ajfabbri commented on a change in pull request #802: HADOOP-16279. S3Guard: 
Implement time-based (TTL) expiry for entries …
URL: https://github.com/apache/hadoop/pull/802#discussion_r282261248
 
 

 ##########
 File path: 
hadoop-common-project/hadoop-common/src/main/resources/core-default.xml
 ##########
 @@ -1502,11 +1502,11 @@
 </property>
 
 <property>
-    <name>fs.s3a.metadatastore.authoritative.dir.ttl</name>
+    <name>fs.s3a.metadatastore.metadata.ttl</name>
 
 Review comment:
   Two things: 1. need to make sure this is ok with compatibility rules 
(changing public API, essentially.. not sure which releases have happened--if 
any--since this was added).  2. I think we want to separate (a) "is 
authoritative" from (b) "does metadata exist" in metadata store.  I think ideal 
would be having something like ```fs.s3a.metadatastore.authoritative.ttl`` 
which says how long S3A treats MS data as fresh enough to skip S3 query, and 
then some other parameter (existing "prune age" may be sufficient) saying when 
metadata should be deleted. I'll start a bigger discussion on the JIRA.

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org


With regards,
Apache Git Services

---------------------------------------------------------------------
To unsubscribe, e-mail: common-issues-unsubscr...@hadoop.apache.org
For additional commands, e-mail: common-issues-h...@hadoop.apache.org

Reply via email to