SAP Management Cockpit

The SAP Management Cockpit is the single-pane-of-glass entry point for the entire Fivetran SAP demo estate. It is a static HTML dashboard hosted on sapidesecc8 that shows the live reachability and service health of every SAP system, links to each system's individual cockpit page, and provides access to monitoring, email alerting, and administrative tools — all from one URL.

Any Sales Engineer working with the SAP demo environment starts here.

https://sapidesecc8.fivetran-internal-sales.com/sap_skills/docs/SAP_Management_Cockpit.html

---

What's on the dashboard

System status grid

Six server cards, each with a live reachability indicator that updates on page load:

Server card	System	What it links to
S/4HANA 2023	sapidess4 — SAP S/4HANA SPS05	S/4HANA cockpit + setup guide
ECC 6.0 EHP8	sapidesecc8 — SAP ECC on Oracle 19c	ECC cockpit + setup guide
ECC SQ1	sap-sql-ides — SAP ECC on SQL Server 2012	SQ1 cockpit + setup guide
SAPRouter	saprouter (34.46.174.105)	SAPRouter cockpit + documentation
HVR Hub	saphvrhub — HVR/HVA replication control plane	HVR Hub cockpit + documentation
SSH Tunnel	ts-sap-hana-s4-ssh-tunnel (10.142.0.37)	SSH Tunnel guide

Each card shows a colored dot — green (reachable), red (unreachable), or gray (check in flight) — updated automatically when the page loads via 6 parallel API calls to /sap_skills/api/reachability?system={system}.

External links

Two quick-access link cards are hardcoded into the page:

SAP links: SAP Notes search (me.sap.com), Software Downloads Center, S/4HANA SSO login via Okta SAML.

Fivetran links: Fivetran Dashboard, HVR Hub Console (saphvrhub:4340), POV App, Slab.

Monitoring tabs

The dashboard includes tabs that surface:

• Service Monitor — per-service health across all 6 servers (see below)
• Monitoring Agent — email distribution list management

---

How the reachability checks work

On page load, JavaScript calls the backend for each server in parallel:

GET /sap_skills/api/reachability?system=s4hana
→ { "reachable": true }

The backend (gcs_explorer_server.py on sapidesecc8) maps each system identifier to a probe appropriate for that server's stack:

system param	Server	Probe method
s4hana	sapidess4	SSH + sapcontrol GetProcessList
ecc8	sapidesecc8	Local sapcontrol
sq1	sap-sql-ides	SOAP sapcontrol port 50013 + WinRM
saprouter	saprouter (34.46.174.105)	SSH + systemctl is-active saprouter
hvrhub	saphvrhub	SSH + systemctl is-active
sshtunnel	ts-sap-hana-s4-ssh-tunnel (10.142.0.37)	SSH

---

Service Monitor

The Service Monitor tab shows fine-grained health — not just "is the server up?" but "is the SAP Dispatcher running? Is the Oracle DB process alive? Is the HVR Hub service active?" — updated every 5 minutes by a cron-driven bash script.

How it works:

1. /usr/local/bin/sap_health_check.sh runs every 5 minutes on sapidesecc8
2. It probes all 6 servers and writes a snapshot to /usr/sap/sap_skills/health_status.json
3. The SAP_Monitoring.html page reads that JSON and renders status cards with green/red/gray dots

Services probed per server:

Server	Services checked
sapidess4	SAP Dispatcher, Gateway, ICM, IGS; HANA tenant FIV; HANA tenant PIT
sapidesecc8	SAP Dispatcher, Gateway, ICM; Oracle DB (ora_pmon process); Web Server (HTTPS curl)
saprouter	Server reachability; saprouter systemd service
sap-sql-ides	Server ping; SAP instance (SOAP port 50013); SQL Server DB (WinRM)
saphvrhub	Server reachability; HVR Hub service; PostgreSQL 14
SSH Tunnel	Server reachability

Status indicators:

Indicator	Meaning
Green dot	Service running
Red dot	Service down or unreachable
Gray dot	Check in progress or initial state
REACHABLE pill	Server responded to SSH/ping
UNREACHABLE pill	Server timed out

health_status.json is also stored in the Postgres doc_pages table — that's why its version number is in the thousands (overwritten every 5 minutes by the cron).

---

Email monitoring and alerts

The Monitoring Agent section of the cockpit manages the mailing list for system alerts, certificate expiry warnings, and health notifications. All outbound mail from the SAP estate is relayed through smtp2go using the verified domain fivetran-internal-sales.com.

Property	Value
SMTP provider	smtp2go (mail.smtp2go.com:2525)
From address	sap-skills@fivetran-internal-sales.com
Distribution list	SAPSpecialists → antonio.carbone@fivetran.com, richard.brouwer@fivetran.com
Vault key	smtp2go

Email APIs (all POST):

Endpoint	Purpose
/sap_skills/api/get_mailing_list	Read a list by name
/sap_skills/api/update_mailing_list	Update a list (rewrites mailing_lists.json)
/sap_skills/api/send_test_email	Send a test message to a list

---

Backend architecture

The cockpit is a static HTML + JavaScript page — it has no server-side rendering. All dynamic behaviour comes from JavaScript calling backend APIs on gcs_explorer_server.py.

Property	Value
URL	https://sapidesecc8.fivetran-internal-sales.com/sap_skills/docs/SAP_Management_Cockpit.html
Hosting server	sapidesecc8, port 443
Filesystem path	/usr/sap/sap_skills/docs/SAP_Management_Cockpit.html
Authoritative copy	Postgres doc_pages table, path /sap_skills/docs/SAP_Management_Cockpit.html
Backend service	gcs-explorer.service (gcs_explorer_server.py)
Theme colour	Teal (#0d7377)

Two HTML files make up the cockpit:

• SAP_Management_Cockpit.html — the live dashboard (server cards, status grid, monitoring tabs)
• SAP_Management_Cockpit_Documentation.html — internal documentation page

---

Deploying changes

HTML-only edits take effect immediately — no restart needed. The portal canonicalises content through Postgres first:

# 1. Edit locally
vim ~/SAP_Skills/docs/SAP_Management_Cockpit.html

# 2. Upsert into Postgres
ssh sapidesecc8 "sudo -u postgres psql -d portal" <<EOF
INSERT INTO doc_pages (path, content, content_type, sha256, version, updated_at, updated_by)
VALUES ('/sap_skills/docs/SAP_Management_Cockpit.html',
        pg_read_file('/tmp/cockpit.html'), 'text/html; charset=utf-8',
        '$(sha256sum < /tmp/cockpit.html | cut -d' ' -f1)', 1, now(),
        'antonio.carbone@fivetran.com')
ON CONFLICT (path) DO UPDATE SET version = doc_pages.version + 1,
  content = EXCLUDED.content, sha256 = EXCLUDED.sha256, updated_at = now();
EOF

# 3. Copy to disk
scp ~/SAP_Skills/docs/SAP_Management_Cockpit.html \
    root@sapidesecc8:/usr/sap/sap_skills/docs/

After editing gcs_explorer_server.py (backend changes — e.g. new reachability probe):

ssh sapidesecc8 'systemctl restart gcs-explorer'

---

Adding a new server

Five steps — order matters (the UI will 404 if HTML ships before the backend handler):

1. Create a per-server cockpit page at ~/SAP_Skills/docs/SAP_NewServer.html (model after an existing one such as SAP_HVRHub.html)
2. Add a server card to SAP_Management_Cockpit.html inside .status-grid, including <span id="reach-newserver"> for the status dot
3. Add a reachability handler in gcs_explorer_server.py — extend the system → probe function map with the appropriate check method for that server's stack
4. Add the JS call in DOMContentLoaded:

   checkReachability('newserver', 'reach-newserver');

5. Deploy — upsert both HTMLs into doc_pages, scp to disk, then systemctl restart gcs-explorer

To also add the server to the Service Monitor, additionally:

• Extend /usr/local/bin/sap_health_check.sh with a probe block for the new server
• Add a rendering section in SAP_Monitoring.html to display the new key from health_status.json

---

Common issues

Symptom	Cause	Fix
All status dots stay gray	gcs-explorer not running, or reachability handler crashing	systemctl status gcs-explorer; check journalctl -u gcs-explorer -n 50
One server shows red but is alive	Wrong probe method for that server's stack	Inspect the handler in gcs_explorer_server.py; switch to an SSH-based check
HTML edits don't appear after scp	Browser cache or ETag mismatch	Hard-reload (Cmd-Shift-R); verify doc_pages.sha256 matches the deployed file
sapcontrol checks spuriously red on sapidess4	XSA / instance not started yet after HANA restart	Wait 60–120 s after HDB start; run sapcontrol -nr 03 -function GetProcessList manually
SQ1 reachability flips green/red intermittently	WinRM timeout race; SOAP port 50013 slow under load	Increase per-check timeout in gcs_explorer_server.py
health_status.json has stale last_update	Cron not running or sap_health_check.sh failing	crontab -l on sapidesecc8; tail -50 /var/log/sap_health_check.log
Mailing list edit not reflected in UI	Cached browser state	Hard-reload; verify mailing_lists.json updated_at advanced in doc_pages
New server card breaks layout	.status-grid auto-fit minmax(280px, 1fr) doesn't balance 7 cards	Use explicit grid-template-columns or split into two grids

---

Individual system cockpits

Each server card on the Management Cockpit links to a dedicated per-system cockpit. Those pages have their own documentation and skills — don't edit them here:

Server	Cockpit HTML	Skill
sapidess4 (S/4HANA)	SAP_S4HANA_2023.html + _Guide.html	SAP_Skills:sap-skills-portal
sapidesecc8 (ECC Oracle)	SAP_ECC6_EHP8.html + _Guide.html	SAP_Skills:sap-skills-portal
sap-sql-ides (ECC SQL Server)	SAP_ECC6_SQ1.html + _Guide.html	sq1-license
saprouter	SAP_SAPRouter.html + SAP_SAPRouter_Documentation.html	saprouter
saphvrhub	SAP_HVRHub.html + _Documentation.html	—
SSH Tunnel	SAP_SSHTunnel.html + _Guide.html	—

---

Related

Resource	Link
SAP Skills Portal	sapidesecc8.fivetran-internal-sales.com/sap_skills
SAP Skills Portal repo	fivetran/SAPSkillsPortal
Service Monitor page	/sap_skills/docs/SAP_Monitoring.html

---

Contacts

Antonio Carbone and Richard Brouwer — Fivetran SAP Specialists.