IDC3013I: Duplicate data set name

What it means

Every name in a catalog has to be unique. When IDCAMS defines a cluster, an alternate index, a GDG base, a non-VSAM entry or an alias, or renames an entry, catalog management adds the new name. If the name is already there, catalog management refuses, and IDCAMS writes IDC3013I.

The next message, IDC3009I, carries return code 8 with reason code 38, the code for a name that already exists when adding an entry. The module shown is usually IGG0CLEH. The command fails, nothing is defined, and IDCAMS sets a nonzero condition code.

A VSAM cluster uses more than one name. The cluster has a name, and so do its data component and, for a KSDS, its index component. Any of them can clash with an existing entry.

The usual story is a rerun. The job deletes and redefines a cluster, the DELETE didn't happen or failed, and the DEFINE then meets the cluster left from the previous run. IBM APARs also show the message from software that renames data sets into a catalog that already holds the target name.

Common causes

• The job was rerun and the data set from the earlier run still exists, because the DELETE before the DEFINE failed or was skipped.
• The DELETE ran against the wrong catalog or with a misspelled name, so it removed nothing and the DEFINE failed.
• The data or index component name chosen for the new cluster is already used by another entry.
• An ALTER NEWNAME or a rename job used a target name that already exists in the catalog.
• Two jobs define the same name, for example a test job and a production job sharing a qualifier.

How to find the problem

1. Read the name in the DEFINE or ALTER command just above IDC3013I in SYSPRINT.
2. Check IDC3009I below it. Return code 8 with reason 38 confirms a duplicate name. Another code means a different problem.
3. Run LISTCAT ENTRIES('PROD.PAYROLL.MASTER') ALL for the cluster name and for the data and index component names, to see what already holds the name.
4. Check the messages for the DELETE step or command that ran before the DEFINE, to see why it didn't remove the old entry.

How to fix it

• Delete the existing entry first if it is the leftover from a failed run, then run the DEFINE again.
• Make the job safe to rerun. Put a DELETE before the DEFINE and follow it with IF LASTCC = 8 THEN SET MAXCC = 0, so a missing data set doesn't stop the job but a real duplicate is cleared.
• Pick a different name, or give the data and index components names of their own.
• For a rename, check the target name with LISTCAT before running the rename.

What it looks like

IDC3013I DUPLICATE DATA SET NAME
IDC3009I ** VSAM CATALOG RETURN CODE IS 8 - REASON CODE IS IGG0CLEH-38

Last reviewed 2026-10-18 by the WND Software team. Written in our own words. For the full reference, check IBM’s documentation for your z/OS release.