Dump data is always historical. Parsers stamped undated events with
time.Now() (Inspur component.log memory/fan status, IDL/H3C SEL with
unparsable time, NVIDIA summary/bug-report, generic, XigmaNAS, Unraid,
Redfish replay warnings), so an August dump opened in September showed
DIMM warnings dated the day of analysis.
- Events without a source time keep a zero timestamp.
- Yearless syslog timestamps take the year from the collection time via
parser.ParseYearlessInDefaultArchiveLocation; zero when unknown.
- Archive/uploaded-snapshot collected_at is no longer defaulted to now;
only live API collection uses the current time.
ADL-070.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Event ingestion now uses the source syslog PRI and an explicit-offset
timezone timeline instead of assuming host-local time. CPU PPIN is
exported as the source-backed CPU serial, and an active NVMe fault SEL
event promotes the matching drive's storage status.
See ADL-056, ADL-057.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
ParseIDLLog deduped by eventID|eventType|description, which is identical
across every occurrence of a recurring alarm (e.g. the flapping PCIe
presence check on Inspur HGX boards). That collapsed every later
occurrence into the first one, so applyGPUStatusFromEvents computed GPU
health from a stale, truncated event list and marked failed GPUs as OK.
Three related fixes for IDL event processing:
1. idl.go: include EventType in dedup key so Deassert events are no
longer silently dropped as duplicates of their Assert counterparts.
2. gpu_status.go: treat Deassert events as clearing all GPU faults —
previously the code re-applied the same faulty GPU set from the
description, leaving GPUs stuck in Critical even after alarm cleared.
3. reanimator_models/converter: add bmc_event_summary section to the
Reanimator export — a deduplicated Critical/Warning event table with
Active/Resolved status derived from Assert/Deassert pairs.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>