Skip to content

Latest commit

 

History

History
292 lines (220 loc) · 10.4 KB

host-management.md

File metadata and controls

292 lines (220 loc) · 10.4 KB
Raw

Host Management with OpenBMC

This document describes the host-management interfaces of the OpenBMC object structure, accessible over REST.

Inventory

The system inventory structure is under the /xyz/openbmc_project/inventory hierarchy.

In OpenBMC the inventory is represented as a path which is hierarchical to the physical system topology. Items in the inventory are referred to as inventory items and are not necessarily FRUs (field-replaceable units). If the system contains one chassis, a motherboard, and a CPU on the motherboard, then the path to that inventory item would be:

inventory/system/chassis0/motherboard0/cpu0

The properties associated with an inventory item are specific to that item. Some common properties are:

  • Version: A code version associated with this item.
  • Present: Indicates whether this item is present in the system (True/False).
  • Functional: Indicates whether this item is functioning in the system (True/False).

The usual list and enumerate REST queries allow the system inventory structure to be accessed. For example, to enumerate all inventory items and their properties:

curl -b cjar -k https://${bmc}/xyz/openbmc_project/inventory/enumerate

To list the properties of one item:

curl -b cjar -k https://${bmc}/xyz/openbmc_project/inventory/system/chassis/motherboard

Sensors

The system sensor structure is under the /xyz/openbmc_project/sensors hierarchy.

This interface allows monitoring of system attributes like temperature or altitude, and are represented similar to the inventory, by object paths under the top-level sensors object name. The path categorizes the sensor and shows what the sensor represents, but does not necessarily represent the physical topology of the system.

For example, all temperature sensors are under sensors/temperature. CPU temperature sensors would be sensors/temperature/cpu[n].

These are some common properties:

  • Value: Current value of the sensor
  • Unit: Unit of the value and "Critical" and "Warning" values
  • Scale: The scale of the value and "Critical" and "Warning" values
  • CriticalHigh & CriticalLow: Sensor device upper/lower critical threshold bound
  • CriticalAlarmHigh & CriticalAlarmLow: True if the sensor has exceeded the critical threshold bound
  • WarningHigh & WarningLow: Sensor device upper/lower warning threshold bound
  • WarningAlarmHigh & WarningAlarmLow: True if the sensor has exceeded the warning threshold bound

A temperature sensor might look like:

curl -b cjar -k https://${bmc}/xyz/openbmc_project/sensors/temperature/pcie
{
  "data": {
    "CriticalAlarmHigh": 0,
    "CriticalAlarmLow": 0,
    "CriticalHigh": 70000,
    "CriticalLow": 0,
    "Scale": -3,
    "Unit": "xyz.openbmc_project.Sensor.Value.Unit.DegreesC",
    "Value": 28187,
    "WarningAlarmHigh": 0,
    "WarningAlarmLow": 0,
    "WarningHigh": 60000,
    "WarningLow": 0
  },
  "message": "200 OK",
  "status": "ok"
}

Note the value of this sensor is 28.187C (28187 * 10^-3).

Unlike IPMI, there are no "functional" sensors in OpenBMC; functional states are represented in the inventory.

To enumerate all sensors in the system:

curl -b cjar -k https://${bmc}/xyz/openbmc_project/sensors/enumerate

List properties of one inventory item:

curl -b cjar -k https://${bmc}/xyz/openbmc_project/sensors/temperature/ambient

Event Logs

The event log structure is under the /xyz/openbmc_project/logging/entry hierarchy. Each event is a separate object under this structure, referenced by number.

BMC and host firmware on POWER-based servers can report event logs to the BMC. Typically, these event logs are reported in cases where host firmware cannot start the OS, or cannot reliably log to the OS.

The properties associated with an event log are as follows:

  • Message: The type of event log (e.g. "xyz.openbmc_project.Inventory.Error.NotPresent").
  • Resolved : Indicates whether the event has been resolved.
  • Severity: The level of problem ("Info", "Error", etc.).
  • Timestamp: The date of the event log in epoch time.
  • associations: A URI to the failing inventory part.

To list all reported event logs:

$ curl -b cjar -k https://${bmc}/xyz/openbmc_project/logging/entry/
{
  "data": [
    "/xyz/openbmc_project/logging/entry/3",
    "/xyz/openbmc_project/logging/entry/2",
    "/xyz/openbmc_project/logging/entry/1",
    "/xyz/openbmc_project/logging/entry/7",
    "/xyz/openbmc_project/logging/entry/6",
    "/xyz/openbmc_project/logging/entry/5",
    "/xyz/openbmc_project/logging/entry/4"
  ],
  "message": "200 OK",
  "status": "ok"
}

To read a specific event log:

$ curl -b cjar -k https://${bmc}/xyz/openbmc_project/logging/entry/1
{
  "data": {
    "AdditionalData": [
      "CALLOUT_INVENTORY_PATH=/xyz/openbmc_project/inventory/system/chassis/powersupply0",
      "_PID=1136"
    ],
    "Id": 1,
    "Message": "xyz.openbmc_project.Inventory.Error.NotPresent",
    "Resolved": 0,
    "Severity": "xyz.openbmc_project.Logging.Entry.Level.Error",
    "Timestamp": 1512154612660,
    "associations": [
      [
        "callout",
        "fault",
        "/xyz/openbmc_project/inventory/system/chassis/powersupply0"
      ]
    ]
  },
  "message": "200 OK",
  "status": "ok"
}

To delete an event log (log 1 in this example), call the delete method on the event:

curl -b cjar -k -H "Content-Type: application/json" -X POST \
    -d '{"data" : []}' \
    https://${bmc}/xyz/openbmc_project/logging/entry/1/action/Delete

To clear all event logs, call the top-level deleteAll method:

curl -b cjar -k -H "Content-Type: application/json" -X POST \
    -d '{"data" : []}' \
    https://${bmc}/xyz/openbmc_project/logging/action/deleteAll

Host Boot Options

With OpenBMC, the Host boot options are stored as D-Bus properties under the control/host0/boot path. Properties include BootMode and BootSource.

Host State Control

The host can be controlled through the host object. The object implements a number of actions including power on and power off. These correspond to the IPMI power on and power off commands.

Assuming you have logged in, the following will power on the host:

curl -c cjar -b cjar -k -H "Content-Type: application/json" -X PUT \
  -d '{"data": "xyz.openbmc_project.State.Host.Transition.On"}' \
  https://${bmc}/xyz/openbmc_project/state/host0/attr/RequestedHostTransition

To power off the host:

curl -c cjar -b cjar -k -H "Content-Type: application/json" -X PUT \
  -d '{"data": "xyz.openbmc_project.State.Host.Transition.Off"}' \
  https://${bmc}/xyz/openbmc_project/state/host0/attr/RequestedHostTransition

To issue a hard power off (accomplished by powering off the chassis):

curl -c cjar -b cjar -k -H "Content-Type: application/json" -X PUT \
  -d '{"data": "xyz.openbmc_project.State.Chassis.Transition.Off"}' \
  https://${bmc}/xyz/openbmc_project/state/chassis0/attr/RequestedPowerTransition

To reboot the host:

curl -c cjar -b cjar -k -H "Content-Type: application/json" -X PUT \
  -d '{"data": "xyz.openbmc_project.State.Host.Transition.Reboot"}' \
  https://${bmc}/xyz/openbmc_project/state/host0/attr/RequestedHostTransition

More information about Host State Management can be found here: https://github.com/openbmc/phosphor-dbus-interfaces/tree/master/xyz/openbmc_project/State

Host Clear GARD

On OpenPOWER systems, the host maintains a record of bad or non-working components on the GARD partition. This record is referenced by the host on subsequent boots to determine which parts should be ignored.

The BMC implements a function that simply clears this partition. This function can be called as follows:

  • Method 1: From the BMC command line:

    busctl call org.open_power.Software.Host.Updater \
  /org/open_power/control/gard \
  xyz.openbmc_project.Common.FactoryReset Reset

  • Method 2: Using the REST API:

    curl -b cjar -k -H 'Content-Type: application/json' -X POST -d '{"data":[]}' https://${bmc}/org/open_power/control/gard/action/Reset

Implementation: https://github.com/openbmc/openpower-pnor-code-mgmt

Host Watchdog

The host watchdog service is responsible for ensuring the host starts and boots within a reasonable time. On host start, the watchdog is started and it is expected that the host will ping the watchdog via the inband interface periodically as it boots. If the host fails to ping the watchdog within the timeout then the host watchdog will start a systemd target to go to the quiesce target. System settings will then determine the recovery behavior from that state, for example, attempting to reboot the system.

The host watchdog utilizes the generic phosphor-watchdog repository. The host watchdog service provides 2 files as configuration options into phosphor-watchdog:

/lib/systemd/system/[email protected]/poweron.conf
/etc/default/obmc/watchdog/poweron

poweron.conf contains the "Conflicts" relationships to ensure the watchdog service is stopped at the correct times. poweron contains the required information for phosphor-watchdog (more information on these can be found in the phosphor-watchdog repository).

The 2 service files involved with the host watchdog are:

[email protected]
[email protected]

phosphor-watchdog@poweron starts the host watchdog service and obmc-enable-host-watchdog starts the watchdog timer. Both are run as a part of the [email protected]. Service dependencies ensure the service is started before the enable is called.

The default watchdog timeout can be found within the dbus interface specification (Interval property).

The host controls the watchdog timeout and enable/disable once it starts.