Routers, switches and firewalls
Read a system uptime, an interface's traffic counter or its operational status, or a processor load figure - the values that tell you a piece of network gear is alive and doing its job.
SNMP monitoring
HostTracker's SNMP monitoring tool reads one numeric OID from a router, switch, firewall, UPS unit or printer over SNMP v1, v2c or v3, on the schedule you set - keeps every reading in the check history and alerts you the moment the device stops answering.
How HostTracker reads one SNMP OID, from the request to the alert
Most hardware on a network cannot run a collector script or a browser-based check, but almost all of it already speaks SNMP. This kind of SNMP network monitoring extends HostTracker beyond websites and servers to the equipment underneath them.
Read a system uptime, an interface's traffic counter or its operational status, or a processor load figure - the values that tell you a piece of network gear is alive and doing its job.
A UPS reports its own battery status and its estimated runtime remaining over SNMP. Watching that number is how you find out the mains failed before the battery runs out.
A printer's toner or ink level and a NAS's storage used are both plain SNMP values, from the Printer-MIB and the same host-resources tree servers use.
Find the OID, open the device to HostTracker's check network, then create the monitor.
An OID is a dotted-decimal address, not a name - HostTracker reads it exactly as written. If you do not already know the one you need, walk the device first:
snmpwalk · the system group
snmpwalk -v2c -c public 192.0.2.10 1.3.6.1.2.1.1That lists the system group on a v2c device using the community string public - swap in your own community string, host and, for v3, the security credentials instead of -c. Standard MIB-2 and common vendor OIDs to start from:
1.3.6.1.2.1.1.3.01.3.6.1.2.1.2.2.1.10.<ifIndex>1.3.6.1.2.1.2.2.1.8.<ifIndex>1.3.6.1.2.1.25.3.3.1.2.<n>1.3.6.1.2.1.25.2.3.1.6.<n>1.3.6.1.2.1.33.1.2.3.01.3.6.1.2.1.33.1.2.1.01.3.6.1.2.1.43.11.1.1.9.1.1All three answer a Get or GetNext request the same way - what differs is how the device decides to trust the request.
On authPriv, both keys need at least 8 characters. Reach for v3 on anything you cannot put entirely behind a trusted network - v1 and v2c put the community string on the wire unencrypted, so anyone who can see the traffic can see it too.
Honest about the trade-off, and what you get for it.
SNMP monitoring runs from HostTracker's fixed internal check network, not the 300+ checkpoints used for uptime and ping checks - so there is no location to pick.
Most teams expose the device to HostTracker through a firewall rule scoped to that network, or by terminating a VPN endpoint the check dials into.
Each poll is stored alongside your other monitors' results, so an interface counter or a supply level has a trend, not just a last value.
Email, SMS, voice call, Slack, Telegram, webhooks and more - the same alert channels every HostTracker monitor uses.
Websites, servers, APIs, certificates. One check type per page, the same locations, alerts and reports behind all of them.
Trusted by teams at
Every chapter opens in place, so the page stays short.
An OID is an address into a tree, and every device publishes a different slice of it. MIB-2 - the base tree defined in RFC 1213 - is where sysUpTime and the interface counters live, and almost every SNMP device answers it whether it is a router, a switch or a Linux box with an agent installed. Past that, a manufacturer's own MIB adds device-specific values: a UPS-MIB for battery gear, a Printer-MIB for supply levels, and vendor MIBs of their own for things like temperature sensors or fan speed.
The fastest way to find the right number on a specific device is to ask it. snmpwalk lists every OID a
device is willing to answer under a starting point; snmpget reads one OID you already know:
snmpwalk -v2c -c public 192.0.2.10 1.3.6.1.2.1.1
snmpget -v2c -c public 192.0.2.10 1.3.6.1.2.1.1.3.0
Whatever you find, HostTracker needs the numeric form - 1.3.6.1.2.1.1.3.0, never the symbolic name
sysUpTime.0 a MIB browser might show you next to it. Copy the number, not the label.
SNMP v1 and v2c authenticate with nothing more than a community string, sent on the wire exactly as typed - a password anyone watching the traffic can read. SNMP v3 replaces that with a security name and a security level you choose per credential:
Both the authentication key and the privacy key need to be at least 8 characters. If the device supports it, authPriv with SHA-256 and AES-256 is the strongest combination on offer; older embedded gear sometimes only speaks MD5 and DES, which is still worth using over a plain v1 or v2c community string.
Being straightforward about where this sits today: an SNMP check reads and records the value the device reports. It does not evaluate a rule against that value the way an HTTP, Database or Counter check does - there is no "alert when this number crosses X" on an SNMP monitor yet. What it does alert on is the device itself: when the request stops getting a reply, an alert goes out to your configured contacts - email, SMS, voice call, Slack, Telegram, webhooks and more - the same channels every HostTracker check uses.
That still covers the case that matters most for hardware: knowing a device fell off the network before someone notices the hard way. And because every reading is kept with the check history, you can open a router's uptime counter or a UPS's battery status and read the trend yourself, even without an automatic rule watching it. For a value that does need a threshold and a debounce - CPU, memory, disk space, a port's connect time - a server load monitor against a small collector endpoint gives you that today.
It sends a single SNMP Get or GetNext request for one OID you choose - a router's uptime, a UPS's battery status, a printer's toner level - on the schedule you set, over SNMP v1, v2c or v3. Every reading joins the check's history, and an alert goes out through your configured channels the moment the device stops answering.
An OID (object identifier) is a dotted-decimal address into a device's SNMP tree, like 1.3.6.1.2.1.1.3.0 for system uptime. HostTracker reads a numeric OID only - it does not accept a symbolic name such as sysUpTime.0. Run snmpwalk or snmpget against the device to see what it exposes, check the vendor's MIB documentation, or start from the standard MIB-2 and vendor-MIB OIDs in the table on this page.
SNMP v1 and v2c with a community string, and SNMP v3 with a security name, one of the noAuthNoPriv, authNoPriv or authPriv security levels, MD5, SHA-1 or SHA-256 authentication, and DES or AES-128, AES-192 or AES-256 privacy. v3 is the one to reach for on anything you cannot put behind a trusted network, since v1 and v2c send the community string in the clear.
No, and you should not. The check runs from HostTracker's fixed internal check network, not the public checkpoint fleet, so the device only needs to accept SNMP from that network specifically. Most teams do this with a firewall allow-rule scoped to HostTracker's check network, or a VPN endpoint the check dials into, rather than opening SNMP to the internet at large.
It reads a numeric value from the OID you give it. If the device answers with something that is not a number, the check still confirms the device replied, but there is no value to chart or a rule to evaluate against it - treat that as a reachability signal, not a reading. For a value you want to see charted over time, point the check at a numeric OID such as an interface counter, a processor load figure or a UPS runtime estimate.
As often as you set the monitor's interval, the same range HostTracker's other checks offer. A router's uptime counter is fine polled every 30 minutes; a UPS you actually want a warning from is worth checking every minute.
No. Website and ping checks run from that public fleet; an SNMP monitor runs from HostTracker's own fixed internal check network instead, because the devices it watches are rarely internet-facing and there is nothing to compare between locations for one router's uptime counter. That is also why an SNMP monitor has no location picker.
SNMP is one of HostTracker's monitor types, available on the plans that include it rather than as a paid add-on to another monitor. Check which check types your plan includes on the pricing page, or start a 30-day free trial - no credit card required - to try it against your own equipment.
Add an SNMP monitor for the routers, switches, UPS units and printers that cannot run a collector, alongside your website, server and database checks.
Part of HostTracker's website monitoring software.