accumulo-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Sean Busbey <bus...@cloudera.com>
Subject Re: Monitor finding wrong hostname
Date Wed, 09 Jul 2014 16:06:00 GMT
Excellent!

Did this change also fix your gc problem?


On Wed, Jul 9, 2014 at 11:02 AM, David Medinets <david.medinets@gmail.com>
wrote:

> It's working! I have four single-node accumulo instances running on my
> laptop. I can visit the monitor page for each one. Changes in one instance
> are isolated from the others. It takes less than 250 milliseconds to start
> a new instance.
>
> david@zareason-verix545:~/projects/docker-builds/accumulo$ time
> ./make_container.sh paul paul
> d317a83cb91abc1e908d70ef05b54934b7ff89fb6920b53b62f5432305715fef
>
> real    0m0.249s
> user    0m0.022s
> sys    0m0.019s
>
>
>
> On Wed, Jul 9, 2014 at 11:49 AM, David Medinets <david.medinets@gmail.com>
> wrote:
>
> > Doh! Perhaps the problem is my supervisor configuration:
> >
> > command=/usr/lib/accumulo/bin/accumulo master --address namenode
> >
> >
> >
> > On Wed, Jul 9, 2014 at 11:46 AM, David Medinets <
> david.medinets@gmail.com>
> > wrote:
> >
> >> Thanks. After updating the accumulo-env.sh the process is pulling from
> >> IPV4. But I have no idea why the hostname is being reported as
> 'namenode'
> >> instead of 'grail'.
> >>
> >>
> >> java.lang.reflect.InvocationTargetException
> >>     at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
> >>     at
> >>
> sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:57)
> >>     at
> >>
> sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)
> >>     at java.lang.reflect.Method.invoke(Method.java:606)
> >>     at org.apache.accumulo.start.Main$1.run(Main.java:103)
> >>     at java.lang.Thread.run(Thread.java:744)
> >> Caused by: java.net.UnknownHostException: namenode
> >>      at java.net.Inet4AddressImpl.lookupAllHostAddr(Native Method)
> >>
> >>     at java.net.InetAddress$1.lookupAllHostAddr(InetAddress.java:901)
> >>     at
> >> java.net.InetAddress.getAddressesFromNameService(InetAddress.java:1293)
> >>     at java.net.InetAddress.getAllByName0(InetAddress.java:1246)
> >>     at java.net.InetAddress.getAllByName(InetAddress.java:1162)
> >>
> >> I've updated the github project with my latest attempt.
> >>
> >>
> >> On Wed, Jul 9, 2014 at 11:21 AM, Josh Elser <josh.elser@gmail.com>
> wrote:
> >>
> >>> Actually, we do, and he isn't :)
> >>>
> >>> https://github.com/medined/docker-builds/blob/master/
> >>> accumulo/accumulo.conf/accumulo-env.sh
> >>>
> >>> David, update your accumulo-env.sh. You also have cruft in there that
> is
> >>> no longer necessary (e.g. ACCUMULO_LOG_HOST)
> >>>
> >>>
> >>> On 7/9/14, 11:18 AM, Sean Busbey wrote:
> >>>
> >>>> Josh and I are talking about this config at the end of hte
> >>>> ACCUMULO_GENERAL_OPTS in the example accumulo-env.sh[1]:
> >>>>
> >>>> -Djava.net.preferIPv4Stack=true
> >>>>
> >>>>
> >>>> [1]:
> >>>> https://github.com/apache/accumulo/blob/1.5.1/conf/
> >>>> examples/512MB/native-standalone/accumulo-env.sh#L52
> >>>>
> >>>>
> >>>> On Wed, Jul 9, 2014 at 10:16 AM, Sean Busbey <busbey@cloudera.com>
> >>>> wrote:
> >>>>
> >>>>  Sure, but we don't know if David's current configs include everything
> >>>>> from
> >>>>> the examples.
> >>>>>
> >>>>>
> >>>>> On Wed, Jul 9, 2014 at 10:11 AM, Josh Elser <josh.elser@gmail.com>
> >>>>> wrote:
> >>>>>
> >>>>>  I think I had made the change before 1.5.1 came out which updates
> the
> >>>>>> example configs to set the JVM preferIpv4Stack (or w/e that
option
> >>>>>> is), too.
> >>>>>>
> >>>>>>
> >>>>>> On 7/9/14, 10:37 AM, Sean Busbey wrote:
> >>>>>>
> >>>>>>  The stacktrace shows it's trying to use IPv6.
> >>>>>>>
> >>>>>>> Have you disabled ipv6 on the host? Do your Accumulo role
options
> >>>>>>> include
> >>>>>>> the flag that says to use the ipv4 network stack?
> >>>>>>>
> >>>>>>>
> >>>>>>>
> >>>>>
> >>>>> --
> >>>>> Sean
> >>>>>
> >>>>>
> >>>>
> >>>>
> >>>>
> >>
> >
>



-- 
Sean

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message