Given the following tables in a deployment :
IntegrationTestBulkLoad
IntegrationTestIngest
IntegrationTestIngestStripeCompactions
IntegrationTestLoadAndVerify
IntegrationTestMTTR
If I pass IntegrationTestIngest as parameter, regions
of IntegrationTestIngestStripeCompactions would be picked up.
Here is modified version of Samir's script that I tested on 0.98 deployment
(note the comma following table name):
#!/bin/bash
table="$1"
echo "scan 'hbase:meta', {FILTER=>\"PrefixFilter('$table,')\"}" > cmd_file
hbase shell < cmd_file | grep $table |awk '{print $1}' | cut -d "," -f3 |
cut -d "." -f2 |tail -n+2 | uniq
On Fri, Apr 25, 2014 at 7:05 AM, Samir Ahmic <[email protected]> wrote:
> Hi David,
>
> Here is quick bash script that will extract encoded region name.
>
> #!/bin/bash
> table="$1"
> echo "scan '.META.', {FILTER=>\"PrefixFilter('$table')\"}" > cmd_file
> hbase shell < cmd_file | grep $table |awk '{print $1}' | cut -d "," -f3 |
> cut -d "." -f2 |tail -n+2 | uniq
>
> Regards
>
>
>
>
> On Fri, Apr 25, 2014 at 3:21 PM, Ted Yu <[email protected]> wrote:
>
> > Minor comment about usage of the PrefixFilter, if there is table 't11' in
> > your cluster, rows for 't11' would also be returned.
> >
> > You can specify PrefixFilter('t1,')
> >
> > Cheers
> >
> >
> > On Fri, Apr 25, 2014 at 3:19 AM, David Koch <[email protected]>
> wrote:
> >
> > > Hi, cool - thanks, it works nicely - taking this even further. Is
> there a
> > > way to get only the "plain text" region name - for instance, doing a
> > plain
> > > scan results in something like:
> > >
> > > my_table,\xA0\x03\x93\xC7\x81aBb\x8Cr\x1D\xB0\x93\x9D\x
> > > column=info:regioninfo, timestamp=1398333985425, value={NAME =>
> > >
> > >
> >
> 'my_table,\xA0\x03\x93\xC7\x81aBb\x8Cr\x1D\xB0\x93\x9D\xBC\x0E,1392713217395.09326af68a5973257f8d88155
> > > BC\x0E,1392713217395.09326af68a5973257f8d881552f809cd. 2f809cd.',
> > > STARTKEY => '\xA0\x03\x93\xC7\x81aBb\x8Cr\x1D\xB0\x93\x9D\xBC\x0E',
> > ENDKEY
> > > => '\xBF\xB9\x92\xA4N\x11C\x8E\x9A$a\xEF\xB2\xB7R\xC6', ENCODED =>
> > > 09326af68a5973
> > >
> > > 257f8d881552f809cd,}
> > >
> > > and all I require is "09326af68a5973257f8d881552f809cd" - which I
> believe
> > > is the HFile name. I currently extract this using some ugly stuff on
> the
> > > command line on the hbase shell output.
> > >
> > > Thanks,
> > >
> > > /David
> > >
> > >
> > >
> > > On Fri, Apr 25, 2014 at 12:05 PM, Bharath Vissapragada <
> > > [email protected]> wrote:
> > >
> > > > You can use a prefix filter something like as follows
> > > >
> > > > hbase(main):013:0> scan '.META.', {FILTER => "PrefixFilter('t1')"}
> > > >
> > > > Gives all the rows which corresponds to table 't1' as all the row
> keys
> > > > start with "t1".
> > > >
> > > >
> > > > On Fri, Apr 25, 2014 at 3:19 PM, David Koch <[email protected]>
> > > wrote:
> > > >
> > > > > Hi,
> > > > >
> > > > > How do I get all rows from .META. pertaining to a specific table? I
> > > don't
> > > > > want to scan all of .META. and then do a grep on the output since
> > > .META.
> > > > is
> > > > > quite large but I don't know how to use HBase filters from hbase
> > shell.
> > > > >
> > > > > Thanks,
> > > > >
> > > > > /David
> > > > >
> > > >
> > > >
> > > >
> > > > --
> > > > Bharath Vissapragada
> > > > <http://www.cloudera.com>
> > > >
> > >
> >
>
