Home | About | Sematext search-lucene.com search-hadoop.com
 Search Hadoop and all its subprojects:

Switch to Threaded View
HBase >> mail # user >> Regions not getting reassigned if RS is brought down


Copy link to this message
-
Re: Regions not getting reassigned if RS is brought down
Filed https://issues.apache.org/jira/browse/HBASE-4109

On Fri, Jul 15, 2011 at 1:53 PM, Stack <[EMAIL PROTECTED]> wrote:

> Good on you lads.  Can we get a fix in for 0.90.4?
> St.Ack
>
> On Fri, Jul 15, 2011 at 1:02 PM, Shrijeet Paliwal
> <[EMAIL PROTECTED]> wrote:
> > So the problem is if you are using an interface anything other than
> > 'default' (literally that keyword) DNS.java 's getDefaultHost will return
> a
> > string which will
> > have a trailing period at the end. Now to me it seems javadoc of
> reverseDns
> > in DNS.java (see below) is conflicting with what that function is
> actually
> > doing.
> > It is returning a PTR record while claims it returns a hostname. The PTR
> > record always has period at the end , RFC:
> > http://irbs.net/bog-4.9.5/bog47.html
> >
> >  /**
> >   * Returns the hostname associated with the specified IP address by the
> >   * provided nameserver.
> >   *
> >   * @param hostIp
> >   *            The address to reverse lookup
> >   * @param ns
> >   *            The host name of a reachable DNS server
> > *   * @return The host name associated with the provided IP*
> >   * @throws NamingException
> >   *             If a NamingException is encountered
> >   */
> >  public static String reverseDns(InetAddress hostIp, String ns)
> >    throws NamingException {
> >    //
> >    // Builds the reverse IP lookup form
> >    // This is formed by reversing the IP numbers and appending
> in-addr.arpa
> >    //
> >    String[] parts = hostIp.getHostAddress().split("\\.");
> >    String reverseIP = parts[3] + "." + parts[2] + "." + parts[1] + "."
> >      + parts[0] + ".in-addr.arpa";
> >
> >    System.out.println("reverse ip is :" + reverseIP);
> >
> >    DirContext ictx = new InitialDirContext();
> >    Attributes attribute > >      ictx.getAttributes("dns://"               // Use "dns:///" if the
> > default
> >                         + ((ns == null) ? "" : ns) +
> >                         // nameserver is to be used
> >                         "/" + reverseIP, new String[] { "PTR" });
> >    ictx.close();
> >
> > *    return attribute.get("PTR").get().toString();*
> >  }
> >
> >
> > Related issue (I havent gone through it completely but glancing hints it
> is
> > related).
> > https://issues.apache.org/jira/browse/HBASE-2599 . Thanks Karthick for
> > pointing this out.
> >
> > A quicky is to recognize that default host has a trailing period and drop
> it
> > when we call it here:
> >  String machineName = DNS.getDefaultHost(conf.get(
> >        "hbase.regionserver.dns.interface", "default"), conf.get(
> >        "hbase.regionserver.dns.nameserver", "default"));
> >
> > I will open an issue shortly.  Thoughts?
> >
> > -Shrijeet
> > On Fri, Jul 15, 2011 at 10:25 AM, Stack <[EMAIL PROTECTED]> wrote:
> >
> >> Thanks for digging in Shrijeet.  We don't do this name matching well
> >> in 0.90.x  Sorry for pain caused.  on your observation below about
> >> RegionServerTracker, if you figure an improvement, that'd be great.
> >>
> >> Thanks,
> >> St.Ack
> >>
> >> On Thu, Jul 14, 2011 at 9:07 PM, Shrijeet Paliwal
> >> <[EMAIL PROTECTED]> wrote:
> >> > I have narrowed it down to following :
> >> >
> >> >  // Server to handle client requests
> >> >    String machineName = DNS.getDefaultHost(conf.get(
> >> >        "hbase.regionserver.dns.interface", "default"), conf.get(
> >> >        "hbase.regionserver.dns.nameserver", "default"));
> >> >
> >> > I am not using the default interface for RS. I have changed it to
> 'eth1'
> >> > . The machineName is getting set as 'server-2.rfiserve.net.'
> >> > Notice the extra period in the end.
> >> >
> >> > Because of above there is an inconsistency in the way zookeeper
> recorded
> >> the
> >> > regionserver address and way ServerManager had it in its cached list
> of
> >> > onlineservers.
> >> > You will notice the extra dot in zookeeper entry but not in the
> >> ServerManager
> >> > list.
> >> >
> >> > [zk: localhost:2181(CONNECTED) 3] ls /hbase/rs
> >>