RSPCCHAIN table — Process Chain Header Table in BW
RSPCCHAIN is the BW header table for process chains: one row identifies a chain by its technical ID and version (active or modified). It does not store the step sequence, execution history, or run status of a chain - those live in separate node, variant, and log tables. It is a design-time metadata table, read through transaction RSPC rather than queried directly in most troubleshooting.
This page covers what RSPCCHAIN stores in the BW process chain framework, which fields and joins actually exist, and the reflex mistakes consultants make when they try to use this table to explain why a chain did not run. It also covers how the table behaves across the active/modified version split and where it sits relative to execution logging.
Published 15 Sept 2026· 998 words
What it stores
One row in RSPCCHAIN represents the header of one process chain at one version, active or modified, in one client. A process chain built in the chain maintenance tool is a directed graph of processes - a start process plus a network of dependent load, DTP, activation, and custom-ABAP steps - but RSPCCHAIN itself only anchors that this chain exists under a given technical name and carries chain-level attributes rather than the graph itself. Every chain has an active-version row that the background scheduler actually runs, and while someone is editing the chain in the maintenance transaction a modified-version row also exists until it is activated. Deleting a chain removes its header row here along with its dependent definition and log tables, though metadata buffers may lag behind the deletion.
Key fields
- MANDT - client
- CHAIN_ID - technical name of the process chain, the key referenced from every other process chain table, from the background job that schedules it, and from the log tables that record its runs
- VERSION - version indicator distinguishing the active version executed on schedule from the modified version being edited and not yet activated
How it joins the data model
- RSPCCHAIN-CHAIN_ID = RSPCPROCESSLOG-CHAIN_ID - matches the chain header to every execution log entry recorded when the chain runs
- RSPCCHAIN joins internally to the chain's node and variant definition tables on CHAIN_ID together with VERSION, so an active row and a modified row never get mixed when reconstructing what will actually execute
- There is no direct join from RSPCCHAIN to an InfoProvider table; a chain that loads data reaches RSDODSO or RSDCUBE indirectly, through the load requests its steps trigger, which are tracked in RSREQDONE
- CHAIN_ID has no foreign key relationship to RSDIOBJ or the extraction metadata tables (ROOSOURCE, ROOSFIELD, ROOSGEN) - those describe the source side of extraction, not the orchestration layer
How to read it safely
Always restrict on MANDT. CHAIN_ID is the natural and only meaningful selection - get the exact technical name from the chain maintenance transaction before querying anything, since chain IDs are not descriptive. Restrict VERSION to A for what actually runs; only pull M when specifically investigating whether an edit was ever activated. The table itself is small, one or two rows per chain, so it is never a performance concern on its own; if a selection against process-chain data feels slow, the volume is almost certainly in the log table, not here.
How to prove it in the data
Symptom: a chain edited yesterday does not appear to run the new logic today. Select RSPCCHAIN by the exact CHAIN_ID with both VERSION values. If a row exists for M but not for A, or the A row's content clearly predates the edit, the change was saved but never activated in the chain maintenance transaction, so the scheduler is still executing the last activated definition, not the visible one on screen.
ECC vs S/4HANA
The process chain concept and its maintenance transaction carried forward into S/4HANA-based BW and into BW/4HANA largely unchanged conceptually, and RSPCCHAIN as a design-time header table remains part of that framework. Because it is pure metadata rather than reportable transactional or master data, it is not typically exposed through a compatibility CDS view the way InfoProvider or extraction tables are. Version-specific structural changes to the table across BW/4HANA generations are not something to assert with confidence without checking the specific release in front of the reader.
Common pitfalls
- Assuming that saving a chain edit makes it live - the active version keeps running until the modified version is explicitly activated, and RSPCCHAIN will show both rows simultaneously with no obvious flag pointing at which one the scheduler actually uses next
- Treating RSPCCHAIN as an execution history table - it holds the definition, not run status, start time, or duration; those questions belong to the process chain log table, not this one
- Trying to read the step sequence or dependency graph out of RSPCCHAIN - the order of processes, their predecessor and successor relationships, and their variants live in separate node and link tables, so this table cannot answer what runs after a given step
- Assuming CHAIN_ID uniqueness spans systems - a technical ID is unique per system and client, and a chain transported inconsistently between systems can carry the same ID with a different internal definition, which a single-system look at RSPCCHAIN will not reveal
- Expecting an immediate disappearance of a deleted chain from schedulable lists - buffering in the chain maintenance tool can leave a deleted chain visible or schedulable until the buffer is refreshed, which reads as a phantom chain to anyone unfamiliar with the caching behavior
Whose problem this is
The BW competency center or basis-adjacent scheduling team that owns process chain design and job scheduling owns questions against this table. Functional consultants raise the symptom - a chain step that did not fire - but resolving it usually means someone with authorization in the chain maintenance transaction checking activation status and the background job, not a raw select against RSPCCHAIN.
Related SAP objects
Reviewed pages this object connects to in the ERPClimb knowledge graph.
Source: ERPClimb — https://erpclimb.com/sap-tables/rspcchainERPClimb is an independent platform and is not affiliated with SAP SE. Reference pages are written and reviewed by SAP consultants for learning and troubleshooting.