Skip to content
← Back to projects
RMM Supervision — vue principale
Company internship

Infrastructure Monitoring with Datto RMM

Datto RMM · Splashtop · MalwareBytes · MIDRANGE GROUP

Context

At MIDRANGE GROUP, a managed service provider (MSP) based in the Île-de-France region, I worked within the Exploitation team alongside the team's systems & network administrator. The technical floor was split into two units: Technical Support (handling client calls, ticket creation, and L1/L2 resolution) and Exploitation (server maintenance, security, and L2/L3 escalation support). My role in Exploitation gave me direct hands-on access to the company's RMM platform.

Technical challenges

  • Distinguishing a genuine problem from background noise across 550+ endpoints spread over multiple client sites, all monitored remotely, requires a consistent daily triage method, not just a glance at the dashboard.
  • Antivirus coverage gaps only became visible by cross-referencing the Default Dashboard's aggregate counts against the software inventory of individual devices — no single view surfaced both at once.
  • A Quick Job has to be scoped to the exact device group that needs the fix; pushing MalwareBytes fleet-wide when only a handful of machines were missing it would have wasted bandwidth and cluttered the job history for the whole team.
  • Splashtop sessions depend on the end user being present to grant access in real time, so remote troubleshooting has to work around whoever is actually at the keyboard, not around an ideal maintenance window.
  • Datto RMM policies and Quick Job templates are owned at the Exploitation team level, shared across the MSP's full client base — every fix had to work within that existing framework rather than introducing a one-off exception for a single device or site.

Platform: Datto RMM

Datto RMM (Remote Monitoring and Management) is an enterprise-grade platform designed for MSPs and IT teams to remotely monitor and manage endpoints, networks, and systems at scale. I used its core capabilities day to day: real-time device health monitoring, automated patch management, software inventory, remote access, security status tracking, and job automation via Quick Jobs and policies.

Platform Details

Devices monitored
550+ endpoints (desktops, laptops, servers) monitored remotely across multiple client sites
Reference site
MID-S0A (MIDRANGE GROUP internal site)
Agent version
4.4.2195.2195
Remote access tool
Splashtop (integrated into Datto RMM)
Antivirus
Webroot SecureAnywhere Endpoint / MalwareBytes

How I worked — day to day

  1. 1
    Step 1 — Morning fleet review
    Each day started with the Default Dashboard: offline device count by type, antivirus coverage across all client sites, and any new alerts raised overnight by the Datto RMM agents.
  2. 2
    Step 2 — Device-level triage
    For any site or device flagged during the morning review, I opened the individual device profile to check hardware specs, OS version, patch status, and software inventory before deciding on an action.
  3. 3
    Step 3 — Remote remediation via Quick Job
    When a device was missing MalwareBytes or another required agent, I launched a Quick Job scoped to that device (or a small group) to deploy it remotely — no physical access, no user interruption.
  4. 4
    Step 4 — End-user support via Splashtop
    For issues that needed hands-on troubleshooting, I connected to the affected machine via Splashtop with the user's consent, working side by side with Théo on the more complex incidents.

Remote Access: Splashtop

Splashtop is integrated natively into Datto RMM and provides secure remote desktop access to managed endpoints. I used it for: remote technical support for end users (viewing and controlling the client's screen to resolve issues without physical presence); server management tasks (accessing and checking server configurations remotely); and collaborative troubleshooting alongside Théo on complex incidents affecting client infrastructure.

Impact

Antivirus coverage extended across the full monitored fleet (550+ devices), with remote deployment requiring zero on-site travel.

Outcomes

Fleet visibility
Real-time monitoring of 550+ devices with health, patch, and antivirus status at a glance
Remote deployment
MalwareBytes deployed to unprotected endpoints via Quick Jobs — zero on-site travel
Remote support
End-user issues resolved via Splashtop without physical intervention
MSP operations
Direct operational view of MSP work: multi-client management, SLA-driven support, tiered escalation
Skills
Datto RMM, remote monitoring, patch management, endpoint security, Splashtop, MSP workflows

Best Practices

  • Cross-check the dashboard's aggregate numbers against individual device profiles before declaring a fleet healthy — summary counts hide per-device patch or software drift.
  • Scope every Quick Job to the smallest device group that solves the problem; a fleet-wide push when only a few machines need it adds noise without adding value.
  • Explain what you're about to do before taking remote control via Splashtop — a short heads-up to the user avoids confusion mid-session and builds trust.
  • Log every intervention, even a routine Quick Job — in an MSP, the audit trail in the job history matters as much as the fix itself.
  • Escalate ambiguous cases early instead of guessing — triaging fast and handing a complex case to the right specialist beats improvising a fix under uncertainty.

Known Limitations & Decisions

  • Monitoring policies and Quick Job templates were owned at the Exploitation team level, not by any single operator — day-to-day execution (dashboard checks, Quick Jobs, remote sessions) was mine, while policy design belonged to the team lead, the same division of responsibility used across the MSP's client base.
  • Antivirus coverage gaps were closed reactively, once flagged on the dashboard, rather than through a proactive compliance policy that would auto-remediate — that policy-level fix was outside the internship's scope.
  • Splashtop support assumes the device is online and a user is present to grant access; offline machines or unattended desks had to wait for the next Quick Job window instead of an immediate session.
  • RMM-only interventions were tracked through Datto RMM's own job history rather than logged as separate Autotask tickets — that ticketing trail applied to the Support Technique-side incidents, not to routine Exploitation tasks.

Tools & references

Screenshots

RMM Supervision — vue 2