Dealing with AMPS
This section describes dealing with down and nonparticipant AMPs.
The impact of down AMPs on Teradata MultiLoad tasks depends on:
If all of the target tables are fallback and not more than one AMP is down, then Teradata MultiLoad tasks continue to run as long as there is not more than one AMP down, either logically or physically, in a cluster. The down AMP does not participate in the application phase if:
If all of the target tables are fallback and two or more AMPs are down, then Teradata MultiLoad tasks do not run, or terminate if two or more AMPs are down, either logically or physically, in a cluster.
Note: In the application phase, if AMPs are down to the extent that data on the disk is corrupted, then the affected tables must be manually restored.
If one or more of the target tables is nonfallback and one or more AMPs are down, then Teradata MultiLoad tasks terminate and they cannot be restarted until all of the AMPs are back up.
Note: Teradata MultiLoad also terminates if I/O errors corrupt the target tables in the application phase.
There are three ways for an AMP to become nonparticipant for a Teradata MultiLoad task.
An AMP can only become a nonparticipant if the target table is defined as having fallback protection because Teradata MultiLoad does not run after the acquisition phase. The AMPCHECK APPLY and ALL options would prevent the occurrence of nonparticipant AMPs in this situation.
In effect, Teradata MultiLoad treats a nonparticipant AMP as if it were a down AMP. Thus, Teradata MultiLoad does not run if a cluster has any combination of more than one AMP that is:
If more than one AMP in a cluster becomes a nonparticipant during the application phase, the Teradata MultiLoad task cannot continue. The target tables are considered unusable, and must be recovered from archives.