ACF2 CPF journals failing to sync / not updated
search cancel

ACF2 CPF journals failing to sync / not updated


Article ID: 242692


Updated On:




There is a CPF connection between TESTSYS1 and TESTSYS2. The sync is one-way from TESTSYS1 and TESTSYS2. 

Today, the sync stopped working sometime today when it was noticed that a dataset rule change on TESTSYS1 did not get updated on TESTSYS2. The rule change did successfully complete for TESTSYS1.

Why is CPF failing to update rules on the other system?


Release : 16.0

Component : ACF2 for z/OS


Restarting the CCILGR task caused the LOGGER database to reorg and ACF2 commands were successfully sent afterwards.

The logical flow for CPF is that after a command leaves ACF2 it goes to CAICCI which then writes to the LOGGER file. The CPF SEND journal for the node only gets written to after the LOGGER database is written to. If the command is executed on the home node (TESTSYS1 in the example scenario above), it is executed prior to sending the command to remote nodes (ex. TESTSYS2).

Verify the following:

  1. Is the command present in the SEND journal for the node?
  2. Is CPF ACTIVE for ACF2? (This can be verified with a TSO ACF SHOW CPF command)
  3. Is Journaling turned on in ACF2? (This can be verified with a TSO ACF SHOW CPF command. If journaling is not turned on, then it will need to be turned on to diagnose CPF sending and receiving issues.)

The outcome of these findings will lead to whether the issue is related to ACF2 or isolated to CAICCI and the LOGGER database. If no changes have been made to either systems ACF2 CPF records or CAICCI settings, the most common issue is the CAICCI LOGGER database needing a reorg. The Common Components and Services for z/OS documentation Configure Assured Delivery section provides information on why and how to reorg the LOGGER database.