[sheepdog-users] plan for Sheepdog 1.0 and new organization

Wenhao wenhao at zelin.io
Wed May 22 19:39:27 CEST 2013

Hi Kazutaka,

Here are some other improvements in my mind for the release. Put here
into wide discussions:

1.  QEMU sheepdog driver reconnect mechanism

Currently, if sheep exits, all the VMs need to be restarted. This
feature is necessary for online sheepdog upgrade and avoid restarting
VMs when sheep unexpected crashes.

2. Zookeeper Driver Panic.

If zk handle times out, the sheep will panic and exits now because it
doesn't know how to handle this time-out event. This needs a rework of
zookeeper driver to resend the event when timeout and should every
node detect duplicate events. @Kai is working on this.

3. Recovery I/O control.

When a node is joined/left or a new md disk is pluged/unpluged, this
will trigger the recovery event which will takes lots of bandwidth in
our test. This makes all the VM IO very slow. As Hongyi suggested, we
could either allocate a proportional bandwidth to the recovery I/O or
put them in a low priority.

The same could apply to other I/O channel, like vdi check.

4. A parameter to adjust recovery trigger interval.

This one is related to 3.

Now sheepdog always do eager recovery when a node is found lost. In
many cases, the node could join very soon. It is useful if there is a
parameter that the Admin could specify how long the recovery should


On Wed, May 22, 2013 at 11:24 AM, MORITA Kazutaka
<morita.kazutaka at gmail.com> wrote:
> Hi all,
> First of all, sorry for not having managed to release Sheepdog as I
> announced before.  I thought that I had to release Sheepdog 0.6 as
> soon as possible, but currently, Sheepdog becomes relatively stable,
> and there are some actual users who use Sheepdog for production.  I
> think it is a good time to release Sheepdog 1.0.
> There are still something to do for the release.  It would be nice if
> someone would help us the following issues especially from a user
> perspective:
>  - update wiki
>  - clean up sheep log outputs (check log level and contents, do spell
>    check, and so on)
>  - update man (collie.8 and sheep.8)
>  - modify /etc/init.d/sheepdog so that we can shutdown cluster cleanly
>    (adding '/etc/init.d/sheepdog stop-all'?)
>  - clean up collie command outputs
> I'm planning to release it on May 31, but let me know if someone has
> any opinions.
> We also plan to announce a new sheepdog organization in near future.
> The organization would be more formal and business friendly community,
> and I think of moving the current sheepdog homepage
> (http://www.osrg.net/sheepdog) to the organization.  We are still in
> the middle of discussing it.  If you are interested, please feel free
> to contact me or Yuan.
> Thanks,
> Kazutaka
> --
> sheepdog-users mailing lists
> sheepdog-users at lists.wpkg.org
> http://lists.wpkg.org/mailman/listinfo/sheepdog-users

More information about the sheepdog-users mailing list