So really, I don't think it's from you code. Looking at the stacktrace, seems that Gora is creating one connection per task, and your number of concurrent tasks might be bigger than 30.
Since tasks might run in different nodes, it's a big difficult to share the connection betweek the tasks. so I guess you 2 best options are to reduce the tasks to make sure you have less than 30 at a time, or to increase the ZK limit... JM 2013/11/12 Jean-Marc Spaggiari <[email protected]> > I look very quickly at your code and I have not seend anything wrong so > far. I will take a deeper look. > > For the number of maps/slots available, you can take a look at the > jobtracker interface http://youserver:50030/jobtracker.jsp > > I will take another look at your code now and keep you posted. > > > 2013/11/12 glumet <[email protected]> > >> My data are stored in hBase. And I am trying to put them from hBase to >> Solr. >> I have written my own mapper and reducer (I put it here if you are >> interested http://pastebin.com/jsKqiZTw). I have to admit that I do not >> know >> where I can check how >> many maps do I have at the same time and where I can change a value... is >> it >> possible that problem is in bad re-using of >> org.apache.hadoop.conf.Configuration in my code? >> >> >> >> -- >> View this message in context: >> http://apache-hbase.679495.n3.nabble.com/hBase-the-server-has-too-many-connections-maxClientConn-property-set-to-0-does-not-help-tp4052728p4052730.html >> Sent from the HBase User mailing list archive at Nabble.com. >> > >
