mesos-reviews mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Jiang Yan Xu <...@jxu.me>
Subject Re: Review Request 47082: LostSlaveMessage should be sent to affected frameworks only.
Date Fri, 10 Jun 2016 16:08:36 GMT


> On June 10, 2016, 5:47 a.m., Neil Conway wrote:
> > Something else to consider here: this change means that frameworks will miss a `slaveLost()`
signal that they care about in some circumstances. For example, suppose an agent has a persistent
volume on an agent; the master fails over, and the agent fails to reregister with the master.
We'll remove the agent, but we _won't_ send `SlaveLostMessage` to the framework in this case,
because the master doesn't know the framework has a persistent volume on the agent.
> > 
> > Since `slaveLost()` is unreliable to begin with, I don't think this is a show-stopper,
but it's a bit unfortunate...

Yeah in this case we'll have to fall back to broadcasting, also a bit unfortunate but there's
not a better way IMO.


- Jiang Yan


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/47082/#review136992
-----------------------------------------------------------


On June 8, 2016, 6:08 p.m., Anindya Sinha wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/47082/
> -----------------------------------------------------------
> 
> (Updated June 8, 2016, 6:08 p.m.)
> 
> 
> Review request for mesos and Jiang Yan Xu.
> 
> 
> Bugs: MESOS-5143
>     https://issues.apache.org/jira/browse/MESOS-5143
> 
> 
> Repository: mesos
> 
> 
> Description
> -------
> 
> When a slave is removed, master sends a LostSlaveMessage to affected
> frameworks only (instead of all registered frameworks). An affected
> framework is a framework which satisfied one or more conditions of
> the following:
> 
> 1. There are tasks on this slave belonging to the framework.
> 2. There are pending tasks on this slave belonging to the framework.
> 3. Reserved resources on the slave have a matching role with the
>    role of the framework.
> 4. There are pending offers or pending inverse offers from this slave
>    for the framework.
> 
> 
> Diffs
> -----
> 
>   src/master/master.hpp 790da3ce686401c378ad9c62d497d60893c4ce41 
>   src/master/master.cpp 21ec70491ce6b79be57ff8db51d4b2fa682b32ce 
>   src/tests/master_tests.cpp 34be015aa314a7574e9065efb7b1bb8e1570c5b7 
> 
> Diff: https://reviews.apache.org/r/47082/diff/
> 
> 
> Testing
> -------
> 
> All existing and modified tests passed.
> 
> 
> Thanks,
> 
> Anindya Sinha
> 
>


Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message