Redfish Dump Service Proposal
Ratan Gupta
ratagupt at linux.vnet.ibm.com
Thu Jan 16 22:01:31 AEDT 2020
Hi All,
Further to the previous dump proposal, I have incorporated the following
changes.
* Having seprate LogService redfish resource for each type of dump.
* Enhance the
LogService(_https://redfish.dmtf.org/schemas/LogService.v1_1_3.json_)
property
"LogEntryType": {
"enum": [
"Event",
"SEL",
"Oem"
“*Dump*”
] }
* Enhance the LogService under OEM for further subsystem type
eg: hostboot dump, hypervisor dump
“OEM”: {
“*SystemType*”: { “enum” : [“HostBoot, Hypervisor, etc”] }
}
* Enhance the Log Service for the following
*Properties*:
1. *DumpOverridePolicy : **This can be different for all different
type of dumps.*
2. **MaxDumps: This can be different for all different type of dumps.**
*Actions:*
1. *CreateLog:**If logservice (**LogEntryType **is Dump) and the
subsystem type is “Hypervisor” then create log will create a
Host hypervisor dump.*
2. ***DeleteAll: Delete all the logs from this log service, This is
a addition on the deletion of the single resource(LogEntry).*
* Enhance the logentry*(*_*https://redfi*
<https://redfish.dmtf.org/schemas/LogEntry.v1_5_0.json>__*sh.dmtf.org/schemas/LogEntry.v1_5_0.json*
<https://redfish.dmtf.org/schemas/LogEntry.v1_5_0.json>_*)*
"LogEntryType": {
"enum": [
"Event",
"SEL",
"Oem"
“*Dump*”
] }
* Map the proposed dump properties with existing log entry property
o Proposed Property Existing LogEntry Property
+ ID ID
+ Timestamp Created
+ Reason LogEntryCode(Introduce more
enums in the LogEntry Code for the
dump reason)
* New Properties to be introduced in the logEntry
o *Size*
o *NOTE: *Dump type is not needed as the logservice logentry type
will tell that this service is for dump and the system type will
tell that this service is for which subsytem.
*NOTE: *
*1/ OdataID of log entry redfish resourc**e**will point to the raw dump
file which can be used to offload the dump.*
*2/ CreateLog: **spawns a task and returns the taskID.Client can query
the status for the task.*
Please let me know if I have missed something else.I would be making the
change in the PPT also.
Regards
Ratan Gupta
On 10/01/20 1:37 AM, Gunnar Mills wrote:
>
> A couple comments.
>
> On 1/7/2020 2:08 PM, Bills, Jason M wrote:
>>
>> On 1/7/2020 2:11 AM, Ratan Gupta wrote:
>>
>>>>
>>>>>
>>>>> On Sat, Dec 14, 2019 at 10:57 AM dhruvaraj S <dhruvaraj at gmail.com>
>>>>> wrote:
>>>>>>
>>>>>> On Sat, Dec 14, 2019 at 1:32 AM Bills, Jason M
>>>>>> <jason.m.bills at linux.intel.com> wrote:
>>>>>>>
>>>>>>> I like this as well. I'm trying to support a CPU crashdump that
>>>>>>> would
>>>>>>> fit perfectly with this proposal.
>>>>>>>
>>>>>>> A question and some comments below:
>>>>>>>
>>>>>>> Will Dump only have the two types: BMC and Host? Could this be more
>>>>>>> flexible to allow for multiple different types of dumps from
>>>>>>> various
>>>>>>> components?
>>>>>> + I think dump types should be flexible to cover different types of
>>>>>> host or bmc dumps from different components with varying formats.
>>> Sure we can enhance the type of dump, it is enum in the proposal
>>> which can be enhanced.
>>> What could be other dump type which I can add in the types?
>
> Slide 15: Since DumpType is an enum, should reason be as well? "Type"
> is a pretty typical enum in Redfish. E.g. BaseModuleType from
> https://redfish.dmtf.org/schemas/Memory.v1_8_0.json
>
> Reason seems similar to the LogEntryCode from
> https://redfish.dmtf.org/schemas/LogEntry.v1_5_0.json
>
> Slide 15:
> "Size": 108944B
> Redfish size properties typically have the unit in the name. E.g. From
> https://redfish.dmtf.org/schemas/Memory.v1_8_0.json CacheSizeMiB or
> CapacityMiB.
>
> odata.context is getting dropped. See
> https://github.com/DMTF/Redfish/issues/2722 or
> https://github.com/DMTF/Redfish/commit/ae49f4fb1278fd435f89317c3fa53cac597e3893#diff-e82b4876efbeaa600d3b104a426f7ac5
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ozlabs.org/pipermail/openbmc/attachments/20200116/643b99fe/attachment.htm>
More information about the openbmc
mailing list