DFHTD0380 applid Illegal attempt to read control interval 0 for the intrapartition data set (DD name ddname).
Explanation
Control interval 0 in the intrapartition data set is reserved for transient data control information. The remaining control intervals are allocated to hold data for queues as determined by transient data processing on behalf of application program requests.
An invalid attempt has been made to read control interval 0 for the intrapartition data set with ddname ddname.
System action
This is a critical error and CICS is terminated, even if you have specified in the dump table that CICS should not terminate.
A system dump with dumpcode TD0380 is taken unless you have specifically suppressed dumps in the dump table.
Message DFHME0116, which contains the symptom string for this problem, is produced.
User response
Each CICS record for an intrapartition transient data queue contains pointers which, if the queue is not empty, are relative byte addresses (RBAs) associated with the intrapartition data set with ddname.
Furthermore each allocated control interval, apart from the first, contains one or more user records as well as a queue control record. This latter record, the first in the control interval, contains the forward chain pointer or RBA for the next control interval containing data for the queue.
In each case, transient data assumes that RBAs address record boundaries within the intrapartition data set.
The assumption can be violated in several ways. The type of violation may be determined from:
-
a control interval print of the intrapartition data set, or
-
using Access Method Services, or
-
using the system dump.
Violations include:
-
THE WRONG INTRAPARTITION DATA SET WAS USED If the wrong data set has been used, that is, the data set used for this CICS start up was not used for the previous CICS start up, then it is highly probable that most of the RBAs in the CICS records for an intrapartition transient data queue will not address record boundaries in the intrapartition data set. A cold start or an initial start of CICS must be carried out.
-
THE INTRAPARTITION DATA SET WAS ALTERED If the records have been moved, possibly through data set compression, then it is highly probable that control interval 0 will contain more than one record and that most of the RBAs in the CICS records for an intrapartition transient data queue will not address record boundaries in the intrapartition data set. A cold start or an initial start of CICS must be carried out.
-
A CICS record for an intrapartition transient data queue WAS CORRUPTED If a CICS record for an intrapartition transient data queue has been corrupted, it is highly probable that just one or two RBAs will not address record boundaries in the intrapartition data set. A specialized trap may be required to identify the offending program.
CICS may be restarted. An emergency restart for transient data will result in the RBAs being reconstructed from the system log and the intrapartition data set.
If an activity keypoint was taken between the occurrence of the error and its detection, it may prove necessary for a cold start or an initial start of CICS to be carried out.
-
AN I/O BUFFER WAS CORRUPTED If an I/O buffer has been corrupted, it is highly probable one of the RBAs in the CICS record for an intrapartition transient data queue will not address record boundaries in the I/O buffer. A specialized trap may be required to identify the offending program.
CICS may be restarted. An emergency restart for transient data results in the RBAs being reconstructed from the system log and the intrapartition data set.
Note: If the contents of the I/O buffer were written to the intrapartition data set between the occurrence of the error and its detection, it may prove necessary to perform a cold start or an initial start of CICS.
-
A CICS LOGIC ERROR OCCURRED If you need further assistance from IBM to resolve this problem, see IBM problem support for guidance on how to proceed.
Module
XMEOUT parameters/Message inserts
- applid
- ddname
Destination
Console