lib/dump_utils.robot

*** Settings ***
Documentation  This module provides general keywords for dump.

Library         bmc_ssh_utils.py

*** Variables ***

*** Keywords ***

Create User Initiated Dump
    [Documentation]  Generate user initiated dump and return
    ...  the dump id number (e.g., "5").  Optionally return EMPTY
    ...  if out of dump space.
    [Arguments]   ${check_out_of_space}=${False}

    # Description of Argument(s):
    # check_out_of_space   If ${False}, a dump will be created and
    #                      its dump_id will be returned.
    #                      If ${True}, either the dump_id will be
    #                      returned, or the value ${EMPTY} will be
    #                      returned if out of dump space was
    #                      detected when creating the dump.

    ${resp}=  OpenBMC Get Request  ${REST_DUMP_URI}
    Run Keyword If  '${resp.status_code}' == '${HTTP_NOT_FOUND}'
    ...  Set Global Variable  ${REST_DUMP_URI}  /xyz/openbmc_project/dump/

    ${data}=  Create Dictionary  data=@{EMPTY}
    ${resp}=  OpenBMC Post Request
    ...  ${REST_DUMP_URI}action/CreateDump  data=${data}  quiet=${1}

    Run Keyword If  '${check_out_of_space}' == '${False}'
    ...      Run Keyword And Return  Get The Dump Id  ${resp}
    ...  ELSE
    ...      Run Keyword And Return  Check For Too Many Dumps  ${resp}


Get The Dump Id
    [Documentation]  Wait for the dump to be created. Return the
    ...  dump id number (e.g., "5").
    [Arguments]  ${resp}

    # Description of Argument(s):
    # resp   Response object from action/Create Dump attempt.
    #        Example object:
    #        {
    #           "data": 5,
    #           "message": "200 OK",
    #           "status": "ok"
    #        },
    #        The "data" field conveys the id number of the created dump.

    Should Be Equal As Strings  ${resp.status_code}  ${HTTP_OK}

    Run Keyword If  ${resp.json()["data"]} == ${None}
    ...  Fail  Dump id returned null.

    ${dump_id}=  Set Variable  ${json["data"]}

    Wait Until Keyword Succeeds  3 min  15 sec  Check Dump Existence
    ...  ${dump_id}

    [Return]  ${dump_id}


Check For Too Many Dumps
    [Documentation]  Return the dump_id number, or return ${EMPTY} if dump
    ...  creation failed due to too many dumps.
    [Arguments]  ${resp}

    # Description of Argument(s):
    # resp   Response object from action/Create Dump attempt.
    #        Example object if there are too many dumps:
    #       {
    #           "data": {
    #               "description": "xyz.openbmc_project.Dump.Create.Error.QuotaExceeded"
    #           },
    #           "message": "Dump not captured due to a cap.",
    #           "status": "error"
    #       }

    # If dump was created normally, return the dump_id number.
    Run Keyword If  '${resp.status_code}' == '${HTTP_OK}'
    ...  Run Keyword And Return  Get The Dump Id  ${resp}

    ${exception}=  Set Variable  ${resp.json()["message"]}
    ${at_capacity}=  Set Variable  Dump not captured due to a cap
    ${too_many_dumps}=  Evaluate  $at_capacity in $exception
    Printn
    Rprint Vars   exception  too_many_dumps
    # If there are too many dumps, return ${EMPTY}, otherwise Fail.
    ${status}=  Run Keyword If  ${too_many_dumps}  Set Variable  ${EMPTY}
    ...  ELSE  Fail  msg=${exception}.

    [Return]  ${status}


Verify No Dump In Progress
    [Documentation]  Verify no dump in progress.

    ${dump_progress}  ${stderr}  ${rc}=  BMC Execute Command  ls /tmp
    Should Not Contain  ${dump_progress}  obmcdump


Check Dump Existence
    [Documentation]  Verify if given dump exist.
    [Arguments]  ${dump_id}

    # Description of Argument(s):
    # dump_id  An integer value that identifies a particular dump
    #          object(e.g. 1, 3, 5).

    ${resp}=  OpenBMC Get Request  ${REST_DUMP_URI}
    Run Keyword If  '${resp.status_code}' == '${HTTP_NOT_FOUND}'
    ...  Set Global Variable  ${DUMP_ENTRY_URI}  /xyz/openbmc_project/dump/entry/

    ${resp}=  OpenBMC Get Request  ${DUMP_ENTRY_URI}${dump_id}
    Should Be Equal As Strings  ${resp.status_code}  ${HTTP_OK}


Delete BMC Dump
    [Documentation]  Deletes a given bmc dump.
    [Arguments]  ${dump_id}

    # Description of Argument(s):
    # dump_id  An integer value that identifies a particular dump (e.g. 1, 3).

    ${resp}=  OpenBMC Get Request  ${REST_DUMP_URI}
    Run Keyword If  '${resp.status_code}' == '${HTTP_NOT_FOUND}'
    ...  Set Global Variable  ${DUMP_ENTRY_URI}  /xyz/openbmc_project/dump/entry/

    ${args}=  Set Variable   {"data": []}
    ${resp}=  OpenBMC Post Request
    ...  ${DUMP_ENTRY_URI}${dump_id}/action/Delete  data=${args}

    Should Be Equal As Strings  ${resp.status_code}  ${HTTP_OK}

Delete All Dumps
    [Documentation]  Delete all dumps.

    ${resp}=  OpenBMC Get Request  ${REST_DUMP_URI}
    Run Keyword If  '${resp.status_code}' == '${HTTP_NOT_FOUND}'
    ...  Set Global Variable  ${DUMP_ENTRY_URI}  /xyz/openbmc_project/dump/entry/

    # Check if dump entries exist, if not return.
    ${resp}=  OpenBMC Get Request  ${DUMP_ENTRY_URI}list  quiet=${1}
    Return From Keyword If  ${resp.status_code} == ${HTTP_NOT_FOUND}

    # Get the list of dump entries and delete them all.
    ${dump_entries}=  Get URL List  ${DUMP_ENTRY_URI}
    FOR  ${entry}  IN  @{dump_entries}
        ${dump_id}=  Fetch From Right  ${entry}  /
        Delete BMC Dump  ${dump_id}
    END


Redfish Delete BMC Dump
    [Documentation]  Deletes a given BMC dump via Redfish..
    [Arguments]  ${dump_id}

    # Description of Argument(s):
    # dump_id  An integer value that identifies a particular dump (e.g. 1, 3).

    Redfish.Delete  /redfish/v1/Managers/bmc/LogServices/Dump/Entries/${dump_id}


Redfish Delete All BMC Dumps
    [Documentation]  Delete all BMC dumps via Redfish.

    # Check if dump entries exist, if not return.
    ${resp}=  Redfish.Get  /redfish/v1/Managers/bmc/LogServices/Dump/Entries
    Return From Keyword If  ${resp.dict["Members@odata.count"]} == ${0}

    Redfish.Post  /redfish/v1/Managers/bmc/LogServices/Dump/Actions/LogService.ClearLog


Redfish Delete All System Dumps
    [Documentation]  Delete all system  dumps via Redfish.

    Redfish.Post  /redfish/v1/Systems/system/LogServices/Dump/Actions/LogService.ClearLog


Delete All BMC Dump
    [Documentation]  Delete all BMC dump entries using "DeleteAll" interface.

    ${resp}=  OpenBMC Get Request  ${REST_DUMP_URI}
    Run Keyword If  '${resp.status_code}' == '${HTTP_NOT_FOUND}'
    ...  Set Global Variable  ${REST_DUMP_URI}  /xyz/openbmc_project/dump/

    ${args}=  Set Variable   {"data": []}
    ${resp}=  Openbmc Post Request  ${REST_DUMP_URI}action/DeleteAll  data=${args}
    Should Be Equal As Strings  ${resp.status_code}  ${HTTP_OK}

Dump Should Not Exist
    [Documentation]  Verify that BMC dumps do not exist.

    ${resp}=  OpenBMC Get Request  ${REST_DUMP_URI}
    Run Keyword If  '${resp.status_code}' == '${HTTP_NOT_FOUND}'
    ...  Set Global Variable  ${DUMP_ENTRY_URI}  /xyz/openbmc_project/dump/entry/

    ${resp}=  OpenBMC Get Request  ${DUMP_ENTRY_URI}list  quiet=${1}
    Should Be Equal As Strings  ${resp.status_code}  ${HTTP_NOT_FOUND}

Check Existence Of BMC Dump File
    [Documentation]  Verify existence of BMC dump file.
    [Arguments]  ${dump_id}

    # Description of argument(s):
    # dump_id  BMC dump identifier

    ${dump_check_cmd}=  Set Variable
    ...  ls /var/lib/phosphor-debug-collector/dumps

    # Output of sample BMC Execute command with '2' as dump id is as follows
    # ls /var/lib/phosphor-debug-collector/dumps/2
    # obmcdump_2_XXXXXXXXXX.tar.xz
    ${file_there}  ${stderr}  ${rc}=  BMC Execute Command
    ...  ${dump_check_cmd}/${dump_id}
    Should End With  ${file_there}  tar.xz  msg=BMC dump file not found.

Get Dump Entries
    [Documentation]  Return dump entries list.

    ${resp}=  OpenBMC Get Request  ${REST_DUMP_URI}
    Run Keyword If  '${resp.status_code}' == '${HTTP_NOT_FOUND}'
    ...  Set Global Variable  ${DUMP_ENTRY_URI}  /xyz/openbmc_project/dump/entry/

    ${dump_entries}=  Get URL List  ${DUMP_ENTRY_URI}
    [Return]  ${dump_entries}

Trigger Core Dump
    [Documentation]  Trigger core dump.

    # Find the pid of the active ipmid and kill it.
    ${cmd_buf}=  Catenate  kill -s SEGV $(ps | egrep ' ipmid$' |
    ...  egrep -v grep | \ cut -c1-6)

    ${cmd_output}  ${stderr}  ${rc}=  BMC Execute Command  ${cmd_buf}
    Should Be Empty  ${stderr}  msg=BMC execute command error.
    Should Be Equal As Integers  ${rc}  ${0}
    ...  msg=BMC execute command return code is not zero.

Initiate BMC Dump Using Redfish And Return Task Id
     [Documentation]  Initiate BMC dump via Redfish and return its task ID.

     ${payload}=  Create Dictionary  DiagnosticDataType=Manager
     ${resp}=  Redfish.Post
     ...  /redfish/v1/Managers/bmc/LogServices/Dump/Actions/LogService.CollectDiagnosticData
     ...  body=${payload}  valid_status_codes=[${HTTP_ACCEPTED}]

     # Example of response from above Redfish POST request.
     # "@odata.id": "/redfish/v1/TaskService/Tasks/0",
     # "@odata.type": "#Task.v1_4_3.Task",
     # "Id": "0",
     # "TaskState": "Running",
     # "TaskStatus": "OK"

     [Return]  ${resp.dict['Id']}

Create User Initiated BMC Dump Via Redfish
    [Documentation]  Generate user initiated BMC dump via Redfish and return the dump id number (e.g., "5").

    ${payload}=  Create Dictionary  DiagnosticDataType=Manager
    ${resp}=  Redfish.Post  /redfish/v1/Managers/bmc/LogServices/Dump/Actions/LogService.CollectDiagnosticData
    ...  body=${payload}  valid_status_codes=[${HTTP_ACCEPTED}]

    # Example of response from above Redfish POST request.
    # "@odata.id": "/redfish/v1/TaskService/Tasks/0",
    # "@odata.type": "#Task.v1_4_3.Task",
    # "Id": "0",
    # "TaskState": "Running",
    # "TaskStatus": "OK"

    Wait Until Keyword Succeeds  5 min  15 sec  Check Task Completion  ${resp.dict['Id']}
    ${task_id}=  Set Variable  ${resp.dict['Id']}

    ${task_dict}=  Redfish.Get Properties  /redfish/v1/TaskService/Tasks/${task_id}

    # Example of HttpHeaders field of task details.
    # "Payload": {
    #   "HttpHeaders": [
    #     "Host: <BMC_IP>",
    #      "Accept-Encoding: identity",
    #      "Connection: Keep-Alive",
    #      "Accept: */*",
    #      "Content-Length: 33",
    #      "Location: /redfish/v1/Managers/bmc/LogServices/Dump/Entries/2"]
    #    ],
    #    "HttpOperation": "POST",
    #    "JsonBody": "{\"DiagnosticDataType\":\"Manager\"}",
    #     "TargetUri": "/redfish/v1/Managers/bmc/LogServices/Dump/Actions/LogService.CollectDiagnosticData"
    # }

    [Return]  ${task_dict["Payload"]["HttpHeaders"][-1].split("/")[-1]}

Auto Generate BMC Dump
    [Documentation]  Auto generate BMC dump.

    ${cmd}=  Catenate  busctl --verbose call xyz.openbmc_project.Dump.Manager
    ...  /xyz/openbmc_project/dump/bmc xyz.openbmc_project.Dump.Create CreateDump a{sv} 0
    ${stdout}  ${stderr}  ${rc}=
    ...  BMC Execute Command  ${cmd}
    [Return]  ${stdout}  ${stderr}  ${rc}

Get Dump Size
    [Documentation]  Get dump size.
    [Arguments]  ${dump_uri}

    # Description of argument(s):
    # dump_uri        Dump URI
    #                 (Eg. 	/xyz/openbmc_project/dump/bmc/entry/1).

    # Example of Dump entry.
    # "data": {
    #   "CompletedTime": 1616760931,
    #   "Elapsed": 1616760931,
    #   "OffloadUri": "",
    #   "Offloaded": false,
    #   "Password": "",
    #   "Size": 3056,
    #   "SourceDumpId": 117440513,
    #   "StartTime": 1616760931,
    #   "Status": "xyz.openbmc_project.Common.Progress.OperationStatus.Completed",
    #   "VSPString": ""
    #  },

    Log  ${dump_uri}
    ${dump_data}=  Redfish.Get Properties  ${dump_uri}
    [Return]  ${dump_data["data"]["Size"]}

Get Dump ID
    [Documentation]  Return dump ID.
    [Arguments]   ${task_id}

    # Description of argument(s):
    # task_id        Task ID.

    # Example of HttpHeaders field of task details.
    # "Payload": {
    #   "HttpHeaders": [
    #     "Host: <BMC_IP>",
    #      "Accept-Encoding: identity",
    #      "Connection: Keep-Alive",
    #      "Accept: */*",
    #      "Content-Length: 33",
    #      "Location: /redfish/v1/Managers/bmc/LogServices/Dump/Entries/2"]
    #    ],
    #    "HttpOperation": "POST",
    #    "JsonBody": "{\"DiagnosticDataType\":\"Manager\"}",
    #     "TargetUri":
    # "/redfish/v1/Managers/bmc/LogServices/Dump/Actions/LogService.CollectDiagnosticData"
    # }

    ${task_dict}=  Redfish.Get Properties  /redfish/v1/TaskService/Tasks/${task_id}
    ${key}  ${value}=  Set Variable  ${task_dict["Payload"]["HttpHeaders"][-1].split(":")}
    Run Keyword If  '${key}' != 'Location'  Fail
    [Return]  ${value.strip('/').split('/')[-1]}

Get Task Status
    [Documentation]  Return task status.
    [Arguments]   ${task_id}

    # Description of argument(s):
    # task_id        Task ID.

    ${resp}=  Redfish.Get Properties  /redfish/v1/TaskService/Tasks/${task_id}
    [Return]  ${resp['TaskState']}

Check Task Completion
    [Documentation]  Check if the task is complete.
    [Arguments]   ${task_id}

    # Description of argument(s):
    # task_id        Task ID.

    ${task_dict}=  Redfish.Get Properties  /redfish/v1/TaskService/Tasks/${task_id}
    Should Be Equal As Strings  ${task_dict['TaskState']}  Completed

Get Dump ID And Status
    [Documentation]  Return dump ID and status.
    [Arguments]   ${task_id}

    # Description of argument(s):
    # task_id        Task ID.

    Wait Until Keyword Succeeds  10 min  15 sec  Check Task Completion  ${task_id}
    ${dump_id}=  Get Dump ID  ${task_id}
    [Return]  ${dump_id}  Completed


Create BMC User Dump
    [Documentation]  Generate user initiated BMC dump via Redfish and return
    ...  the task instance Id and response object (e.g., "5").

    ${payload}=  Create Dictionary  DiagnosticDataType=Manager
    ${resp}=  Redfish.Post  /redfish/v1/Managers/bmc/LogServices/Dump/Actions/LogService.CollectDiagnosticData
    ...  body=${payload}  valid_status_codes=[${HTTP_ACCEPTED}]

    ${ip_resp}=  Evaluate  json.loads(r'''${resp.text}''')  json

    Return From Keyword  ${ip_resp["Id"]}  ${resp}


Wait For Task Completion
    [Documentation]  Check whether the state of task instance matches any of the
    ...  expected completion states before maximum number of retries exceeds and
    ...  exit loop in case completion state is met.
    [Arguments]  ${task_id}  ${expected_status}  ${retry_max_count}=300
    ...  ${check_state}=${FALSE}

    # Description of argument(s):
    # task_id                     the task id for which completion is
    #                             to be monitored.
    # expected_status             the task state which is to be considered as the
    #                             end of task life cycle.
    # retry_max_count             the maximum number of retry count to wait for
    #                             task to reach its completion state. Default
    #                             value of retry_max_count is 300.
    # check_state                 if set as TRUE, the task state will be
    #                             monitored whether the task state value is
    #                             valid throughout task life cycle until
    #                             expected completion state is reached.
    #                             Default value of check_state is FALSE.

    FOR  ${retry}  IN RANGE  ${retry_max_count}
        ${resp}=  Redfish.Get Properties  /redfish/v1/TaskService/Tasks/${task_id}
        ${current_task_state}=  Set Variable  ${resp["TaskState"]}
        Rprint Vars  current_task_state

        Run Keyword If  ${check_state} == ${TRUE}  Should Be True
        ...  '${resp["TaskState"]}' in ${allowed_task_state}
        ...  msg=Verify task state is valid

        Exit For Loop If
        ...  '${resp["TaskState"]}' in ${expected_status}

        Sleep  5s
    END