FoundryNet / Error codes / MTConnect UNAVAILABLE
MTConnect error code

UNAVAILABLE — Data item value is UNAVAILABLE (not an error)

Not an error code at all — it is a legitimate data value meaning the agent currently has no reading for that item. Treating it as a number is the most common MTConnect integration bug.

Start free MTConnect → canonical

Diagnosis

When you hit this


Resolution

How to fix it

  1. Never coerce UNAVAILABLE to zero
    A spindle speed of UNAVAILABLE is not zero rpm. Coercing it produces false idle time and corrupts every average computed downstream.
  2. Treat it as absent
    Exclude the item from that sample rather than substituting a value. Forge reports such tags as unresolved rather than inventing a reading.
  3. Check the adapter if it persists
    An item permanently UNAVAILABLE usually means the adapter never maps it, not that the machine lacks the capability.

Normalization

What Forge does about it

A request that fails on some points and succeeds on others still normalizes. The tags that arrived resolve to canonical fields as usual; the ones that did not come back are listed in unresolved_tags and excluded from coverage_pct, so a partial read reports as partial instead of quietly looking complete.


Connect

Normalize MTConnect telemetry

One MCP endpoint serves every protocol adapter. 16,908 curated mappings, 366 canonical fields.

# normalize telemetry from MTConnect — and 13 other protocols
claude mcp add --transport http forge https://mcp.foundrynet.io/mcp \
  --header 'Authorization: Bearer YOUR_FORGE_KEY'
✓ forge connected — 32 tools available

Related

Keep reading

Other MTConnect codes: OUT_OF_RANGE NO_DEVICE UNAUTHORIZED

Equipment that speaks MTConnect: DMG Mori Fanuc Haas Heidenhain Mazak Okuma Siemens Sinumerik Makino Yaskawa Motoman Bridgeport

MTConnect protocol hub

Other protocols: Modbus OPC UA MQTT / Sparkplug B FANUC FOCAS EtherNet/IP PROFINET / S7 BACnet/IP J1939 (CAN) ROS 2 (DDS) SNMP MAVLink HTTP / REST CSV ingest

All error codes · All OEMs · All protocols

Industrial AI fails at the data layer. The Machine Agent explains why, weekly. Read it →
Get the thesis behind this infrastructure — The Machine Agent, weekly.