zookeeper-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Scott Lindner <scott.a.lind...@gmail.com>
Subject Re: Zookeeper cluster failing
Date Tue, 06 Mar 2012 18:13:04 GMT
Hi Ted,

Sorry about that - I should have been more specific.  When I said "taking 2
down" I actually meant taking the entire cluster down and reconfiguring it
to only have a single server.  So to put it another way, when it is
actually set up as a cluster the failures occurred however when it was
configured to run with only a single server it worked fine.  Also when it
was configured to only include the 2 servers that were in sync that also
worked fine.  However any configurations involving the out of sync server
failed to work properly.

Any other ideas / suggestions?

Thanks again,

On Tue, Mar 6, 2012 at 12:49 PM, Ted Dunning <ted.dunning@gmail.com> wrote:

> OK.  Trying again against the correct thread.
> I trimmed down your summary to something that jumped out at me.  If you
> have three ZK servers in a cluster, then taking 2 down should not allow
> normal operation.
> I think you have a config error.
> On Tue, Mar 6, 2012 at 8:15 AM, Scott Lindner <lindner.scott@gmail.com
> >wrote:
> > ...
> > If we took down 2 of the 3 servers in the cluster all clients could
> connect
> > and everything would function fine, but as soon as we introduce an
> > additional server to the cluster client connections would no longer work
> > and the errors described above would repeat.  Also, no configuration that
> > involved the out of sync server would allow the servers to properly
> > communicate to one another, but if we only involved the 2 servers that
> were
> > in sync then communication between the servers seemed to work without
> > error.
> > ...

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