Thanks again!
We're not (yet) using docker (or any other form of containerization for
that matter).
I think in the meantime the approach of sub-advertising resources should
work for us.
As we're spinning clusters up and down on demand on GCE, I can probably do
this pretty easily by adding these
Hi Itamar,
You can pass the amount of CPU and memory that the slave is advertising to
the master for scheduling as part of the resources slave flag. So you can
only schedule 12 cpus and leave 4 for your services if you want.
That said, there are discussions about launching multiple tasks
That's a great point Itamar, and something we discussed quite some time ago
here but never implemented. These are the first two options that spring to mind
that I can remember...
- Are you using docker containers for your tasks? Why not use containers
pre-configured on the box for these
I completely agree with Charles, though I think I can appreciate what you're
trying to do here. Take the log aggregation service as an example, you want
that on every slave to aggregate logs, but want to avoid using yet another
layer of configuration management to deploy it.
I'm of the
We've been experimenting with using Marathon to do this. We've found a
couple issues in making in smooth, but those bugs are all in the process of
being fixed. Within the next few weeks, I think that Marathon will be a
solid choice for this use case.
On Tue, Jan 6, 2015 at 4:05 AM, Itamar
5 matches
Mail list logo
