Re: BOSH / cloudfoundry - stop / start during maintenance

Senthil Velusamy <vrss76@...>

Thanks Stefan!

On Wed, Aug 17, 2016 at 2:28 PM, Stefan Mayr <stefan(a)> wrote:


Am 17.08.2016 um 18:49 schrieb Senthil Velusamy:


We have deployed BOSH / cloudfoundry on vSphere. There are two hosts
(each host is an availability zone) and vm's are distributed between these
two hosts. microBOSH, BOSH and cloudfoundry all three are co-located in
these two hosts.

Currently we stop / start all the vm's running on those two hosts during
maintenance activities. What we are finding is that when the vm's come up
apps come in started state and after that apps getting re-started for a

Are there any recommended practice to cleanly shutdown microBOSH, BOSH,
cloudfoundry components and custom deployed apps during maintenace? Can
someone point me to documentation or best practices to stop / start?


For Cloud Foundry you can take a look at the Pivotal documentation:
It has tables for the correct ordering when stopping or starting the VMs
via bosh. For automation we use a fork of
71faef39e5005694b11564ab00c38458 (Gemfile path has changed) but I just
found some even more generic script in



Join { to automatically receive all group messages.