Archives of the TeradataForum

Message Posted: Thu, 21 Jun 2013 @ 01:00:25 GMT

	<Prev	Next>		<<First	<Prev	Next>	Last>>

Sirupa Ashok-

The fact that this is always occurring and it is associated with a vproc would tend to indicate that you have a disk that is degraded. You now know which node and which AMP is suspect. If you want to delve deeper, you can use the query below. I would run it against the suspect AMP and another AMP on the same node for comparative purposes. I would expect to see a very high count for the outstanding request (DiskoutReqAvg) which would be reflective of successive retries.

As I said earlier, you should consider opening an incident with the GSC. With the information in hand, they should resolve it very quickly.

Regards,

-dave.clark

     select  cast((ProcId/32)*100 as integer) + (ProcID mod 32) (title 'Node',
             format 'ZZ9-99')
            ,VprocNo (title 'Vproc',format 'ZZZ9')
            ,DiskUse (title '% Disk',format 'ZZ9.99')
            ,DiskReads (title 'Reads',format 'ZZZ,ZZZ,ZZZ,ZZ9')
            ,DiskWrites (title 'Writes',format 'ZZZ,ZZZ,ZZZ,ZZ9')
            ,DiskoutReqAvg (title '% Req Out',format 'ZZ9.99')
            ,PctService (title '% OS',format 'ZZ9.99')
            ,PctAmpWT (title '% Amp',format 'ZZ9.99')
     from    table (SYSLIB.MonitorVirtualResource()) as t2
     WHERE VprocType = 'Amp'
       AND VprocNo = xxxx    <==== replace with vproc #
     order by 1,2
     ;

	<Prev	Next>		<<First	<Prev	Next>	Last>>