Hi Ralph, I am not sure if the problem is Phoenix specific. Looks to me that it might be related to https://issues.apache.org/jira/browse/HBASE-11295.
However, this setting value seems to be too large. If the default 60000 ms for "hbase.rpc.timeout" is not working, you might want to multiply the value by 2 or 4 and see if that will help. And it actually not matter what physical location you put your configuration file in, but you have to make sure that the file is included in your Phoeinx client classpath. Thanks, Maryann On Wed, Nov 19, 2014 at 12:18 PM, Perko, Ralph J <[email protected]> wrote: > Hi, > > When using the subquery feature I continually get, what appear to be, RPC > timeout errors on the larger queries. Queries that only return a few > results do great, returning in just a few seconds, but queries that I > expect to return 3 or 4K results are timing out. > > > > Query: > > select * from TABLE1 where (file_id,recnum) in (select file_id, recnum > from TABLE1 where ipaddress in(ip1,ip2,ip3,ip4)); > > > > Error: PhoenixIOException: Failed after retry of > OutOfOrderScannerNextException: was there a rpc timeout? > > > > I have tried setting both: > > > > hbase.rpc.timeout=3600000 > > phoenix.query.timeoutMs=3600000 > > > > Both on the server and client and the settings aren’t sticking > > > > Questions: > > > > 1) Are these the correct settings? > > 2) On the client side I am using <phoenix-home>/bin/hbase-site.xml > - is that correct? > > > > Hbase 0.98 > > Phoenix 4.2 > > 7B recs on 30 nodes > > Using both sqlline.py and psql.py > > > > Thank you for your help, > > Ralph > > > -- Thanks, Maryann
