| Previous | Next |
| hrColumnMaxTruncated | hrKeyChanged |
hrwrnDataHasChanged
Interpretation in the backup state machine
For hrwrnDataHasChanged, a correct handler first decides whether the value is success, warning, or failure, then examines the documented outputs. The decisive boundary is the record or data image changed while the caller was using a prior observation.
The stored Value is 0x8800064A (positive JET status 1610; 1610). The HRESULT carries a warning, so the call may have useful output or a valid cursor state that must be inspected. Current ESE documentation uses JET_wrnDataHasChanged for the corresponding published JET condition.
The first useful distinction is that this is an optimistic-concurrency signal, not automatically physical corruption or a write conflict at the lock manager. Start by capture bookmark, update sequence, transaction boundaries, before/after version identifiers, and which operation detected the change. That separates a reproducible incident from a later generic cleanup or service error.
Start with these observations
Preserve the first result before retries, cleanup, service restart, or file replacement changes the evidence. The diagnostic record should identify the exact API phase and the owner of every handle or artifact involved.
- Code-specific observation: capture bookmark, update sequence, transaction boundaries, before/after version identifiers, and which operation detected the change.
- Transaction boundaries: record it together with this result and the timestamp of the first occurrence.
- Before/after values or hashes: record it together with this result and the timestamp of the first occurrence.
- Update owner and timing: record it together with this result and the timestamp of the first occurrence.
- Record key or bookmark: record it together with this result and the timestamp of the first occurrence.
Use hashes, lengths, IDs, generation numbers, and redacted samples instead of copying directory contents or sensitive database values into routine logs. For this it investigation, a complete provenance chain is often more useful than a second automatic retry.
Objects and state involved
| Diagnostic layer | record versioning and optimistic concurrency |
|---|---|
| Relevant API surface | record retrieval, prepared updates, bookmarks, transactions, and compare-before-write logic |
| Code-specific boundary | the record or data image changed while the caller was using a prior observation |
| Narrow corrective direction | re-read under the intended transaction model and revalidate business predicates before applying an update |
ESE snapshot/version behavior can make a prior observation stale without corrupting the record., a caller must define whether to retry, merge, or reject when data changes.
Recovery or retry plan
- Record it,
0x8800064A, the API name, the current phase, and all live context or file owners. - Verify the decisive condition by capture bookmark, update sequence, transaction boundaries, before/after version identifiers, and which operation detected the change.
- Apply only the narrow correction: re-read under the intended transaction model and revalidate business predicates before applying an update.
- After it, recreate any context invalidated by the failure; do not carry stale HBC, cursor, file, or restore-map state into the retry.
- repeat the smallest non-destructive test that reaches the same boundary, then verify both the return value and the resulting file, cursor, backup, or database state.
Not the same as
It specifically means that this is an optimistic-concurrency signal, not automatically physical corruption or a write conflict at the lock manager. Related values below can appear in the same workflow but require a different response:
hrLogWriteFail | the write-ahead log could not durably accept a required record |
|---|---|
hrColumnSetNull | a column update resulted in a NULL value as an explicit outcome |
hrLogSequenceEnd | the legacy log generation namespace reached its maximum value |
Keep the original constant and hexadecimal value in telemetry. Replacing it with “backup failed” or “database warning” removes the state information needed to select the next legal API call.
Actions that can make diagnosis worse
- do not classify every data-change warning as storage damage.
- do not overwrite the newer record from a stale snapshot.
- do not discard the first lower-level Win32, RPC, or JET result merely because a later cleanup call returned a more familiar error.
Acceptance criteria for a fix
A useful regression test for this HRESULT should force the condition “the record or data image changed while the caller was using a prior observation”, call one documented API transition, and assert the exact HRESULT. The corrected the case should change only the decisive precondition and should verify cleanup as well as the primary output. For the result backup or restore path, also prove that the resulting set can be enumerated and that no file handle or context remains active after finalization.
Technical references
- ESE transaction model — API ordering, file semantics, warning/error interpretation, or recovery behavior relevant to it.
- ESE warning table
- Open-source ESE implementation
Looking for a different code? Search another status or error code.