Re: core standup

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Thu, 5 Dec 2013, Gregory Farnum wrote:
> On Thursday, December 5, 2013, Sage Weil wrote:
>       Missing core standup today (all day meeting).
> 
>       Y:
>       - more wip-tier work
>       - almost done with memstore (a simple in-memory ObjectStore
>       backend to
>         serve as a reference implementation and possibly as a
>       benchmarking tool)
>       - more review of Zheng's mds branch
> 
>       T:
>       - meeting
>       - finish memstore
>       - more wip-tier wrangling (verify the trimming is working
>       properly; add a
>       test for it)
> 
>       B:
>       - wip-crush-tunables needs review
>       - wip-crush needs review
>       - wip-pool-set-* needs some consensus on approach.  the kludgey
>       fix is in
>       the pull requests; an alternative general approach is described
>       by
>              
>       https://github.com/ceph/ceph/pull/895#issuecomment-29846799
> 
> 
> Since we need this for the tiering, I'm going to tackle an alternative
> today that Sam and Joao and I have discussed previously, where the monitors
> in a quorum share their accepted APIs with each other and advertise the
> maximum subset to clients. I'll need to become more familiar with the
> API internals but I don't think it should be too difficult.

Cool.  Is this just a current_version and compat_version defined in 
MonCommands?  And if incompat with the leader, boot the client?

s


[Index of Archives]     [CEPH Users]     [Ceph Large]     [Information on CEPH]     [Linux BTRFS]     [Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]
  Powered by Linux