PHOENIX-6298 : Use timestamp of PENDING_DISABLE_COUNT to calculate elapse time for PENDING_DISABLE state#1060
PHOENIX-6298 : Use timestamp of PENDING_DISABLE_COUNT to calculate elapse time for PENDING_DISABLE state#1060virajjasani wants to merge 1 commit intoapache:masterfrom virajjasani:PHOENIX-6298-master
Conversation
|
💔 -1 overall
This message was automatically generated. |
gjacoby126
left a comment
There was a problem hiding this comment.
Thanks, @virajjasani . Change looks good, but I think we need a test.
I also want to point out (as I'll do on the JIRA) that this only affects the old indexing framework, because the new one doesn't auto-disable indexes. (It repairs write failures at read time.)
|
Thanks for the review @gjacoby126 . Added a test. |
|
💔 -1 overall
This message was automatically generated. |
| final String fullTableName = SchemaUtil.getTableName(schemaName, tableName); | ||
| final String fullIndexName = SchemaUtil.getTableName(schemaName, indexName); | ||
| final MyClock clock = new MyClock(1000); | ||
| EnvironmentEdgeManager.injectEdge(clock); |
There was a problem hiding this comment.
We've seen problems in the past with HBase 2.x and injecting an Environment Edge before creating a table. (It can hang forever) Probably want to move the edge injection until after the create table / create index statement unless you know the HBase 2.x edge bug has been fixed, and that it's fixed in all HBase 2.x versions we support (2.1, 2.2, 2.3 and 2.4)
There was a problem hiding this comment.
I am not sure if this edge bug is fixed in 2.x versions, so kept the injection after create table.
Thanks
|
💔 -1 overall
This message was automatically generated. |
|
💔 -1 overall
This message was automatically generated. |
…apse time for PENDING_DISABLE state
|
💔 -1 overall
This message was automatically generated. |
No description provided.