Add block to queue failed - issue

[shin0066]

Hi,
I am getting "Add block to queue failed" error the eror mesage specify very clearly that scratch disk or buffer pool are full. We are running on MPP system and we use total of 60 nodes -In my config file i have 60 nodes - by looking at the error message -how do i make sure which node has space issues? Does the buffer(11),9 has some meaning?

Code: Select all

buffer(11),9: APT_BufferOperator: Add block to queue failed.  This means that your buffer filesystems all ran out of file space, or that some other system error occurred.  Please ensure that you have sufficient scratchdisks in either the default or "buffer" pools on all nodes in your configuration file.

Thanks in advance.

[ray.wurlod]

The ",9" following the name of the buffer operator tells you that the problem occurred on the tenth-named node in your configuration file (the nodes are numbered beginning from 0).

[shin0066]

Thanks Ray,

so if i start counting from 0 in my config file - when i reached to number 9 node - that is the problem one...

deleted few tempfiles from that scratchdisk area - hopefully job completes now....

thanks again....

[ray.wurlod]

Tip: with these large configuration files it's not a bad idea to add a comment beside each node name indicating its node number.

Code: Select all

/*  Four node configuration beatles.apt */
{

   node "john"  /* node #0 */
   {
         definition
   }

   node "paul"  /* node #1 */
   { 
         definition
   }

   node "george"  /* node #2 */
   {
         definition
   }

   node "ringo"   /* node #3 */
   {
         definition
   }

}
/* If performance is poor, point out that two of them are dead. */