*Request For Feedback*: TimeOwner in phosphor-timemanager

*Request For Feedback*: TimeOwner in phosphor-timemanager

[vishwa]

[-- Attachment #1: Type: text/plain, Size: 3071 bytes --]

Hi All,

Sending this email requesting the feedback on the usage/need of 
"TimeOwner" that is part of phosphor-timemanager.

phosphor-timemanager is a time keeping daemon on the BMC that uses BMC's 
RTC to hold the time. It uses systemd-timesyncd and systemd-timedated.

There are 2 D-Bus objects that implement 
"phosphor-dbus-interfaces/xyz/openbmc_project/Time/EpochTime.interface.yaml" 
and is managed by a common manager.

/xyz/openbmc_projects/time/bmc  : Used when the request is coming in via 
Redfish
/xyz/openbmc_projects/time/host : Used when the request is inband with 
IPMI_SET_SEL_TIME and IPMI_GET_SEL_TIME + PLDM ones.

timemanager consumes 2 settings, namely, TimeSyncMethod and TimeOwner 
defined in phosphor-dbus-interfaces/xyz/openbmc_project/Time/.

TimeSyncMethod: Indicates whether the NTP must be used or should it be 
settable by the user.
                 For example: If the time set request comes and the 
Method was set to NTP, then the write request would be ignored.

TimeOwner: This is where we have seen the community telling why do we 
even have such a thing. Talking to the UX team, looks like, the 
customers have had difficulty understanding this.

There are 4 values for TimeOwners: ( Works irrespective of acting on 
time/bmc or time/host D-Bus objects )

/BMC : BMC is the owner of the time. So, Host *can not *set the time on 
BMC. Inband SET_SEL_TIME would result in error.//
//
//HOST: Host is the owner of the time. So, only the host *can set* the 
time on BMC. Any external Redfish SET requests would result in error/

/SPLIT: This is where we could achieve independent times for BMC and 
Host. This is a combination of BMC + HOST. Both BMC and HOST can set
        the time. Host time is maintained as an offset into BMC's time. 
So, whenever the Host sends the time request, the offset is
        factored into//BMC's time and then the response would be returned.//
//
//BOTH : BMC and/or HOST can SET the time. Whoever did SET it last, will 
be maintained.
/

Personally, I have worked on github issue from Redhat when there were 
failures setting the time from OS using "hwclock" since the Owner was 
set to [BMC]. The solution was to [poweroff] the system, change the 
Owner to [HOST/SPLIT] and boot back.

Note: Since these policies are user settable, they can be changed at 
anytime. However, they will not be taken into account until the system 
is booted again.

*Why did we even have this ?*: During the initial development of 
timemanager, the thought was that, Host would not trust the BMC and 
hence we want a way where BMC and Host can maintain their own time

*Do we need this going forward ?*: I am being asked by UX team about the 
need of this and I mentioned I would get the community feedback on this. 
Although, I feel this level of granularity gives control over how we can 
manage time, it would not justify the complexity if the customers don't 
appreciate it. Removing the TimeOwner would make the code a lot simpler.

Please help with your votes.

Thank you,

!! Vishwa !!


[-- Attachment #2: Type: text/html, Size: 3987 bytes --]

RE: *Request For Feedback*: TimeOwner in phosphor-timemanager

[Pine, Kathryn ElaineX]

[-- Attachment #1: Type: text/plain, Size: 924 bytes --]

> Do we need this going forward ?: I am being asked by UX team about the need of this and I mentioned I would get the community feedback on this. Although, I feel this level of granularity gives control over how we can manage time, it would not justify the complexity if the customers don't appreciate it. Removing the TimeOwner would make the code a lot simpler.
Hi,

I am coming from the UX side of how we set the date time settings page up for phosphor-webui downstream recently. We switched our page to use Redfish, here’s how ours is set up now:

There is no longer a “time owner” and the setting is either:
NTPEnabled: true or false
If false, we are not allowing the user to set the time, because the BMC is synching from the host time and therefore any settings we made to the time on the BMC would be overwritten.
If true, we use the NTP server(s) the user provides.

We are testing this currently.

[-- Attachment #2: Type: text/html, Size: 3652 bytes --]

Re: *Request For Feedback*: TimeOwner in phosphor-timemanager

[vishwa]

[-- Attachment #1: Type: text/plain, Size: 1480 bytes --]

On 12/11/19 12:17 AM, Pine, Kathryn ElaineX wrote:
>
> >*Do we need this going forward ?*: I am being asked by UX team about 
> the need of this and I mentioned I would get the community feedback on 
> this. Although, I feel this level of granularity gives control over 
> how we can manage time, it would not justify the complexity if the 
> customers don't appreciate it. Removing the TimeOwner would make the 
> code a lot simpler.
>
> Hi,
>
> I am coming from the UX side of how we set the date time settings page 
> up for phosphor-webui downstream recently. We switched our page to use 
> Redfish, here’s how ours is set up now:
>
> There is no longer a “time owner” and the setting is either:
>
> NTPEnabled: true or false
>
> If false, we are not allowing the user to set the time, because the 
> BMC is synching from the host time and therefore any settings we made 
> to the time on the BMC would be overwritten.
>
> If true, we use the NTP server(s) the user provides.
>
> We are testing this currently.
>

Hi, Thank you for the response. From what I interpreted:

- TimeOwner is not settable via GUI
- Irrespective of NTP setting, the user is not allowed to set the time 
on BMC
- BMC is syncing the time from Host.
    - This means, the current TimeOwner is "Host" and NTP is off. Is 
that being set as default ?

I am requesting for feedback from the community on the need of TimeOwner 
feature that we have in Settings and timemanager code.

!! Vishwa !!
>

[-- Attachment #2: Type: text/html, Size: 4816 bytes --]

Re: *Request For Feedback*: TimeOwner in phosphor-timemanager

[vishwa]

[-- Attachment #1: Type: text/plain, Size: 1682 bytes --]

Hello community,

Would be really great if there is any feedback on this.

Have a happy holidays !!

!! Vishwa !!

On 12/11/19 3:24 PM, vishwa wrote:
> On 12/11/19 12:17 AM, Pine, Kathryn ElaineX wrote:
>>
>> >*Do we need this going forward ?*: I am being asked by UX team about 
>> the need of this and I mentioned I would get the community feedback 
>> on this. Although, I feel this level of granularity gives control 
>> over how we can manage time, it would not justify the complexity if 
>> the customers don't appreciate it. Removing the TimeOwner would make 
>> the code a lot simpler.
>>
>> Hi,
>>
>> I am coming from the UX side of how we set the date time settings 
>> page up for phosphor-webui downstream recently. We switched our page 
>> to use Redfish, here’s how ours is set up now:
>>
>> There is no longer a “time owner” and the setting is either:
>>
>> NTPEnabled: true or false
>>
>> If false, we are not allowing the user to set the time, because the 
>> BMC is synching from the host time and therefore any settings we made 
>> to the time on the BMC would be overwritten.
>>
>> If true, we use the NTP server(s) the user provides.
>>
>> We are testing this currently.
>>
>
> Hi, Thank you for the response. From what I interpreted:
>
> - TimeOwner is not settable via GUI
> - Irrespective of NTP setting, the user is not allowed to set the time 
> on BMC
> - BMC is syncing the time from Host.
>    - This means, the current TimeOwner is "Host" and NTP is off. Is 
> that being set as default ?
>
> I am requesting for feedback from the community on the need of 
> TimeOwner feature that we have in Settings and timemanager code.
>
> !! Vishwa !!
>>

[-- Attachment #2: Type: text/html, Size: 5461 bytes --]

Re: *Request For Feedback*: TimeOwner in phosphor-timemanager

[Kun Yi]

[-- Attachment #1: Type: text/plain, Size: 2257 bytes --]

On Wed, Dec 18, 2019 at 7:02 AM vishwa <vishwa@linux.vnet.ibm.com> wrote:

> Hello community,
>
> Would be really great if there is any feedback on this.
>
> Have a happy holidays !!
>
> !! Vishwa !!
> On 12/11/19 3:24 PM, vishwa wrote:
>
> On 12/11/19 12:17 AM, Pine, Kathryn ElaineX wrote:
>
> >* Do we need this going forward ?*: I am being asked by UX team about
> the need of this and I mentioned I would get the community feedback on
> this. Although, I feel this level of granularity gives control over how we
> can manage time, it would not justify the complexity if the customers don't
> appreciate it. Removing the TimeOwner would make the code a lot simpler.
>
> Hi,
>
>
>
> I am coming from the UX side of how we set the date time settings page up
> for phosphor-webui downstream recently. We switched our page to use
> Redfish, here’s how ours is set up now:
>
>
>
> There is no longer a “time owner” and the setting is either:
>
> NTPEnabled: true or false
>
> If false, we are not allowing the user to set the time, because the BMC is
> synching from the host time and therefore any settings we made to the time
> on the BMC would be overwritten.
>
> If true, we use the NTP server(s) the user provides.
>
>
>
> We are testing this currently.
>
>
> Hi, Thank you for the response. From what I interpreted:
>
> - TimeOwner is not settable via GUI
> - Irrespective of NTP setting, the user is not allowed to set the time on
> BMC
> - BMC is syncing the time from Host.
>    - This means, the current TimeOwner is "Host" and NTP is off. Is that
> being set as default ?
>
> I am requesting for feedback from the community on the need of TimeOwner
> feature that we have in Settings and timemanager code.
>
> !! Vishwa !!
>
> My 2cents here along with a few questions:
- how does the current time manager interact with systemd-timesyncd?
- From my experience, BMC time can drift a lot easily and it is more useful
to have the BMC time synced externally
- Whether NTP or host time is preferred should be left to the platform
designer to decide
- sounds like it would be more straightforward to have the 'Owner' reflect
one time source: NTP/IPMI/Redfish

-- 
Regards,
Kun

[-- Attachment #2: Type: text/html, Size: 4633 bytes --]

Re: *Request For Feedback*: TimeOwner in phosphor-timemanager

[vishwa]

[-- Attachment #1: Type: text/plain, Size: 3209 bytes --]

On 12/19/19 1:29 AM, Kun Yi wrote:
>

>
> On Wed, Dec 18, 2019 at 7:02 AM vishwa <vishwa@linux.vnet.ibm.com 
> <mailto:vishwa@linux.vnet.ibm.com>> wrote:
>
>     Hello community,
>
>     Would be really great if there is any feedback on this.
>
>     Have a happy holidays !!
>
>     !! Vishwa !!
>
>     On 12/11/19 3:24 PM, vishwa wrote:
>>     On 12/11/19 12:17 AM, Pine, Kathryn ElaineX wrote:
>>>
>>>     >*Do we need this going forward ?*: I am being asked by UX team
>>>     about the need of this and I mentioned I would get the community
>>>     feedback on this. Although, I feel this level of granularity
>>>     gives control over how we can manage time, it would not justify
>>>     the complexity if the customers don't appreciate it. Removing
>>>     the TimeOwner would make the code a lot simpler.
>>>
>>>     Hi,
>>>
>>>     I am coming from the UX side of how we set the date time
>>>     settings page up for phosphor-webui downstream recently. We
>>>     switched our page to use Redfish, here’s how ours is set up now:
>>>
>>>     There is no longer a “time owner” and the setting is either:
>>>
>>>     NTPEnabled: true or false
>>>
>>>     If false, we are not allowing the user to set the time, because
>>>     the BMC is synching from the host time and therefore any
>>>     settings we made to the time on the BMC would be overwritten.
>>>
>>>     If true, we use the NTP server(s) the user provides.
>>>
>>>     We are testing this currently.
>>>
>>
>>     Hi, Thank you for the response. From what I interpreted:
>>
>>     - TimeOwner is not settable via GUI
>>     - Irrespective of NTP setting, the user is not allowed to set the
>>     time on BMC
>>     - BMC is syncing the time from Host.
>>        - This means, the current TimeOwner is "Host" and NTP is off.
>>     Is that being set as default ?
>>
>>     I am requesting for feedback from the community on the need of
>>     TimeOwner feature that we have in Settings and timemanager code.
>>
>>     !! Vishwa !!
>>>
> My 2cents here along with a few questions:
> - how does the current time manager interact with systemd-timesyncd?

TimeManager interacts with systemd-timedated, which in turn engages 
systemd-timesync on setting *NTP *policy.
https://www.freedesktop.org/software/systemd/man/systemd-timesyncd.service.html**

> - From my experience, BMC time can drift a lot easily and it is more 
> useful to have the BMC time synced externally

This is a discussion point for using / not using NTP. It would not 
impact TimeOwner
https://github.com/openbmc/phosphor-time-manager/blob/master/README.md

> - Whether NTP or host time is preferred should be left to the platform 
> designer to decide

If the setting is NTP, then Host can not set the time on BMC unless the 
Owner is SPLIT.
I believe, you are saying more from a NTP / Non NTP as opposed to 
TimeOwner ?

> - sounds like it would be more straightforward to have the 'Owner' 
> reflect one time source: NTP/IPMI/Redfish

This is part of TimeSynchronization. TimeOwner is another layer on who 
owns the time as opposed to how the time is set.
Hopefully the README I pointed to before would help.

Thank you,
!! Vishwa !!
>
> -- 
> Regards,
> Kun

[-- Attachment #2: Type: text/html, Size: 8223 bytes --]

Re: *Request For Feedback*: TimeOwner in phosphor-timemanager

[vishwa]

[-- Attachment #1: Type: text/plain, Size: 3504 bytes --]

Hi,

Happy New Year ...

btw, looks like TimeOwner may not be of interest for the community ?. 
Shall we go ahead remove that setting ?

!! Vishwa !!

On 12/19/19 1:30 PM, vishwa wrote:
> On 12/19/19 1:29 AM, Kun Yi wrote:
>>
>
>>
>> On Wed, Dec 18, 2019 at 7:02 AM vishwa <vishwa@linux.vnet.ibm.com 
>> <mailto:vishwa@linux.vnet.ibm.com>> wrote:
>>
>>     Hello community,
>>
>>     Would be really great if there is any feedback on this.
>>
>>     Have a happy holidays !!
>>
>>     !! Vishwa !!
>>
>>     On 12/11/19 3:24 PM, vishwa wrote:
>>>     On 12/11/19 12:17 AM, Pine, Kathryn ElaineX wrote:
>>>>
>>>>     >*Do we need this going forward ?*: I am being asked by UX team
>>>>     about the need of this and I mentioned I would get the
>>>>     community feedback on this. Although, I feel this level of
>>>>     granularity gives control over how we can manage time, it would
>>>>     not justify the complexity if the customers don't appreciate
>>>>     it. Removing the TimeOwner would make the code a lot simpler.
>>>>
>>>>     Hi,
>>>>
>>>>     I am coming from the UX side of how we set the date time
>>>>     settings page up for phosphor-webui downstream recently. We
>>>>     switched our page to use Redfish, here’s how ours is set up now:
>>>>
>>>>     There is no longer a “time owner” and the setting is either:
>>>>
>>>>     NTPEnabled: true or false
>>>>
>>>>     If false, we are not allowing the user to set the time, because
>>>>     the BMC is synching from the host time and therefore any
>>>>     settings we made to the time on the BMC would be overwritten.
>>>>
>>>>     If true, we use the NTP server(s) the user provides.
>>>>
>>>>     We are testing this currently.
>>>>
>>>
>>>     Hi, Thank you for the response. From what I interpreted:
>>>
>>>     - TimeOwner is not settable via GUI
>>>     - Irrespective of NTP setting, the user is not allowed to set
>>>     the time on BMC
>>>     - BMC is syncing the time from Host.
>>>        - This means, the current TimeOwner is "Host" and NTP is off.
>>>     Is that being set as default ?
>>>
>>>     I am requesting for feedback from the community on the need of
>>>     TimeOwner feature that we have in Settings and timemanager code.
>>>
>>>     !! Vishwa !!
>>>>
>> My 2cents here along with a few questions:
>> - how does the current time manager interact with systemd-timesyncd?
>
> TimeManager interacts with systemd-timedated, which in turn engages 
> systemd-timesync on setting *NTP *policy.
> https://www.freedesktop.org/software/systemd/man/systemd-timesyncd.service.html**
>
>> - From my experience, BMC time can drift a lot easily and it is more 
>> useful to have the BMC time synced externally
>
> This is a discussion point for using / not using NTP. It would not 
> impact TimeOwner
> https://github.com/openbmc/phosphor-time-manager/blob/master/README.md
>
>> - Whether NTP or host time is preferred should be left to the 
>> platform designer to decide
>
> If the setting is NTP, then Host can not set the time on BMC unless 
> the Owner is SPLIT.
> I believe, you are saying more from a NTP / Non NTP as opposed to 
> TimeOwner ?
>
>> - sounds like it would be more straightforward to have the 'Owner' 
>> reflect one time source: NTP/IPMI/Redfish
>
> This is part of TimeSynchronization. TimeOwner is another layer on who 
> owns the time as opposed to how the time is set.
> Hopefully the README I pointed to before would help.
>
> Thank you,
> !! Vishwa !!
>>
>> -- 
>> Regards,
>> Kun

[-- Attachment #2: Type: text/html, Size: 9688 bytes --]

Re: *Request For Feedback*: TimeOwner in phosphor-timemanager

[vishwa]

[-- Attachment #1: Type: text/plain, Size: 3986 bytes --]


Alright.. This is what I am going ahead with for TimeOwner.

Phase-1 and near term goal: Make "BOTH" as default than what we 
currently have ( "BMC" )
Phase-2: Modify the code to remove support for HOST / BMC / SPLIT. Code 
will behave as if it's BOTH

If this breaks any usecase, please let me know.

Thank you,

!! Vishwa !!


On 1/6/20 2:48 PM, vishwa wrote:
> Hi,
>
> Happy New Year ...
>
> btw, looks like TimeOwner may not be of interest for the community ?. 
> Shall we go ahead remove that setting ?
>
> !! Vishwa !!
>
> On 12/19/19 1:30 PM, vishwa wrote:
>> On 12/19/19 1:29 AM, Kun Yi wrote:
>>>
>>
>>>
>>> On Wed, Dec 18, 2019 at 7:02 AM vishwa <vishwa@linux.vnet.ibm.com 
>>> <mailto:vishwa@linux.vnet.ibm.com>> wrote:
>>>
>>>     Hello community,
>>>
>>>     Would be really great if there is any feedback on this.
>>>
>>>     Have a happy holidays !!
>>>
>>>     !! Vishwa !!
>>>
>>>     On 12/11/19 3:24 PM, vishwa wrote:
>>>>     On 12/11/19 12:17 AM, Pine, Kathryn ElaineX wrote:
>>>>>
>>>>>     >*Do we need this going forward ?*: I am being asked by UX team
>>>>>     about the need of this and I mentioned I would get the
>>>>>     community feedback on this. Although, I feel this level of
>>>>>     granularity gives control over how we can manage time, it
>>>>>     would not justify the complexity if the customers don't
>>>>>     appreciate it. Removing the TimeOwner would make the code a
>>>>>     lot simpler.
>>>>>
>>>>>     Hi,
>>>>>
>>>>>     I am coming from the UX side of how we set the date time
>>>>>     settings page up for phosphor-webui downstream recently. We
>>>>>     switched our page to use Redfish, here’s how ours is set up now:
>>>>>
>>>>>     There is no longer a “time owner” and the setting is either:
>>>>>
>>>>>     NTPEnabled: true or false
>>>>>
>>>>>     If false, we are not allowing the user to set the time,
>>>>>     because the BMC is synching from the host time and therefore
>>>>>     any settings we made to the time on the BMC would be overwritten.
>>>>>
>>>>>     If true, we use the NTP server(s) the user provides.
>>>>>
>>>>>     We are testing this currently.
>>>>>
>>>>
>>>>     Hi, Thank you for the response. From what I interpreted:
>>>>
>>>>     - TimeOwner is not settable via GUI
>>>>     - Irrespective of NTP setting, the user is not allowed to set
>>>>     the time on BMC
>>>>     - BMC is syncing the time from Host.
>>>>        - This means, the current TimeOwner is "Host" and NTP is
>>>>     off. Is that being set as default ?
>>>>
>>>>     I am requesting for feedback from the community on the need of
>>>>     TimeOwner feature that we have in Settings and timemanager code.
>>>>
>>>>     !! Vishwa !!
>>>>>
>>> My 2cents here along with a few questions:
>>> - how does the current time manager interact with systemd-timesyncd?
>>
>> TimeManager interacts with systemd-timedated, which in turn engages 
>> systemd-timesync on setting *NTP *policy.
>> https://www.freedesktop.org/software/systemd/man/systemd-timesyncd.service.html**
>>
>>> - From my experience, BMC time can drift a lot easily and it is more 
>>> useful to have the BMC time synced externally
>>
>> This is a discussion point for using / not using NTP. It would not 
>> impact TimeOwner
>> https://github.com/openbmc/phosphor-time-manager/blob/master/README.md
>>
>>> - Whether NTP or host time is preferred should be left to the 
>>> platform designer to decide
>>
>> If the setting is NTP, then Host can not set the time on BMC unless 
>> the Owner is SPLIT.
>> I believe, you are saying more from a NTP / Non NTP as opposed to 
>> TimeOwner ?
>>
>>> - sounds like it would be more straightforward to have the 'Owner' 
>>> reflect one time source: NTP/IPMI/Redfish
>>
>> This is part of TimeSynchronization. TimeOwner is another layer on 
>> who owns the time as opposed to how the time is set.
>> Hopefully the README I pointed to before would help.
>>
>> Thank you,
>> !! Vishwa !!
>>>
>>> -- 
>>> Regards,
>>> Kun

[-- Attachment #2: Type: text/html, Size: 11215 bytes --]