HBASE-8755 introduced new write thread model. It is integrated in the recently released 0.98.0
You can consider giving 0.98.0 a spin. FYI On Feb 26, 2014, at 1:03 AM, shapoor <[email protected]> wrote: > I have a task which comes with a lot of requests. I had the same issue with > 0.94. I managed to solve it with letting my requests waiting longer if the > traffic is high and they have to wait (time_wait expansion). I put the > following properties in hbase-site.xml: > > <property> > <name>hbase.rpc.timeout</name> > <value>120000</value> > <description> > </description> > </property> > > <property> > <name>hbase.regionserver.lease.period</name> > <value>120000</value> > <description> > </description> > </property> > > Now I changed from 0.96.1.1-hadoop2 to 0.94 again and it didn't throw any > exceptions with 0.94 as before. But I guess this setting doesnt work for > 0.96.1.1-hadoop2 as "Nicolas Liochon" sujested in this post. > For the log, I have to switch back to 0.96.1.1-hadoop2 again and run the > task which I will do in the following days and let you know about that. > > thx and regards, > > > > -- > View this message in context: > http://apache-hbase.679495.n3.nabble.com/RegionTooBusyException-Above-memstore-limit-tp4056339p4056395.html > Sent from the HBase User mailing list archive at Nabble.com.
