Ok. It works with branch-1 but not with 0.99.0. Just tried the 2. JMS
2014-09-23 21:10 GMT-04:00 Jean-Marc Spaggiari <[email protected]>: > I tried with 0.99 I will retry with branch-1... > > 2014-09-23 21:05 GMT-04:00 Ted Yu <[email protected]>: > > J-M: >> Clearing out /tmp did the trick. >> >> I tried with tip of branch-1 and I didn't see the problem you mentioned in >> your first email. >> >> Cheers >> >> On Tue, Sep 23, 2014 at 5:58 PM, Jean-Marc Spaggiari < >> [email protected]> wrote: >> >> > So what is the solution to start in pseudo dist? Or is that only >> impacting >> > 0.99.0 and I should try with 0.99.x? >> > >> > @Ted: No, haven't see that. Make sure your /tmp folder has been >> cleared... >> > >> > 2014-09-23 20:52 GMT-04:00 Enis Söztutar <[email protected]>: >> > >> > > We've discussed this in >> > https://issues.apache.org/jira/browse/HBASE-11575. >> > > It seems that the master will not know whether it is an actual >> > distributed >> > > setup or not. >> > > >> > > Enis >> > > >> > > On Tue, Sep 23, 2014 at 5:41 PM, Jean-Marc Spaggiari < >> > > [email protected]> wrote: >> > > >> > > > Thanks for the hint. I will do that. >> > > > >> > > > Should we not set this value to 1 by default if it blocks the >> startup >> > > > process? >> > > > >> > > > 2014-09-23 20:37 GMT-04:00 Ted Yu <[email protected]>: >> > > > >> > > > > Please set hbase.master.wait.on.regionservers.mintostart to 1 >> > > > > >> > > > > Cheers >> > > > > >> > > > > On Tue, Sep 23, 2014 at 5:32 PM, Jean-Marc Spaggiari < >> > > > > [email protected]> wrote: >> > > > > >> > > > > > Hi, >> > > > > > >> > > > > > Trying to run HBase 0.99.0 in pseudo dist, I get this: >> > > > > > 2014-09-23 20:31:25,283 INFO [ActiveMasterManager] >> > > > master.ServerManager: >> > > > > > Waiting for region servers count to settle; currently checked >> in 1, >> > > > slept >> > > > > > for 57227 ms, expecting minimum of 2, maximum of 2147483647, >> > timeout >> > > of >> > > > > > 4500 ms, interval of 1500 ms, selfCheckedIn true >> > > > > > >> > > > > > However, seems that we are only starting a single RS and master >> > never >> > > > > > complete initialization. >> > > > > > >> > > > > > I think it's something we should look at. >> > > > > > >> > > > > > JM >> > > > > > >> > > > > >> > > > >> > > >> > >> > >
