In my Ambari cluster setup, in HBASE I see the Files Local percentage is fluctuating. Though the cluster is not busy at all and am not doing any bulk load or any such heavy operation.
What are the things I need to verify and make changes to keep it stable and above 70%?
Do you mean the file locality can be lost at any point in time, no need to worry about it? Or I need to configure and schedule for the major compaction or memstore flushes.
Is it not going to impact the live queries if I leave it just like that? At present the configs are default configs came with Amabri installations, I haven't modified anything.
May be am asking too many qs, since this is my first cluster, about to go live in prod, so little worried.
+1 to what you said, Sunile.
Rudra: locality may fluctuate, but, without regionserver failure or region splitting, you should see a high percentage of locality. This is the general statement, but there are some exceptions. Unless you are correlating a realized negative performance with low locality, you are likely chasing your tail :).
These are the some noticeable picks from the RS log:
[MemStoreFlusher.1] regionserver.HRegion: Started memstore flush for SYSTEM.STATS,,1471982464802.1ff0385b05d60535d5217e2402753d6d., current region memstore size 1.88 KB, and 1/1 column families' memstores are being flushed. regionserver.HRegionServer: myregionserver.com,16020,1477683211158-MemstoreFlusherChore requesting flush for region SYSTEM.STATS,,1471982464802.1ff0385b05d60535d5217e2402753d6d. after a delay of 17160 [regionserver/myregionserver.com/<IPADDR>:16020-shortCompactions-1477693224761] regionserver.HRegion: Starting compaction on 0 in region SYSTEM.STATS,,1471982464802.1ff0385b05d60535d5217e2402753d6d.