I've fixed the problem. The reason namenode won't start was that I accidentally started the cluster with root account. This somehow changed the ownership of some hadoop-related files(ex: log files, and hadoop.tmp.dir/dfs/name/current/edits) from hadoop:hadoop to root:root. After I fixed the ownership issue, everything went fine. Thanks for the concern.
2010/8/5 Harsh J <[email protected]> > Could you check the NameNode/SecondaryNameNode logs and try to find > the exact issue? Post the errors (if) it contains here, so we can try > to help you better. > > On Thu, Aug 5, 2010 at 9:39 AM, edward choi <[email protected]> wrote: > > I am currently stuck with hadoop namenode that won't start. > > > > When I type "start-all.sh", everything prints out fine. > > But when I type "jps", only JobTracker is activated. > > > > When this happens, I usually format the namenode. > > > > But the problem is that there are 500gigs of date in HDFS. > > So I really want to save the data. > > > > Can anyone give help please? > > > > > > -- > Harsh J > www.harshj.com >
