Graham McIntie
1590f78bdc
- LIKE wildcard injection: sanitize %, _ in search queries (devices, sites, gaiia) - Jason.decode! → Jason.decode with error handling for untrusted input - inspect() leak: replace with generic error messages, log details server-side - SSRF protection: URL validator blocks private IPs, localhost, non-HTTP schemes - SSRF validation added to HTTP monitoring executor and integration credentials - GraphQL complexity limits: always applied, not just in prod - GraphQL introspection: also check GET query params, not just body - Stripe webhook: explicit nil/empty checks for signature and body - Cookie security: secure flag for session (prod), http_only+secure for remember_me - Honeybadger API key: read from env var with fallback - String.to_integer → Integer.parse with fallback for URL params - String.to_atom → whitelist map for HTTP methods - Gaiia webhook: remove secret_len and expected signature from log - Admin API: add rate limiting pipeline - to_atom_keys: per-key fallback instead of all-or-nothing rescue
524 lines
34 KiB
Markdown
524 lines
34 KiB
Markdown
# TowerOps Competitive Analysis
|
|
|
|
**Last Updated**: 2026-03-13
|
|
**Purpose**: Map the ISP/WISP network monitoring and operations tool landscape, identify gaps, and position TowerOps for differentiation.
|
|
|
|
---
|
|
|
|
## Table of Contents
|
|
|
|
1. [LibreNMS](#1-librenms)
|
|
2. [PRTG Network Monitor](#2-prtg-network-monitor)
|
|
3. [Auvik](#3-auvik)
|
|
4. [The Dude (MikroTik)](#4-the-dude-mikrotik)
|
|
5. [Preseem](#5-preseem)
|
|
6. [Sonar](#6-sonar)
|
|
7. [Splynx](#7-splynx)
|
|
8. [Oxidized / RANCID](#8-oxidized--rancid)
|
|
9. [Grafana + Prometheus](#9-grafana--prometheus)
|
|
10. [Competitive Matrix](#competitive-matrix)
|
|
11. [TowerOps Positioning](#towerops-positioning)
|
|
|
|
---
|
|
|
|
## 1. LibreNMS
|
|
|
|
**Website**: librenms.org
|
|
**License**: GPL v3 (open source)
|
|
|
|
### Target Market
|
|
ISPs, WISPs, enterprises, hosting providers, universities — anyone with network infrastructure and the technical staff to self-host.
|
|
|
|
### Pricing Model
|
|
Free and open source. No licensing costs. Community support via Discord and forums. Professional support available through third-party consultants.
|
|
|
|
### Key Strengths
|
|
- **Massive device support**: Auto-discovery via CDP, FDP, LLDP, OSPF, BGP, SNMP, and ARP. 786+ OS profiles.
|
|
- **Billing module**: Built-in bandwidth billing system for ports based on usage or transfer — rare in an open-source monitoring tool.
|
|
- **Distributed polling**: Horizontal scaling with poller groups. Handles large networks (10k+ devices).
|
|
- **Alerting flexibility**: Alerts via email, IRC, Slack, PagerDuty, Telegram, and many more transports.
|
|
- **API access**: Full REST API for automation and integration.
|
|
- **Community**: Large, active community. Active Discord. Regular contributions.
|
|
- **Zero cost**: No per-device, per-sensor, or per-subscriber fees.
|
|
|
|
### Key Weaknesses / Gaps
|
|
- **UX is dated**: Web UI is functional but looks like it was designed in 2015. Not responsive or modern. Navigation is confusing for new users.
|
|
- **RRDtool limitations**: Default time-series backend (RRDtool) has fixed retention periods and is not SQL-queryable. InfluxDB support exists but is less mature.
|
|
- **No multi-tenancy**: Single-organization design. No proper tenant isolation for MSPs or multi-client environments.
|
|
- **Complex setup**: Requires Linux sysadmin skills. Installation, upgrades, and maintenance are manual. No SaaS option.
|
|
- **No agent architecture**: Polling is centralized or distributed-pollers, but no lightweight edge agents for remote sites behind NAT/firewalls.
|
|
- **No subscriber awareness**: Monitors devices and interfaces, not subscribers. No concept of "this customer is having a bad experience."
|
|
- **Config management**: No built-in config backup or change tracking. Requires Oxidized or RANCID alongside it.
|
|
- **Mobile app**: Community-built mobile app exists (by Wavedirect) but not officially maintained.
|
|
|
|
### What TowerOps Can Learn / Differentiate On
|
|
- **Learn**: LibreNMS's device profile coverage (786 OS types) is the gold standard. TowerOps has matched this — maintain parity.
|
|
- **Learn**: Bandwidth billing module is a unique feature that ISPs value. TowerOps integrates with billing platforms (Sonar, Splynx) instead — a different but arguably better approach.
|
|
- **Differentiate**: Modern LiveView UI vs. LibreNMS's aging PHP interface. This is a major selling point.
|
|
- **Differentiate**: Multi-tenancy from day one. LibreNMS can't serve MSPs or multi-org setups cleanly.
|
|
- **Differentiate**: TimescaleDB > RRDtool. SQL-queryable, better compression, better retention policies.
|
|
- **Differentiate**: Agent-based distributed polling for sites behind NAT. LibreNMS requires VPN tunnels or direct access.
|
|
- **Differentiate**: Subscriber-aware monitoring. LibreNMS thinks in devices; TowerOps can think in customers.
|
|
|
|
---
|
|
|
|
## 2. PRTG Network Monitor
|
|
|
|
**Website**: paessler.com/prtg
|
|
**Vendor**: Paessler AG (German company)
|
|
|
|
### Target Market
|
|
Mid-market enterprise IT departments, MSPs, some ISPs. Generalist network monitoring — not ISP/WISP-specific.
|
|
|
|
### Pricing Model
|
|
Subscription-based by sensor count:
|
|
- **PRTG 500** (50 devices): $179/month
|
|
- **PRTG 1000** (100 devices): $325/month
|
|
- **PRTG 2500** (250 devices): $675/month
|
|
- **PRTG 5000** (500 devices): $1,183/month
|
|
- **PRTG 10000** (1000 devices): $1,492/month
|
|
- **Enterprise**: Custom pricing for 10k+ sensors
|
|
- Free tier: 100 sensors (~10 devices)
|
|
|
|
### Key Strengths
|
|
- **Ease of use**: Highly polished UI. Web interface, desktop app, and mobile apps (iOS/Android). Best-in-class for non-technical users.
|
|
- **Sensor-based architecture**: Granular monitoring — each metric is a "sensor." Flexible but can get expensive quickly.
|
|
- **Broad protocol support**: SNMP, WMI, SSH, HTTP, NetFlow, sFlow, jFlow, packet sniffing, and more.
|
|
- **Maps & dashboards**: Built-in map designer with real-time status. Good for NOC displays.
|
|
- **Notifications**: Email, push notifications, HTTP requests, SMS, Slack, Teams, etc.
|
|
- **Windows-native**: Runs on Windows Server. Familiar for Windows-shop IT teams.
|
|
|
|
### Key Weaknesses / Gaps
|
|
- **Expensive at scale**: An ISP with 500 devices needs PRTG 5000 ($1,183/mo) and will burn through sensors quickly. Interface monitoring alone eats 5-10 sensors per device.
|
|
- **Windows-only server**: Core server runs only on Windows. Odd choice for network monitoring in a Linux-heavy ISP world.
|
|
- **Not ISP-aware**: No subscriber mapping, no billing integration, no WISP-specific features. It's a generic IT monitoring tool.
|
|
- **No config management**: No config backup or diff tracking.
|
|
- **Sensor sprawl**: The sensor model means costs escalate unpredictably. Hard to budget for growing networks.
|
|
- **No distributed agents**: Remote probes exist but are Windows-based. No lightweight Linux agent for tower sites.
|
|
- **Limited automation**: Basic auto-discovery but no sophisticated topology mapping or network automation.
|
|
|
|
### What TowerOps Can Learn / Differentiate On
|
|
- **Learn**: PRTG's UX polish is excellent. Maps, dashboards, mobile apps — they invest heavily in usability. TowerOps should aim for this level of polish.
|
|
- **Learn**: The sensor model, while expensive, gives users granular control. Consider offering similar granularity with better pricing.
|
|
- **Differentiate**: Per-device or flat pricing vs. per-sensor. ISPs hate unpredictable costs.
|
|
- **Differentiate**: Linux-native, cloud-native. No Windows dependency.
|
|
- **Differentiate**: ISP/WISP-specific features that PRTG will never build (subscriber awareness, billing integration, tower management).
|
|
|
|
---
|
|
|
|
## 3. Auvik
|
|
|
|
**Website**: auvik.com
|
|
**Type**: Cloud-based SaaS
|
|
|
|
### Target Market
|
|
MSPs (Managed Service Providers) primarily. Some mid-market IT departments. Not ISP/WISP-focused.
|
|
|
|
### Pricing Model
|
|
Quote-based (not publicly listed). Reports suggest:
|
|
- Per-device pricing, tiered by network size
|
|
- Estimated $3-8/device/month depending on volume
|
|
- Multi-tenant by design (MSP model)
|
|
- No free tier; 14-day free trial
|
|
|
|
### Key Strengths
|
|
- **Cloud-first**: No server to manage. Deploy a lightweight collector and go. Full visibility in under an hour.
|
|
- **Auto-discovery & mapping**: Best-in-class automated network topology mapping. Live, auto-updating maps.
|
|
- **64+ pre-configured alerts**: Works out of the box with sensible defaults. Low time-to-value.
|
|
- **Configuration backup**: Automatic config backup with version history and diff comparison. Built-in (no Oxidized needed).
|
|
- **700+ vendor support**: Broad multi-vendor device support out of the box.
|
|
- **NetFlow/traffic analysis**: Deep traffic visibility including encrypted traffic analysis.
|
|
- **MSP integrations**: ConnectWise, Autotask, ServiceNow, IT Glue, PagerDuty, Slack, Teams, etc.
|
|
- **Multi-tenant**: Built for MSPs managing many client networks. Proper tenant isolation.
|
|
- **Modern UX**: Clean, intuitive interface. Junior techs can use it immediately.
|
|
|
|
### Key Weaknesses / Gaps
|
|
- **Not ISP-focused**: Designed for corporate LANs, not carrier networks. No subscriber awareness, no WISP features, no tower management.
|
|
- **Expensive for ISPs**: Per-device pricing adds up fast for ISPs with hundreds/thousands of devices. An ISP with 1000 devices could be paying $3,000-8,000/month.
|
|
- **No SNMP depth**: Good for L2/L3 basics but lacks deep SNMP sensor discovery (temperature, power, optical levels) that ISPs need for tower equipment.
|
|
- **No billing integration**: No connection to ISP billing platforms (Sonar, Splynx, VISP).
|
|
- **Cloud-only**: No self-hosted option. Data sovereignty concerns for some operators.
|
|
- **Limited customization**: Pre-built dashboards are good but custom metrics/sensors are limited compared to PRTG or LibreNMS.
|
|
- **No QoE**: No subscriber quality-of-experience monitoring.
|
|
|
|
### What TowerOps Can Learn / Differentiate On
|
|
- **Learn**: Auvik's time-to-value is exceptional. Deploy collector → see network in minutes. TowerOps's agent model should aim for similar simplicity.
|
|
- **Learn**: Auto-updating topology maps are a killer feature. TowerOps already has LLDP topology discovery — push this further.
|
|
- **Learn**: Built-in config backup eliminates the need for a separate Oxidized/RANCID install. TowerOps should include this.
|
|
- **Differentiate**: ISP/WISP domain expertise vs. Auvik's MSP/corporate focus.
|
|
- **Differentiate**: Deep SNMP sensor discovery (optical, RF, environmental) vs. Auvik's shallow device monitoring.
|
|
- **Differentiate**: Billing platform integrations (Sonar, Splynx, VISP, Gaiia) that Auvik will never build.
|
|
|
|
---
|
|
|
|
## 4. The Dude (MikroTik)
|
|
|
|
**Website**: mikrotik.com (under Tools/Legacy)
|
|
**Vendor**: MikroTik
|
|
|
|
### Target Market
|
|
Small-to-medium WISPs and networks running MikroTik equipment. Hobbyists. Budget-conscious operators.
|
|
|
|
### Pricing Model
|
|
**Free**. Included with MikroTik RouterOS. Server can run on any RouterOS device or as a Windows/Wine application.
|
|
|
|
### Key Strengths
|
|
- **Free**: Zero cost. Hard to compete with free for budget-conscious WISPs.
|
|
- **Deep MikroTik integration**: Direct RouterOS API access. Can monitor and configure MikroTik devices natively without SNMP overhead.
|
|
- **Auto-discovery**: Subnet scanning with service detection. Draws network maps automatically.
|
|
- **Lightweight**: Runs on a MikroTik router itself (CHR or hardware). No separate server needed.
|
|
- **Network maps**: Visual topology with automatic layout. Real-time status overlay.
|
|
- **Multi-protocol**: SNMP, ICMP, DNS, TCP services monitoring.
|
|
- **Widely used in WISP community**: De facto standard for MikroTik-heavy WISPs.
|
|
|
|
### Key Weaknesses / Gaps
|
|
- **Legacy/neglected product**: MikroTik lists it under "Legacy Tools." Development has stagnated. The client is a Windows app (or runs under Wine). No web UI.
|
|
- **MikroTik-centric**: Works great for MikroTik gear but limited for multi-vendor environments (Ubiquiti, Cambium, Cisco, etc.).
|
|
- **No web interface**: Desktop client only. No remote access without VPN/RDP. No mobile app.
|
|
- **Limited alerting**: Basic notifications but no sophisticated alert routing, escalation, or integration with modern tools (PagerDuty, Slack, etc.).
|
|
- **No time-series storage**: Stores current state but lacks historical trending, capacity planning, or long-term data retention.
|
|
- **No API**: No REST API or integration capabilities.
|
|
- **No multi-tenancy**: Single-user design.
|
|
- **Scaling issues**: Performance degrades with large networks (500+ devices). Not designed for ISP-scale operations.
|
|
- **No config backup**: No configuration management features.
|
|
- **No subscriber awareness**: Device-centric only.
|
|
|
|
### What TowerOps Can Learn / Differentiate On
|
|
- **Learn**: The Dude's auto-discovery and visual mapping are beloved by WISPs. Simple, immediate, visual. TowerOps topology should feel this intuitive.
|
|
- **Learn**: Running on the router itself (zero infrastructure) is appealing for tiny WISPs. TowerOps's agent-on-router approach can replicate this for larger operators.
|
|
- **Differentiate**: Everything. The Dude is a legacy tool with no active development. TowerOps is the modern replacement.
|
|
- **Differentiate**: Web UI, multi-vendor, multi-tenant, alerting, API, historical data, subscriber awareness — all things The Dude lacks.
|
|
- **Key opportunity**: Many WISPs have outgrown The Dude but haven't found a replacement that "gets" their world. TowerOps is that replacement.
|
|
|
|
---
|
|
|
|
## 5. Preseem
|
|
|
|
**Website**: preseem.com
|
|
**Type**: Cloud-based SaaS with on-prem appliance for traffic analysis
|
|
|
|
### Target Market
|
|
Regional ISPs and WISPs. Purpose-built for fixed wireless and fiber operators.
|
|
|
|
### Pricing Model
|
|
- **Per-subscriber**: Starting at $0.40/subscriber/month
|
|
- **Minimum**: $300/month
|
|
- **Volume discounts** available
|
|
- **Month-to-month**: No long-term contracts required
|
|
- **Hardware**: Optional pre-loaded appliances or BYO Linux server
|
|
- 30-day free trial
|
|
|
|
### Key Strengths
|
|
- **QoE-focused**: Quality of Experience monitoring is their core differentiator. Measures what subscribers actually experience, not just device uptime.
|
|
- **Traffic shaping**: Adaptive traffic management that reduces congestion and improves subscriber QoE. This is unique — most monitoring tools only observe.
|
|
- **Multi-vendor, multi-access**: Supports fixed wireless (Cambium, Ubiquiti, MikroTik, Baicells, etc.) and fiber. Unified view across technologies.
|
|
- **Billing integrations**: Connects to Sonar, Splynx, Powercode, VISP, BillMax, Azotel, and more. Subscriber-aware from day one.
|
|
- **Churn reduction**: Proven results — customers report 25-30% churn reduction. Direct revenue impact.
|
|
- **Industry recognition**: 5x WISPA Service of the Year winner. Deeply embedded in the WISP community.
|
|
- **Cloud analytics**: Centralized view across all vendors and network elements. No need to log into multiple systems.
|
|
- **Scales to 100+ Gbps**: Handles large traffic volumes.
|
|
|
|
### Key Weaknesses / Gaps
|
|
- **Not a full monitoring platform**: Preseem monitors traffic and QoE but doesn't replace a full NMS. No SNMP polling, no device inventory, no ping monitoring, no alerting for device failures.
|
|
- **Requires inline deployment**: The appliance sits inline or uses port mirroring. Adds a potential point of failure and complexity.
|
|
- **No config management**: No config backup or change tracking.
|
|
- **No topology mapping**: No network topology discovery or visualization.
|
|
- **No on-call/incident management**: No paging, escalation, or incident workflows.
|
|
- **Pricing adds up**: At $0.40/subscriber, a 5,000-subscriber ISP pays $2,000/month just for QoE. On top of their NMS, billing platform, etc.
|
|
- **Complementary, not complete**: ISPs still need LibreNMS/PRTG/etc. alongside Preseem. More tools = more cost and complexity.
|
|
|
|
### What TowerOps Can Learn / Differentiate On
|
|
- **Learn**: Subscriber-aware monitoring is incredibly valuable. Preseem proved the market wants this. TowerOps should integrate subscriber context from billing platforms.
|
|
- **Learn**: QoE metrics (latency, jitter, packet loss per subscriber) matter more than device uptime to ISP operators. Build this into TowerOps.
|
|
- **Learn**: Preseem's billing integrations are a model. They connect to every major ISP billing platform — TowerOps already integrates with Sonar, Splynx, VISP, and Gaiia.
|
|
- **Differentiate**: TowerOps can be the unified platform that includes SNMP monitoring + subscriber awareness + alerting + topology — eliminating the need for Preseem as a separate tool.
|
|
- **Complement or compete**: In the near term, TowerOps + Preseem could work together. Long term, TowerOps could build QoE features and reduce the need for Preseem.
|
|
|
|
---
|
|
|
|
## 6. Sonar
|
|
|
|
**Website**: sonar.software
|
|
**Type**: Cloud-based SaaS
|
|
|
|
### Target Market
|
|
ISPs of all types: cable, fiber, wireless, VoIP, MDU, enterprise. Focused on billing and operations, not monitoring.
|
|
|
|
### Pricing Model
|
|
Per-subscriber tiered pricing with $500/month minimum (with contract):
|
|
- **1-5,000 subscribers**: $1.25/subscriber/month
|
|
- **5,001-10,000**: $1.00/subscriber/month
|
|
- **10,001-50,000**: $0.75/subscriber/month
|
|
- **50,000+**: Enterprise/custom pricing
|
|
|
|
### Key Strengths
|
|
- **Comprehensive ISP operations**: Billing, CRM, ticketing, inventory, field service scheduling, provisioning — all in one platform.
|
|
- **Financial tools**: 25+ pre-built financial reports. Tax management across jurisdictions. Payment processing.
|
|
- **Field service**: Job scheduling, dispatch, GPS tracking, work orders, checklists. Mobile app for field techs.
|
|
- **Network visibility**: Basic monitoring, fault detection, device provisioning, performance tracking.
|
|
- **DataConnect**: Snowflake data warehouse integration for advanced analytics.
|
|
- **Customer portal**: Self-service portal for subscribers.
|
|
- **US/Canada support**: Real humans, no outsourcing, Mon-Fri 8am-6pm CT.
|
|
|
|
### Key Weaknesses / Gaps
|
|
- **Monitoring is basic**: Sonar's network monitoring is rudimentary compared to a dedicated NMS. Basic ICMP/SNMP, no deep sensor discovery, no QoE.
|
|
- **Not a replacement for an NMS**: ISPs using Sonar still need LibreNMS, PRTG, or similar for real network monitoring.
|
|
- **Expensive at scale**: A 5,000-subscriber ISP pays $6,250/month. A 10,000-subscriber ISP pays ~$7,500/month. This is before monitoring, QoE, or other tools.
|
|
- **Cloud-only**: No self-hosted option.
|
|
- **No topology/mapping**: No network topology visualization.
|
|
- **No config management**: No device config backup or change tracking.
|
|
|
|
### What TowerOps Can Learn / Differentiate On
|
|
- **Learn**: Sonar owns the "ISP operations platform" positioning. TowerOps should position as the monitoring/network operations complement, not a billing competitor.
|
|
- **Learn**: Sonar's integration ecosystem is a model. Being the platform that other tools integrate with is powerful.
|
|
- **Differentiate**: TowerOps is the deep network monitoring that Sonar can't do. Sonar does billing; TowerOps does monitoring. They're complementary.
|
|
- **Integration opportunity**: TowerOps's existing Sonar integration should be a first-class feature. Pull subscriber data from Sonar → correlate with network health → give NOC teams subscriber context.
|
|
- **Differentiate**: TowerOps can surface network problems that impact billing (e.g., "these 50 subscribers have been experiencing degraded service for 3 days — expect churn").
|
|
|
|
---
|
|
|
|
## 7. Splynx
|
|
|
|
**Website**: splynx.com
|
|
**Type**: Cloud-hosted or self-hosted SaaS
|
|
|
|
### Target Market
|
|
Small-to-medium ISPs worldwide. Strong presence in Africa, Eastern Europe, Southeast Asia, and Latin America. 1,000+ ISP customers globally.
|
|
|
|
### Pricing Model
|
|
Per-subscriber, subscription-based (prices in USD):
|
|
- Scales with active subscriber count (intervals of 100, entry level up to 400 subscribers)
|
|
- Includes license + maintenance fee (hosting, support, backups)
|
|
- Cloud or on-premise deployment options
|
|
- Add-ons: ACS (TR-069) from $100/month, Whalebone security from $80/month
|
|
- All features included in base license (no feature tiers)
|
|
|
|
### Key Strengths
|
|
- **All-in-one ISP platform**: Billing, RADIUS, TR-069 (ACS), IPAM, CRM, ticketing, field service, inventory — comprehensive.
|
|
- **Networking features**: Built-in RADIUS server, bandwidth management, QoE, IPAM, network monitoring, config backups, change control.
|
|
- **Self-hosted option**: Unlike Sonar (cloud-only), Splynx can be self-hosted. Important for operators in regions with data sovereignty requirements.
|
|
- **Customer portal & mobile app**: White-labeled mobile app for subscriber self-service.
|
|
- **Field tech mobile app**: Dedicated app for technicians with scheduling, work orders, maps.
|
|
- **Global reach**: 1,000+ ISPs worldwide. Strong in developing markets where cost matters.
|
|
- **Fast support**: 85% of tickets resolved in under 60 minutes.
|
|
- **TR-069 / ACS**: CPE device provisioning and management. Increasingly important for fiber ISPs.
|
|
|
|
### Key Weaknesses / Gaps
|
|
- **Monitoring is secondary**: Network monitoring exists but is not the primary focus. Basic compared to dedicated NMS tools.
|
|
- **Not deep SNMP**: Doesn't match LibreNMS or TowerOps for SNMP sensor discovery, vendor-specific MIBs, or environmental monitoring.
|
|
- **Smaller than Sonar in US market**: Sonar dominates the US ISP billing space. Splynx is stronger internationally.
|
|
- **No topology mapping**: No automated network topology discovery or visualization.
|
|
- **No QoE**: No subscriber quality-of-experience monitoring or traffic shaping.
|
|
- **UI less polished than Sonar**: Functional but not as modern as Sonar's interface.
|
|
|
|
### What TowerOps Can Learn / Differentiate On
|
|
- **Learn**: Splynx's all-in-one approach (billing + networking + CRM) appeals to small ISPs who don't want 5 separate tools. TowerOps should integrate deeply enough that it feels like part of the stack.
|
|
- **Learn**: Self-hosted option matters for international markets. TowerOps's agent architecture and potential self-hosted option could appeal to the same market.
|
|
- **Differentiate**: TowerOps provides the deep monitoring that Splynx's built-in monitoring can't match. Position as the "upgrade your monitoring" path for Splynx users.
|
|
- **Integration opportunity**: TowerOps's existing Splynx integration should pull subscriber data, service plans, and customer context to enrich network monitoring.
|
|
|
|
---
|
|
|
|
## 8. Oxidized / RANCID
|
|
|
|
**Website**: github.com/ytti/oxidized (Oxidized), shrubbery.net/rancid (RANCID)
|
|
**License**: Open source
|
|
|
|
### Target Market
|
|
Any network operator needing config backup. ISPs, enterprises, MSPs. Technical users comfortable with CLI tools.
|
|
|
|
### Pricing Model
|
|
**Free** (open source). Self-hosted only.
|
|
|
|
### Key Strengths
|
|
- **Oxidized**: Modern RANCID replacement. Ruby-based. Supports 130+ OS types. Git-backed config storage with blame/diff. REST API. Hooks for notifications (Slack, email). Syslog integration to trigger config fetch on change events.
|
|
- **RANCID**: Battle-tested (20+ years). CVS/SVN-backed. Widely deployed. Simple and reliable.
|
|
- **Config versioning**: Git integration means full history, diff, blame. Know who changed what and when.
|
|
- **Lightweight**: Single-purpose tools that do one thing well.
|
|
- **Event-driven**: Oxidized can react to syslog events (config change detected → immediately fetch new config).
|
|
|
|
### Key Weaknesses / Gaps
|
|
- **Config backup only**: No monitoring, no alerting, no topology, no dashboards. Single-purpose.
|
|
- **Oxidized maintenance concerns**: GitHub repo has a "Maintainer Wanted" notice. Bus factor risk.
|
|
- **No web UI**: CLI/API only. No visual diff viewer out of the box (though tools like oxidized-web exist).
|
|
- **Manual setup**: Requires configuration, cron jobs, and integration work.
|
|
- **No compliance features**: No automated compliance checking against config templates.
|
|
- **RANCID is legacy**: CVS-based, Perl scripts, showing its age. Most operators have migrated to Oxidized.
|
|
|
|
### What TowerOps Can Learn / Differentiate On
|
|
- **Learn**: Config backup with git-backed versioning and diff is table stakes for ISPs. TowerOps already has config change tracking for MikroTik — expand this to all supported vendors.
|
|
- **Learn**: Event-driven config fetch (syslog trigger → immediate backup) is smart. TowerOps agents could do this natively.
|
|
- **Learn**: "Git blame" for config changes (who changed what line) is incredibly valuable for troubleshooting and compliance.
|
|
- **Differentiate**: TowerOps can build config backup into the platform, eliminating a separate Oxidized install. One less tool to maintain.
|
|
- **Differentiate**: Config compliance checking (compare running config against a template, flag drift) — something neither Oxidized nor RANCID does.
|
|
|
|
---
|
|
|
|
## 9. Grafana + Prometheus
|
|
|
|
**Website**: grafana.com, prometheus.io
|
|
**License**: Open source (Grafana AGPL, Prometheus Apache 2.0)
|
|
|
|
### Target Market
|
|
DevOps teams, SREs, cloud-native organizations, and technically sophisticated ISPs who want full control and customization.
|
|
|
|
### Pricing Model
|
|
**Free** (self-hosted open source). Grafana Cloud offers managed hosting:
|
|
- Free tier: 10k metrics, 50GB logs
|
|
- Pro: ~$29/month per active series
|
|
- Enterprise: Custom pricing
|
|
- Most ISPs self-host to avoid costs.
|
|
|
|
### Key Strengths
|
|
- **Visualization**: Grafana is the gold standard for dashboards. Flexible, beautiful, and infinitely customizable.
|
|
- **Prometheus data model**: Labels-based metrics with powerful PromQL query language. Excellent for high-cardinality data.
|
|
- **Ecosystem**: Thousands of exporters (SNMP Exporter, Node Exporter, Blackbox Exporter, etc.). Alertmanager for alerting. Loki for logs.
|
|
- **Cloud-native**: Designed for containerized, microservices architectures. Kubernetes-native.
|
|
- **Scalability**: Horizontal scaling with Thanos, Cortex, or Mimir for long-term storage.
|
|
- **Community**: Massive community. Extensive documentation. Frequent releases.
|
|
- **Flexibility**: Can monitor anything. Network, servers, applications, custom metrics.
|
|
|
|
### Key Weaknesses / Gaps
|
|
- **DIY assembly required**: Not a product — it's a toolkit. ISPs must assemble SNMP Exporter + Prometheus + Alertmanager + Grafana + dashboards. Significant engineering effort.
|
|
- **No auto-discovery**: Prometheus uses static config or service discovery (Consul, DNS, file). No SNMP/LLDP auto-discovery of network devices.
|
|
- **SNMP support is bolted on**: snmp_exporter works but requires SNMP MIB compilation into a generator config. Painful compared to LibreNMS's plug-and-play.
|
|
- **No network awareness**: No topology mapping, no device inventory, no interface table management. You're staring at metrics, not a network.
|
|
- **Alert fatigue**: Alertmanager is powerful but complex to configure. No built-in alert correlation or deduplication intelligence.
|
|
- **No subscriber awareness**: Purely infrastructure metrics. No concept of customers or service impact.
|
|
- **No config management**: No config backup or change tracking.
|
|
- **Maintenance burden**: Self-hosted Prometheus requires capacity planning, storage management, retention tuning. It's a full-time job at scale.
|
|
- **Dashboard sprawl**: Without discipline, you end up with 200 dashboards and no one knows which to look at.
|
|
|
|
### When ISPs Choose Grafana+Prometheus Over a Product
|
|
- They have DevOps/SRE staff who already know the stack
|
|
- They want custom dashboards that no product can provide
|
|
- They're already using it for server/application monitoring and want to extend to network
|
|
- They're philosophically opposed to vendor lock-in
|
|
- They have unique metrics or integrations that products don't support
|
|
- Cost-sensitive and willing to trade engineering time for license fees
|
|
|
|
### What TowerOps Can Learn / Differentiate On
|
|
- **Learn**: Grafana's dashboard flexibility is unmatched. TowerOps should offer embeddable Grafana-style dashboards or a dashboard builder that approaches this flexibility.
|
|
- **Learn**: The Prometheus ecosystem's breadth is incredible. TowerOps should be able to export metrics in Prometheus format for users who want Grafana alongside TowerOps.
|
|
- **Differentiate**: TowerOps is a product, not a toolkit. Auto-discovery, device inventory, subscriber awareness, alerting, topology — all built-in. No assembly required.
|
|
- **Differentiate**: SNMP-first design. TowerOps's 786 OS profiles and vendor-specific MIB support is years of work that Prometheus users would have to replicate.
|
|
- **Differentiate**: ISP domain knowledge. Grafana doesn't know what a tower site is or what subscriber churn means. TowerOps does.
|
|
- **Coexistence strategy**: Offer Prometheus-compatible metric export. Let users who love Grafana use it alongside TowerOps. Don't force them to choose.
|
|
|
|
---
|
|
|
|
## Competitive Matrix
|
|
|
|
| Feature | TowerOps | LibreNMS | PRTG | Auvik | The Dude | Preseem | Sonar | Splynx | Oxidized | Grafana+Prom |
|
|
|---------|----------|----------|------|-------|----------|---------|-------|--------|----------|-------------|
|
|
| **SNMP Monitoring** | ✅ Deep | ✅ Deep | ✅ Deep | ⚠️ Basic | ⚠️ Basic | ❌ | ⚠️ Basic | ⚠️ Basic | ❌ | ⚠️ Bolt-on |
|
|
| **Device Auto-Discovery** | ✅ | ✅ | ✅ | ✅ | ✅ | ❌ | ❌ | ❌ | ❌ | ❌ |
|
|
| **Topology Mapping** | ✅ LLDP | ✅ | ❌ | ✅ Best | ✅ | ❌ | ❌ | ❌ | ❌ | ❌ |
|
|
| **Multi-Tenancy** | ✅ | ❌ | ❌ | ✅ | ❌ | ❌ | ✅ | ✅ | ❌ | ❌ |
|
|
| **Subscriber Awareness** | ✅ | ❌ | ❌ | ❌ | ❌ | ✅ Best | ✅ | ✅ | ❌ | ❌ |
|
|
| **QoE Monitoring** | 🔜 | ❌ | ❌ | ❌ | ❌ | ✅ Best | ❌ | ⚠️ | ❌ | ❌ |
|
|
| **Config Backup** | ✅ MikroTik | ❌ | ❌ | ✅ | ❌ | ❌ | ❌ | ✅ | ✅ Best | ❌ |
|
|
| **Billing Integration** | ✅ | ❌ | ❌ | ❌ | ❌ | ✅ | ✅ Native | ✅ Native | ❌ | ❌ |
|
|
| **Alerting/On-Call** | ✅ | ✅ | ✅ | ✅ | ⚠️ | ❌ | ❌ | ❌ | ❌ | ✅ |
|
|
| **Web UI** | ✅ Modern | ⚠️ Dated | ✅ | ✅ | ❌ Desktop | ✅ | ✅ | ✅ | ❌ CLI | ✅ |
|
|
| **Mobile App** | ✅ | ⚠️ | ✅ | ✅ | ❌ | ✅ | ✅ | ✅ | ❌ | ✅ |
|
|
| **API** | ✅ REST+GraphQL | ✅ REST | ✅ | ✅ | ❌ | ✅ | ✅ | ✅ | ✅ REST | ✅ |
|
|
| **Agent/Distributed** | ✅ | ⚠️ Pollers | ⚠️ Probes | ✅ Collector | ❌ | ✅ Appliance | ❌ | ❌ | ❌ | ✅ Exporters |
|
|
| **Open Source** | ❌ | ✅ | ❌ | ❌ | ✅ Free | ❌ | ❌ | ❌ | ✅ | ✅ |
|
|
| **ISP/WISP Focus** | ✅ | ⚠️ | ❌ | ❌ | ⚠️ MikroTik | ✅ | ✅ | ✅ | ❌ | ❌ |
|
|
|
|
---
|
|
|
|
## TowerOps Positioning
|
|
|
|
### Where TowerOps Fits
|
|
|
|
TowerOps occupies a unique position in the landscape: **the only purpose-built network monitoring platform designed specifically for ISPs and WISPs, with modern architecture and subscriber awareness.**
|
|
|
|
The current landscape forces ISPs to choose between:
|
|
1. **Generic monitoring tools** (LibreNMS, PRTG, Grafana) that don't understand ISP operations
|
|
2. **ISP billing platforms** (Sonar, Splynx) with rudimentary monitoring bolted on
|
|
3. **QoE-only tools** (Preseem) that complement but don't replace an NMS
|
|
4. **Legacy WISP tools** (The Dude) that can't scale
|
|
|
|
TowerOps eliminates the "duct-tape stack" of LibreNMS + Oxidized + Preseem + Grafana that most ISPs cobble together.
|
|
|
|
### Target Customer Profile
|
|
|
|
**Primary**: Regional ISPs and WISPs with 500-50,000 subscribers
|
|
- Running mixed equipment: MikroTik, Ubiquiti, Cambium, Cisco, Juniper
|
|
- Currently using LibreNMS or The Dude and outgrowing them
|
|
- Using Sonar or Splynx for billing and wanting monitoring that integrates
|
|
- Frustrated with managing 3-5 separate tools for network operations
|
|
- Small NOC team (1-5 people) that needs efficiency, not complexity
|
|
|
|
**Secondary**: MSPs managing ISP/WISP clients
|
|
- Need multi-tenant monitoring with ISP-specific features
|
|
- Currently using Auvik but hitting its limits for carrier networks
|
|
|
|
### Unique Value Propositions
|
|
|
|
1. **ISP-native, not ISP-adapted**: Built from day one for ISP/WISP operations. Not a generic monitoring tool with ISP features bolted on. Sites, towers, sectors, subscribers — first-class concepts.
|
|
|
|
2. **Unified platform**: SNMP monitoring + topology mapping + config management + alerting + on-call + subscriber awareness in one product. Replaces LibreNMS + Oxidized + PagerDuty + custom Grafana dashboards.
|
|
|
|
3. **Modern architecture**: Phoenix LiveView for real-time UI, TimescaleDB for time-series (SQL-queryable, not RRDtool), distributed agent architecture for sites behind NAT/firewalls.
|
|
|
|
4. **Subscriber-aware monitoring**: Pull subscriber data from Sonar/Splynx/VISP/Gaiia and correlate with network health. "These 47 subscribers on Tower-12 Sector-3 are experiencing degraded service" — not just "interface utilization is high."
|
|
|
|
5. **Agent-based distributed monitoring**: Lightweight agents deployed at tower sites, headends, or POPs. Monitor behind NAT, through firewalls, with local data collection and central aggregation. No VPN tunnels or direct SNMP access needed.
|
|
|
|
6. **Deep SNMP + fast polling**: 786 OS profiles with vendor-specific sensor discovery. 60-second default polling (vs LibreNMS's 5-minute). Environmental monitoring (temperature, power, optical levels) that Auvik and billing platforms can't do.
|
|
|
|
7. **Billing platform integrations**: First-class integrations with Sonar, Splynx, VISP, and Gaiia. Not just API calls — bidirectional sync that enriches both platforms.
|
|
|
|
### Features Competitors Lack
|
|
|
|
| Feature | Why It Matters | Who's Missing It |
|
|
|---------|---------------|-----------------|
|
|
| **Site/tower hierarchy** | ISPs think in sites, not flat device lists | LibreNMS, PRTG, Auvik, Grafana |
|
|
| **Agent-based polling behind NAT** | Tower sites often lack direct connectivity | LibreNMS, PRTG, The Dude |
|
|
| **Subscriber ↔ device correlation** | "Which customers are affected by this outage?" | LibreNMS, PRTG, Auvik, Grafana |
|
|
| **Built-in on-call/escalation** | No separate PagerDuty/Opsgenie needed | LibreNMS, Preseem, Sonar, Splynx |
|
|
| **Config backup + deep monitoring** | Eliminates Oxidized as separate tool | LibreNMS, PRTG, Preseem |
|
|
| **Multi-tenant with ISP features** | MSPs managing WISPs need both | LibreNMS, The Dude, Preseem |
|
|
| **60-second polling** | Faster fault detection, better capacity data | LibreNMS (5min default) |
|
|
| **TimescaleDB time-series** | SQL-queryable historical data, continuous aggregates | LibreNMS (RRDtool), The Dude (none) |
|
|
| **Capacity planning** | Interface utilization trends and forecasting | The Dude, Preseem, billing platforms |
|
|
|
|
### Competitive Positioning Summary
|
|
|
|
```
|
|
ISP/WISP Specific
|
|
↑
|
|
|
|
|
Preseem | TowerOps ★
|
|
(QoE only) | (Full NMS + ISP features)
|
|
|
|
|
Sonar/Splynx |
|
|
(Billing + basic |
|
|
monitoring) |
|
|
|
|
|
Generic ←──────────────┼──────────────→ Deep Monitoring
|
|
|
|
|
The Dude | LibreNMS
|
|
(Legacy) | (Open source NMS)
|
|
|
|
|
Auvik | PRTG
|
|
(MSP focus) | (Enterprise IT)
|
|
|
|
|
Grafana+Prometheus
|
|
(DIY toolkit)
|
|
|
|
|
Generic IT
|
|
```
|
|
|
|
### Go-To-Market Priorities
|
|
|
|
1. **Migrate The Dude users**: Largest addressable audience of frustrated WISP operators. Low switching cost (The Dude is free but limited). Marketing message: "You've outgrown The Dude."
|
|
|
|
2. **Replace LibreNMS for ISPs**: Technical operators who've invested in LibreNMS but want modern UX, subscriber awareness, and less maintenance. Message: "Everything LibreNMS does, plus everything it doesn't."
|
|
|
|
3. **Complement Sonar/Splynx**: Don't compete with billing platforms. Be the monitoring layer they recommend. Message: "The monitoring platform that actually talks to your billing system."
|
|
|
|
4. **Displace the duct-tape stack**: ISPs running LibreNMS + Oxidized + Grafana + PagerDuty + Preseem. Message: "One platform instead of five."
|