28 06:16:19,142 INFO namenode.FSEditLogLoader (FSEditLogLoader.java:loadEditRecords(266)) - replaying edit log: 1637472/2048342 transactions completed. (80%)

[#2]

@mike_bronson7 

Can you please check the logs for any ERROR / WARNINGs ?

Usually after some time when the NameNode comes out of SafeMode then we should see the Active/StandBy status.

So can you please confirm if you still see the same behaviour even after some time?

Was it showing fine earlier?   Do you have the FQDN of namenode hosts in Mixed Case?

[#3]

Dear Jay

on both namenode we see the following

2020-01-28 06:17:19,950 INFO namenode.FSEditLogLoader (FSEditLogLoader.java:loadEditRecords(266)) - replaying edit log: 772706/2404698 transactions completed. (32%)
2020-01-28 06:17:20,951 INFO namenode.FSEditLogLoader (FSEditLogLoader.java:loadEditRecords(266)) - replaying edit log: 908585/2404698 transactions completed. (38%)
2020-01-28 06:17:21,952 INFO namenode.FSEditLogLoader (FSEditLogLoader.java:loadEditRecords(266)) - replaying edit log: 1060473/2404698 transactions completed. (44%)
2020-01-28 06:17:22,953 INFO namenode.FSEditLogLoader (FSEditLogLoader.java:loadEditRecords(266)) - replaying edit log: 1193701/2404698 transactions completed. (50%)

[#4]

about dns verification

yes we checked the dns names nad they are FQDN ,

just to sync you , the cluster filed due electricity failure  ( cluster was working fine before electricity failure )

so after we copy the cuurent folder from good namenode to bad namenode

still both namenode not became active/standby