kadirozde commented on code in PR #5545:
URL: https://github.com/apache/hbase/pull/5545#discussion_r1442652833
##########
hbase-server/src/main/java/org/apache/hadoop/hbase/regionserver/HStoreFile.java:
##########
@@ -138,6 +140,12 @@ public class HStoreFile implements StoreFile {
// Indicates if the file got compacted
private volatile boolean compactedAway = false;
+ // Indicate if the file contains live cell versions for a given column
+ // in a row. MemStore flushes generate files with all cell versions. However,
+ // compactions can generate two files, one with the live cell versions and
the other
+ // with the remaining (historical) cell versions.
+ private volatile boolean hasLiveVersions = true;
Review Comment:
Even when `hbase.hstore.defaultengine.enable.dualfilewriter` is enabled, the
files generated by memstore flushes do not include the HAS_LIVE_VERSIONS HFile
metadata key since memstore does not use compaction writers. However, these
files will include live version cells. When the HAS_LIVE_VERSIONS HFile
metadata key is not included in the HFile, we still want to return true for
HStoreFile#hasLiveVersion(). That is why the default is true. I will add more
comments for this.
--
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.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
