risingwavelabs/risingwave · error
must be either all snapshot_backfill or no…
Error message
must be either all snapshot_backfill or no snapshot_backfill. Curr: {stream_scan:?} Prev: {prev_stream_scan:?} What it means
`collect_snapshot_backfill_info_impl` walks all StreamScan nodes across fragments and requires them to be uniformly either all snapshot-backfill scans or all non-snapshot-backfill scans. A mix (one scan has snapshot backfill info, another doesn't, or vice versa) makes epoch filling ambiguous, so it errors, naming the current and previous conflicting scans.
Solutions
- Make all MV scans in the job consistent: either all snapshot backfill or none (recreate the offending MV).
- Check each upstream MV's stream_scan_type and align creation settings (snapshot backfill config) across the chain.
- If a version upgrade changed scan types, recreate dependent MVs so the graph is uniform.
Defensive patterns
Strategy: validation
Validate before calling
// Ensure all upstream MV scans in the chain use a uniform snapshot-backfill setting // before creating dependent MVs.
Prevention
- Create MVs in a chain with consistent snapshot-backfill configuration.
- Audit upstream MV scan types before adding a dependent MV.
- Recreate mixed chains so all scans are uniform.
When it happens
Trigger: Building/refreshing a graph where some fragments contain StreamScan nodes of type SnapshotBackfill and others contain plain StreamScan nodes over MVs — e.g. a job mixing snapshot-backfill MV scans with regular MV scans.
Common situations: Creating MVs that chain over other MVs where only some upstreams were created with snapshot backfill enabled; inconsistent stream_scan_type after plan changes or version upgrades.
Understand the failure class
Background: "Invalid state transition" errors: "status must be X, actually Y", "already rejected/charging/uninstalled", "cannot ... while running" — what they mean when a library rejects your call — this error's family across 31 libraries.
Related errors
- new column id not found in scan output
- upstream table id not covered
- upstream table id not set
- auto schema change with drop column only supports Project…
- Different number of internal tables. New
AI-assisted analysis of risingwavelabs/risingwave@6469eb736d (2026-09-11).
Data as JSON: /api/errors/6f5a3c53289022e7.
Report an issue: GitHub.
Appendix: source
Thrown at src/meta/src/stream/stream_graph/fragment.rs:1271
}
_ => false,
};
match &mut prev_stream_scan {
Some((prev_snapshot_backfill_info, prev_stream_scan)) => {
match (prev_snapshot_backfill_info, is_snapshot_backfill) {
(Some(prev_snapshot_backfill_info), true) => {
prev_snapshot_backfill_info
.upstream_mv_table_id_to_backfill_epoch
.insert(
stream_scan.table_id,
stream_scan.snapshot_backfill_epoch,
);
true
}
(None, false) => true,
(_, _) => {
result = Err(anyhow!("must be either all snapshot_backfill or no snapshot_backfill. Curr: {stream_scan:?} Prev: {prev_stream_scan:?}").into());
false
}
}
}
None => {
prev_stream_scan = Some((
if is_snapshot_backfill {
Some(SnapshotBackfillInfo {
upstream_mv_table_id_to_backfill_epoch: HashMap::from_iter(
[(
stream_scan.table_id,
stream_scan.snapshot_backfill_epoch,
)],
),
})
} else {
None
},View on GitHub (pinned to 6469eb736d)