pdh: report duplicate instances by occurrence instead of renaming them (#2742)

Signed-off-by: Jan-Otto Kröpke <mail@jkroepke.de>
This commit is contained in:
Jan-Otto Kröpke
2026-10-16 09:58:43 +02:00
committed by GitHub
parent d8dbf852d1
commit 43d77f6869
32 changed files with 832 additions and 266 deletions
+8
View File
@@ -33,6 +33,14 @@ Edition (`PatchLevel` 16.0.1000.6), named instance `NONDEFAULT`.
15 including `_Total`. `SQL Errors` instances are error classes
(`User Errors`, `Info Errors`, `Kill Connection Errors`,
`DB Offline Errors`); `_Total` equalled their sum (38 = 34 + 4).
- No row type declares `Occurrence` (see `internal/pdh`), because no object
repeats an instance name in one sample: database names are unique per
instance, a database belongs to at most one availability group
(`Database Replica`), `Availability Replica` instances are replicas of the
instance's availability groups, and the other instances are fixed lock
types, error classes, wait statistics or the single unnamed instance. A
sample on this host had no duplicates in any of these objects. If pdh
ever logs dropped duplicates for an `MSSQL$` object, declare the field.
- `Wait Statistics` is unusual: every counter is a wait class and every
instance is a statistic, `Average wait time (ms)`,
`Cumulative wait time (ms) per second`, `Waits in progress` and