/sap-data-volume

name: sap-data-volume
description: |
  Read-only DB growth + archivability evidence over RFC — the table-growth watch nobody
  does until the disk is full. `growth` ranks tables by row count (EM_GET_NUMBER_OF_ENTRIES,
  fast DB-stat counts that work on cluster tables too), diffs against a persisted local
  snapshot for a rows/day trend, and flags known log-table offenders (BALDAT, EDID4, CDPOS,
  DBTABLOG) and Z tables that lack housekeeping. `archivability` maps each archiving object
  to its tables (ARCH_DEF), pulls its run history (ADMI_RUN/ADMI_STATS), and reports whether
  a TAANA age analysis exists. Replaces manual DB02/DB15/SARA spelunking. Prerequisites:
  pinned profile via /sap-login (RFC); SAP NCo 3.1 (32-bit). No GUI, no Z-object, no
  dev-init — safe to point at production. `analyze` (schedule TAANA) is v1.5; `archive-run`
  (SARA write/delete) is v2 and refused in v1.
argument-hint: "growth [--top N] [--tables A,B] [--z-only] [--max-tables N] [--physical]  |  archivability [--object O] [--table T] [--residence-days N]"

SAP Data Volume — Growth & Archivability (read-only)

Automate the read-only DB-volume evidence pass: ranked table growth with a local snapshot trend, log-table + Z-housekeeping flags, and per-archiving-object history + analysis coverage — all over RFC with honest COULD_NOT_CHECK where a source is absent on the release.

Task: $ARGUMENTS

**Pure read-only on SAP** in every v1 mode (RFC_READ_TABLE + EM_GET_NUMBER_OF_ENTRIES). The only local write is the growth snapshot file. `--physical` delegates report RSTABLESIZE to /sap-run-report, which asks its **own** Rule-5 confirm — this skill never pre-answers it.

---

Shared Resources

| File | Token / call | Purpose | |---|---|---| | `<SAP_DEV_CORE_SHARED_DIR>/rules/skill_operating_rules.md` | *(rule)* | Mandatory operating rules — read-only here | | `<SKILL_DIR>/references/sap_data_volume_rfc.ps1` | `-Action growth\|archivability … -OutDir` | The two RFC modes → growth.tsv / archivability.tsv | | `<SKILL_DIR>/references/sap_data_volume_offenders.tsv` | catalog | Known log/growth tables + archiving object + ECC physical cluster; override at `{custom_url}\sap_data_volume_offenders.tsv` | | `/sap-run-report` (Skill tool) | RSTABLESIZE / RSSPACECHECK | `--physical` size pass (its own Rule-5 confirm) | | `/sap-sp02` (Skill tool) | spool → file | `--physical` spool collection | | `<SAP_DEV_CORE_SHARED_DIR>/scripts/sap_artifact_lib.ps1` | Step 6 | Artifact index for /sap-evidence-pack | | `<SAP_DEV_CORE_SHARED_DIR>/scripts/sap_log_helper.ps1` | Step 0.5 / Final | Structured run logging |

---

Step 0 — Resolve Work Directory, OUT, Snapshot Dir

powershell -NoProfile -ExecutionPolicy Bypass -Command ". '<SAP_DEV_CORE_SHARED_DIR>\scripts\sap_settings_lib.ps1'; . '<SAP_DEV_CORE_SHARED_DIR>\scripts\sap_connection_lib.ps1'; Write-Output ('WORK_DIR=' + (Get-SapWorkDir)); Write-Output ('STAMP=' + (Get-Date -Format 'yyyyMMddHHmmss')); Write-Output ('RUN_TEMP=' + (Get-SapRunTemp))"

Resolve the pinned connection's SID+client. Set `{RUN_TEMP}` = the `RUN_TEMP=` value printed above (`Get-SapRunTemp` mints + creates the per-run scratch dir holding the log state file; mint it once here and reuse — re-minting breaks the `-Action end` state-file lookup); `{OUT}` = `Get-SapArtifactDir -ScopeKey SYS_<SID>_<CLIENT> -Skill sap-data-volume`. **Snapshot dir (Bucket A, durable — NOT temp):** `{work_dir}\cache\data_volume\<SID>_<CLIENT>` (the engine derives the real SID+client live from RFC_SYSTEM_INFO / USR02, so pass this dir and it self-creates).

Step 0.5 — Start Logging

powershell -ExecutionPolicy Bypass -File "<SAP_DEV_CORE_SHARED_DIR>\scripts\sap_log_helper.ps1" -Action start -StateFile "{RUN_TEMP}\sap_data_volume_run.json" -Skill sap-data-volume -ParamsJson "{}"

---

Step 1 — Parse Arguments & Mode

• **growth** (default): `--top N` (report cutoff, default 20), `--tables A,B` (explicit

scope; else offenders + Z*/Y* tables), `--z-only` (only Z*/Y*), `--max-tables N` (default 200), `--physical` (Step 5 size pass), `--z-threshold N` (Z-housekeeping row threshold, default 1000000).

• **archivability**: `--object O1,O2`, `--table T1,T2` (reverse-map via ARCH_DEF), or

neither (default = objects with run history), `--residence-days N`.

• **analyze** → **v1.5**: say `NOT_YET_IMPLEMENTED — schedules a TAANA batch job (Rule 5);

ships in v1.5 with a recorded GUI leg` and STOP.

• **archive-run** → **v2**: **hard refuse** — `archive-run drives SARA write/delete and is

deferred to phase 2; refused in v1` and STOP. Never attempt it.

Step 2 — Ensure RFC Profile

Pinned via /sap-login. A GUI session is required **only** for the `--physical` delegation chain (/sap-run-report), not for the RFC reads.

Step 3 — growth (read-only)

C:\Windows\SysWOW64\WindowsPowerShell\v1.0\powershell.exe -ExecutionPolicy Bypass -File "<SKILL_DIR>\references\sap_data_volume_rfc.ps1" -Action growth -MaxTables <N> -OffenderFile "<SKILL_DIR>\references\sap_data_volume_offenders.tsv" -SnapshotDir "{work_dir}\cache\data_volume\<SID>_<CLIENT>" -SharedDir "<SAP_DEV_CORE_SHARED_DIR>" -OutDir "{OUT}"

Add `-Tables "A,B"` / `-ZOnly` / `-ZThreshold N` per args. Parse `DATAVOL:` (per-table row/class/sizecat/delta/perday), `DATAVOL_FLAG:` (Z_TABLE_NO_HOUSEKEEPING / LOG_TABLE_LARGE), `DATAVOL_SNAPSHOT:` (BASELINE_CREATED first run — trend `n/a`, honest — else DELTA), `DATAVOL_SKIP:` (table not in DD02L), and `STATUS:`. A connect failure → exit 2, `RFC_LOGON_FAILED`, **no verdict**. A `COUNT_FAILED` row is tri-state — never rendered as 0. The engine writes `growth.tsv` and the snapshot.

Render `growth_report.md` into `{OUT}`: top-N table with rows / class / size-cat / delta / per-day, then the flags section, then (on ECC) the cluster-attribution note for CDPOS/EDID4 (physical size is on CDCLS/EDI40 — `EM_GET_NUMBER_OF_ENTRIES` still counts them fine, verified live). **Grounding rule: every row traces to gr