executing-plans
executing-plans
Use when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available
sap-cloud-connectorOperate the SAP Cloud Connector (SCC), the on-premise agent linking SAP BTP to on-prem systems — start, stop, restart, the admin UI on 8443, Master/Shadow HA, ports and auto-restart, on Linux/Windows/macOS (not AIX). Use for "start/stop/restart cloud connector", "scc is down", "cloud connector admin UI 8443", "SCC high availability master shadow".
Independent GitHub snapshot · Automated classification · Original documentation governs
SCC is a Java agent that establishes an outbound TLS tunnel from your network to an SAP BTP subaccount, exposing selected on-prem systems to cloud apps under fine-grained access control. It runs on its own JVM (not an SAP kernel instance).
Platform note: SCC runs on Linux, Windows and macOS — not AIX (unlike the SAP kernel components). If the on-prem hosts are AIX, SCC still runs on a separate Linux/Windows host. [C2]
Guardrail: SCC is the single bridge between BTP and on-prem — stopping it breaks all cloud→on-prem connectivity (cloud Fiori, integration flows, RFC/HTTP to the backend). Use the Shadow for zero-downtime; confirm before stopping a Master. Access-control changes expose/hide internal systems — treat as security-sensitive.
Linux — as the SCC/OS admin:
# systemd distributions:
systemctl start|stop|restart scc_daemon
# System V distributions:
service scc_daemon start|stop|restart
Windows: the Cloud Connector is a Windows service (auto-start after install) — start/stop via
Services.msc or net start|stop <SCC service>.
macOS / portable install: the <scc>/ install dir scripts (go.sh / daemon control).
The daemon auto-restarts on host reboot by default (disable per KBA 3474163 if required). [C1]
https://<host>:8443 # default admin UI port; default user "Administrator"
| Purpose | Port |
|---|---|
| Administration UI (HTTPS) | 8443 (default; changeable — KBA 2955431) |
| Outbound tunnel to BTP region | 443 (TLS, outbound only — no inbound firewall opening) |
| Exposed backend systems | the virtual host/port you define per access-control mapping |
Check state and trigger/monitor failover from the admin UI (Master/Shadow status).
ljs_trace.log (main Java trace, rotates ~20×50 MB), plus scc_* and audit logs, in the SCC log/
directory — downloadable from the admin UI (Log And Trace Files). Details in
sap-log-reference.
sap-web-dispatcher (HTTP reverse proxy), sap-saprouter (NI proxy).Every command executed must be traceable to one of exactly three things:
If a command is backed by none of those, do not run it — say what backing is missing and stop. "It's probably fine", "this is standard", and "I recall the syntax" are not backing. When the backing is a source, name it (page or Note number) alongside the command; when it is the user, quote the instruction.
If executing would require assuming anything the user did not state, you are obliged to confirm first. Never fill a gap with a plausible default. Common gaps that force a stop:
A wrong assumption here is not a typo — it is the difference between reading a log and stopping production.
Asking the user for something the system can answer is a failure. Before you raise a question, ask the user to go and look, or request Computer Use / GUI access, you must first try to determine it programmatically. Only what genuinely cannot be derived — intent, authorization, a business decision, a value that exists only in the user's head — is a legitimate question.
| Determine programmatically (do NOT ask) | Ask the user (cannot be derived) |
|---|---|
Which DB — echo $dbms_type, profile dbms/type | Which client to act on |
SIDs / instances / hosts / ports — sapcontrol … GetSystemInstanceList, ls /usr/sap | Whether this system is in scope / approved |
Is it up, is the DB up — GetProcessList, R3trans -d | PRD change approval, downtime window |
Kernel / release / patch — disp+work -version, saphostexec -version | The intended recovery point or retention policy |
Which clients exist — table T000 | Which of those clients is meant |
Free space, log locations, parameter values — df -h, sappfpar, profile | Business impact / urgency |
Order, always: verify programmatically → ask only what remains → never assume.
Work down this ladder. Take the highest rung that does the job — and within that rung, the lowest privilege that suffices. Never skip a rung because you assume it is unavailable (see the burden of proof below).
| # | Path | Privilege | Notes |
|---|---|---|---|
| 1 | REST / OData — GET first | Narrowest. Scoped service user | Read-only by construction when you stay on GET. $metadata gives you the contract |
| 2 | SOAP / web service | Scoped service user | Typed contract via ?wsdl. Client-cert auth where offered — no password in a script |
| 3 | RFC / BAPI (creds exec, JCo, pyrfc) | RFC user with S_RFC | For ABAP writes, prefer this over 1–2: BAPIs have real commit/rollback semantics and land in SM19/SM20 |
| 4 | OS shell as the correct user → DB utility | ⚠️ Escalates — see below | The chain matters more than the rung |
| 5 | Browser automation (headless or in-app) | Interactive user | Session-based UIs only. Fragile across releases |
| 6 | Computer Use / screen driving | Interactive user | Last resort. Not repeatable, not diffable, breaks on any UI change |
⚠️ Rung 4 is a chain, and each link widens the blast radius
ssh <host> ← host access → su - <sid>adm ← SAP admin: can stop/start the system → su - ora<sid> / syb<sid> ← DB owner: can drop data → sudo / root ← everything → hdbsql | isql | dbmcli | sqlplus | db2 ← the actual commandStop at the least-privileged user that can run the command. Most read-only checks need only
<sid>adm; DB utilities usually need the DB owner; root is almost never the right answer andsaproot.shis the rare legitimate exception. Say which user you used and why.
Two axes, and they do not agree
The ladder ranks by automation quality — repeatable, reviewable, loggable, diffable. Privilege runs on a different axis and is worst in the middle: rung 4 (OS/root) can destroy a system, while rung 6 (Computer Use) is merely an interactive user clicking. So Computer Use ranks last for reproducibility, not because it is the most dangerous.
The practical rule: prefer the highest rung, but never escalate privilege to climb it. A read-only OData call beats an RFC that needs a write-capable user; an
<sid>admshell beats a root shell. If climbing a rung requires more privilege than the task needs, stay where you are and say so.
🛑 You must demonstrate the absence of a programmatic path, not assume it
"There is no API for this" is a finding that requires evidence, not a default. Before dropping to rung 5 or 6, actually probe:
- HTTP status codes tell you the access mode.
401→ Basic auth works, scriptable.302regardless of credentials → session UI, browser needed.404→ not deployed.503→ deployed but stopped.- Look for a contract: append
?wsdl,$metadata,/api,?sap-client=and see what answers.- Ask the platform what it exposes:
sapcontrol -function J2EEGetApplicationAliasList,hdbcons help,btp --help,xs help,<tool> -h.- A Swing or WebDynpro UI being un-automatable does not mean the objects are. The editor and the API are different doors — check for the second before declaring the first is the only one.
Programmatic execution is repeatable, reviewable, loggable and diffable; screen-driving is none of those. When you do drop to a lower rung, say which rung you are on and what you probed to rule out the higher ones — so the user can correct you if they know of a path you missed.
Work that produces evidence (logs, traces, command output, screenshots, reports) has two reasonable endings. Ask which the user wants rather than guessing:
Don't dump large output into the conversation unasked, and don't silently discard evidence either — for troubleshooting and any change with a rollback, (a) is usually the right default to offer.
Identify the right OS user before running anything, and switch with a login shell. Wrong-user
execution is a top cause of SAP failures, and the damage outlives the command: files created by root
under /usr/sap, /sapmnt or a DB directory break every later start by the real owner. A login shell
also matters because each user carries the environment the tools need (SAPSYSTEMNAME, ORACLE_HOME/
ORACLE_SID, SYBASE, DB2INSTANCE, library paths) — without it, commands fail or act on the wrong system.
| What you're operating | UNIX user | Windows |
|---|---|---|
SAP instances — sapcontrol, startsap/stopsap, tp, R3trans, disp+work, sappfpar, cleanipc | <sid>adm (lower-case SAP SID) | <SID>adm; services run as SAPService<SID> |
SAP HANA — HDB, hdbsql, hdbnsutil | <sid>adm of the HANA SID (e.g. h10adm — may differ from the SAP SID) | n/a (HANA server is Linux-only) |
Oracle — sqlplus, lsnrctl, BR*Tools | ora<dbsid> (BR*Tools also runs as <sid>adm; generic installs may use oracle) | <SID>adm; DB runs as a service |
SAP ASE — isql, startserver, Backup Server | syb<dbsid> | syb<dbsid> / SAPService<SID> |
IBM Db2 — db2start/db2stop, db2 CLP | db2<dbsid> (the instance owner = DB2INSTANCE) | same; Db2 runs as a service |
SAP MaxDB / liveCache — dbmcli, x_server | sdb (software owner, group sdba) + a DBM operator at DB level | install/service account |
MS SQL Server — sqlcmd, service control | n/a (Windows-only for SAP) | <SID>adm / the SQL Server service account |
SAP Host Agent — saphostexec, saphostctrl | root | Administrator / SAPHostExec service |
Rules
su - <user> (the - is what loads the environment) or sudo -iu <user>.
Windows: use the correct account, or an elevated shell only where documented.root only where the procedure explicitly says so — e.g. saproot.sh after a kernel extract, SAP Host
Agent install/upgrade. Never as a shortcut around a permission error; that is how root-owned files get
created and break the system later.whoami / id, plus the env actually being set (echo $SAPSYSTEMNAME,
echo $ORACLE_SID, echo $DB2INSTANCE, echo $SYBASE).<sid>adm:"), and if the required user is not
available, say so and stop — do not substitute another user.SAP Notes supersede this file. Landscapes differ by release, patch level, DB and OS, and SAP changes procedures via Notes/KBAs between doc revisions.
If the SAP Notes MCP is configured, use it before acting on anything version-specific — especially any destructive step, or when a command here doesn't behave as documented:
search the topic (e.g. the component + symptom, or a Note number cited below).fetch the promising Note IDs for the current text, validity (affected releases/components),
prerequisites and side effects.attachments array. SAP routinely puts the actual deliverable in an attachment rather
than the Note body — sizing guides, SQL script collections, configuration PDFs, spreadsheets. A Note
whose text says "see the attached document" is not fully read until you have it.Downloading an attachment: fetch returns attachments[].url and attachments[].filename;
fetch_attachment retrieves the bytes. Pass the URL verbatim — the URLs are opaque and cannot be
constructed. If your MCP build predates that tool, open the URL in a signed-in browser instead and say the
file was fetched manually.
⚠️ Two ways a hand-rolled fetch goes wrong — both verified. 1. Trusting the status code. An unauthenticated request returns HTTP 200 with a small HTML login stub, not an error. Check the content type and magic bytes, or you save a JavaScript redirect page under a
…/services/attachment.htm?iv_key=…&iv_guid=…— so the URL basename isattachment.htmeven when the payload is a 24-page PDF. Take the name fromattachments[].filenameor the response'sContent-Dispositionheader, never from the URL path.
No MCP available? Look the Note up on me.sap.com/notes/<id> and say the check was skipped rather than
assuming this file is current.
systemctl/service scc_daemon, Windows service). https://me.sap.com/notes/2485510To confirm/deepen: the SAP Cloud Connector documentation for your version (system requirements + Master/Shadow setup) and KBA 2485510 for the exact service/daemon names on your OS.