Skip to content

Latest commit

 

History

History
91 lines (62 loc) · 2.57 KB

monitor-hw.md

File metadata and controls

91 lines (62 loc) · 2.57 KB

Monitor hardware

monitor-hw is a monitoring daemon that exports Redfish data for Prometheus.

Synopsis

$ monitor-hw [--listen=<address>] [--interval=<interval>] [vendor-specific options...]

Description

monitor-hw exposes hardware status obtained through Redfish API as Prometheus metrics.

When monitor-hw is invoked, it starts a background routine to gather Redfish data of the server where it runs. This routine periodically traverses Redfish data from the BMC via HTTPS, and stores them in memory.

monitor-hw also starts an HTTP server to export data to Prometheus. When its /metrics path is accessed, it reads the stored Redfish data and reports them as the metrics of the server for Prometheus.

As monitor-hw uses the standard API of Redfish instead of dedicated tools like omreport, it can support multiple types of servers from multiple vendors.

Vendor specific behaviors

Dell actions

monitor-hw invokes instsvcdrv-helper for racadm.

monitor-hw periodically resets iDRAC because it occasionally hangs.

QEMU actions

monitor-hw behaves as a mock server.

Options

--listen=<address> specifies the address, i.e. host and TCP port, where monitor-hw listens for metrics retrieval. This accepts the same forms of <host>:<port> and :<port> as Addr of http.Server. The default is :9105.

--interval=<interval> specifies the interval of Redfish data traversal in seconds. The default is 60 seconds. This interval means the time between the end of a certain traversal operation and the beginning of the next one, so the observed interval of metrics update will be somewhat longer than this interval.

Dell options

--reset-interval specifies the interval of resetting iDRAC in hours. The default is 24 hours.

--no-reset specifies a file name watched by monitor-hw. While the file exists, monitor-hw does not reset iDRAC.

Configuration files

monitor-hw reads the host and the user information of the BMC from the configuration files. See the description of configuration files for details.

Internals

monitor-hw first detects the hardware type of the server where it runs. It traverses and interprets Redfish data according to the collection rule for that hardware type. Collection rules are compiled from YAML files under redfish/rules. See the description of collection rules for details.