Home Page for the TeradataForum
 

Archives of the TeradataForum

Message Posted: Fri, 30 Jan 2009 @ 08:38:06 GMT


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


Subj:   Memfails - who is the culprit?
 
From:   steve.tennant

Firstly please forgive newbie ignorance.

I am using a Teradata box which is apparently experiencing many memfails at certain times. They occur on any weeknight, but not every weeknight, and at the same time (around 18:50).

I have looked in the DBQLog* tables and can not see any significant activity which corresponds with these times.

So my questions are:

Are there things which could cause many (>1,000,000) memfails which are not recorded in the DBQ* tables?

Am I looking in the wrong place?

Am I misunderstanding the issue?


Here is an extract from resusagespma for the last few weeks.

     Date          Time   Node   MinMB Mem Free   SUM Page Drops   SUM Proc Swap    Mem Fails
     2009-01-09   1,850   110        0.67             8.00             2.00      2,238,160.00
     2009-01-15   1,850   109        0.51            12.00             4.00        989,658.00
     2009-01-22   1,850   109        0.77            11.00             3.00        862,952.00
     2009-01-20   1,850   109        0.78            11.00             3.00        270,458.00
     2009-01-05   1,840   109        0.91             5.00             3.00        101,102.00
     2009-01-02   1,840   110        0.34            10.00             1.00         15,671.00
     2009-01-27     400   109       16.25             6.00             3.00          2,038.00
     2009-01-06   1,850   109       85.84             6.00             3.00            132.00
     2009-01-21   1,150   110       49.91            15.00             1.00             28.00
     2009-01-28   1,140   109       23.52             5.00             3.00              2.00
     2009-01-30   1,150   109       50.99            17.00             3.00              0.00
     .........

As you can see just a few nights have this problem, and the nights vary each week. Now there could easily be users who have jobs scheduled to run from a number of sources and there could be old or forgotten admin jobs (backups, stats, etc) running. So what I need to know is where to look to identify what is causing this.

And of course, should I care? My view of memfails is "I'm paging, so what" but when I see the huge numbers at precisely the same times I really want to know what is going on.

Thanks in advance for you help.


Steve T



     
  <Prev Next>  
<<First
<Prev
Next> Last>>  
 
 
 
 
 
 
 
 
  
  Top Home Privacy Feedback  
 
 
Copyright for the TeradataForum (TDATA-L), Manta BlueSky    
Copyright 2016 - All Rights Reserved    
Last Modified: 27 Dec 2016