There are three ways for an AMP to become nonparticipant for an Update driver task:
When... | Then the associated AMP becomes a nonparticipant... |
---|---|
Any AMP is down at the end of the acquisition phase or the beginning of the application phase | If the TD_AMP_CHECK None option is specified. Because the Update driver does not run after the acquisition phase if an AMP is offline and the target table is nonfallback, an AMP can become a nonparticipant only if the target table is defined as having fallback protection. The TD_AMP_CHECK Apply and All options would prevent the occurrence of nonparticipant AMPs in this situation. |
I/O errors occur in certain Update driver tables during the application phase, a head/disk assembly (HDA) fails during the application phase | When the I/O recovery operation stops the Update task. |
A head/disk assembly (HDA) fails during the application phase | But it returns after the disk is replaced and the Disk Copy and Table Rebuild utilities are run. |
The Update driver treats a nonparticipant AMP as if it is an offline AMP. The Update driver does not execute if a cluster has any combination of more than one AMP that is:
- Offline
- Nonparticipant
If more than one AMP in a cluster becomes a nonparticipant during the application phase, the Update driver tasks do not continue; the target tables are considered unusable, and must be recovered from archives.