There had been discussion to support killing tasks from the operator API,
but it's not in place yet.

In the interim, you can manually kill them from the host.

On Thu, Oct 26, 2017 at 11:24 AM, Zhitao Li <[email protected]> wrote:

> Each active framework on HTTP scheduler API is allocated with a stream id.
> This is included in the header "Mesos-Stream-Id" in initial subscribed
> response.
>
> If you obtain this stream id, another process can "impersonate" this
> framework to submit kill requests (using framework id, task id and the same
> credential if auth is enabled).
>
> I cannot find a good place in Mesos master which logs this stream id, so
> probably report it in your framework side?
>
> Hope this is helpful as we have done similar things recently.
>
>
>
> On Thu, Oct 26, 2017 at 2:10 AM, Manuel Montesino <
> [email protected]> wrote:
>
>> Hi,
>>
>>
>> We have a framework with some tasks that we would like to kill but not
>> all framework (teardown), so we would like to use the kill method of the
>> http api scheduler, the problem is that is needed to be suscribed. Creating
>> a new framework in stream mode and executing the kill method it's not
>> allowed to kill task for another framework (403 Forbidden).
>>
>>
>> So, is possible to subscribe/connect to an existing framework, or other
>> mode to operate into it from the http api?.
>>
>>
>> Thanks in advance.
>>
>>
>> *Manuel Montesino*
>> Devops Engineer
>>
>> *E* *manuel.montesino@piksel(dot)com*
>>
>> Marie Curie,1. Ground Floor. Campanillas, Malaga 29590
>> *liberating viewing* | *piksel.com <http://piksel.com>*
>>
>> [image: Piksel_Email.png]
>>
>> This message is private and confidential. If you have received this
>> message in error, please notify the sender or [email protected] and
>> remove it from your system.
>>
>> Piksel Inc is a company registered in the United States, 2100 Powers
>> Ferry Road SE, Suite 400, Atlanta, GA 30339
>> <https://maps.google.com/?q=2100+Powers+Ferry+Road+SE,+Suite+400,+Atlanta,+GA+30339&entry=gmail&source=g>
>>
>
>
>
> --
> Cheers,
>
> Zhitao Li
>

Reply via email to