Here is mine btw: 21:30:44 10.101.7.1 j...@rdag1:~ $ hadoop fs -lsr /hbase/.META./1028785192/info/ | wc -l 1825
And, I think yes, most of the META queries are hitting cache: 2011-01-03 21:33:18,022 DEBUG org.apache.hadoop.hbase.io.hfile.LruBlockCache: Block cache LRU eviction started; Attempting to free 100.78 MB of total=850.16 MB 2011-01-03 21:33:18,042 DEBUG org.apache.hadoop.hbase.io.hfile.LruBlockCache: Block cache LRU eviction completed; freed=93.18 MB, total=756.98 MB, single=266.76 MB, multi=374.35 MB, memory=200.84 MB It would be usesful for regionserver log transactions to Meta region, so that we could track responses. -Jack On Mon, Jan 3, 2011 at 9:25 PM, Stack <[email protected]> wrote: > Jack: > > Meta table affinity? > > Do you see hbase client logging of its going to .META. Jack? If you > check any of the lookups, do seem sensible? > > St.Ack > > On Mon, Jan 3, 2011 at 7:32 PM, Jack Levin <[email protected]> wrote: >> I suspect I have the same issue. And I would like very much to have meta >> table affinity (having extra debug logs for clients talking to meta would be >> super nice too) >> >> -Jack >> >> >> On Jan 3, 2011, at 8:31 PM, Wayne <[email protected]> wrote: >> >>> We are finding that the node that is responsible for the .META. table is >>> going in GC storms causing the entire cluster to go AWOL until it recovers. >>> Isn't the master supposed to serve up the .META. table? Is it possible to >>> Pin this table somewhere that only handles this? Our master server and >>> zookeeper servers are separate from our 10 region server nodes but in the >>> end one of the region servers is responsible for the .META. table and we >>> sometimes see all requests drop to zero except on the server handling the >>> .META. table and the requests jump up to the number of regions+1 and back >>> down. This has lasted for as long as 5 minutes before the cluster goes back >>> to responding to requests normally. When we had a 1GB region size with LZO >>> it was 90% in this AWOL state. >>> >>> Do we have our cluster set up correctly? Is it supposed to behave like this? >>> >>> >>> Thanks for any advice that can be provided. >> >
