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

Switch to Threaded View
Accumulo >> mail # user >> Iterator name already in use with AccumuloInputFormat?


Copy link to this message
-
Re: Iterator name already in use with AccumuloInputFormat?
To summarize the issue: whenever you add a RegExFilter using the
recommended method AccumuloInputFormat.addIterator in 1.4, and you haven't
explicitly specified a name for your iterator, an error is thrown saying
the name is in use.  This is because the input format always sets a
RegExFilter to handle the deprecated method of setting regex on the input
format, even if that method isn't used.  See ACCUMULO-1267.  This issue can
be avoided by specifying a name for the iterator when creating the
IteratorSetting.
On Thu, Apr 11, 2013 at 11:13 AM, Keith Turner <[EMAIL PROTECTED]> wrote:

> On Thu, Apr 11, 2013 at 2:10 PM, Chris Sigman <[EMAIL PROTECTED]> wrote:
> > I've just run a job for the second time where I've called addIterator
> with a
> > RegExFilter, and it's saying that the filter name is already in use.
>  When I
> > try to scan using the shell though, the iterator's not there... what's
> going
> > on?
>
> Are you using the same java process to start two map reduce jobs when
> you see this?
>
> >
> > Thanks,
> > --
> > Chris
>