[ https://issues.apache.org/jira/browse/HBASE-3710?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Doug Meil updated HBASE-3710: ----------------------------- Attachment: book.xml.patch > Book.xml - fill out descriptions of metrics > ------------------------------------------- > > Key: HBASE-3710 > URL: https://issues.apache.org/jira/browse/HBASE-3710 > Project: HBase > Issue Type: Improvement > Reporter: Doug Meil > Assignee: Doug Meil > Priority: Minor > Attachments: book.xml.patch > > > I filled out the skeleton of the metrics in book.xml, but I'd like one of the > committers to fill in the rest of the details on what these mean. Thanks! :-) > For example.. > ------- > I'm assuming that these are referring to the LRU block cache in memory. I'd > like to docs to state this (for the sake of clarity) and also the units > (e.g., MB). > hbase.regionserver.blockCacheCount > hbase.regionserver.blockCacheFree > hbase.regionserver.blockCacheHitRatio > hbase.regionserver.blockCacheSize > ------- > This is read latency from HDFS, I assume... > hbase.regionserver.fsReadLatency_avg_time > hbase.regionserver.fsReadLatency_num_ops > hbase.regionserver.fsSyncLatency_avg_time > hbase.regionserver.fsSyncLatency_num_ops > hbase.regionserver.fsWriteLatency_avg_time > hbase.regionserver.fsWriteLatency_num_ops > ---------------- > point in time utilized (i.e., as opposed to max or trailing) memstore I > assume, would be nice to document. > hbase.regionserver.memstoreSizeMB > --------------- > obvious, but might as well document it > hbase.regionserver.regions > -------------- > This is any Put, Get, Delete, or Scan operation, I assume? > hbase.regionserver.requests > -------------- > detail on these would be nice, especially for tips on if there are any > critical numbers/ratios to watch for. > hbase.regionserver.storeFileIndexSizeMB > hbase.regionserver.stores -- This message is automatically generated by JIRA. For more information on JIRA, see: http://www.atlassian.com/software/jira