> -----Original Message-----
> From: IBM Mainframe Discussion List On Behalf Of Jack Kelly
>
> or the changed bit was reset and hsm thought that the dsn had
> not changed and then reconnected?
Opened a PMR with DFSMShsm, and among other things they explained how
hsm determines a dataset's eligibility for "Fast Subsequent Migration"
(FSM), or "reconnection" to the most recent migration copy:
" [H]ere is the general
logic behind the eligibility checking for FSM (aka reconnection):
The "change bit" (DS1IND02) is checked in the Format1 DSCB of the
dataset being evaluated for possible reconnection.
If the bit is on, the dataset is migrated normally, that is, no
reconnection is attempted.
If the bit is off, DFSMShsm must determine if that is the case because
the dataset has not been opened for output since being recalled, or if
DFSMShsm turned the bit off when it backed up the dataset after it was
recalled. This determination is made using information in our BCDS.
If the BCDS indicates that the dataset was backed up on or after the
date of recall, the dataset is migrated normally (no reconnection).
If the BCDS shows that no backup copy exists at all, or that the most
recent backup copy was made prior to the recall, the dataset is deemed
eligible for reconnection since there is no evidence that it has changed
since being recalled."
We had "moved" the volume containing the updated dataset to the new
datacenter via DFSMSdss full-volume dump of the source volume to tape,
followed by full-volume restore to the new target volume in the new
datacenter. HSM level2 consulted with DSS level2, and were informed
that DSS does NOT reset any "change bits" unless explicitly instructed
to do so via the RESET keyword in the DUMP command. Note that we did
NOT specify RESET anywhere, at any time.
Yesterday I performed a test on our "sandbox" system and found that the
DSS DUMP FULL / RESTORE FULL sequence DOES reset ALL "change bits" on
the target volume, WITHOUT having specified RESET. I forwarded doc of
my test to IBM, and DSS now has the PMR. They are examining my doc and
will attempt to duplicate my result in their lab.
-jc-
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
