Instead of `lbtargetgroups`, try `lb_target_groups` - the markup
formatting on the bosh.io docs page is rendering poorly.

Danny

On Wed, Mar 29, 2017 at 9:10 AM, Tsvetan Valchev <bgtsetso(a)gmail.com>
wrote:

Hi,

In my cloud config descriptor in the vm_types section I have specified
that I want the created via bosh deployment VM instances to be added to a
specific load balancer target group using the lbtargetgroups attribute,
e.g.:
vm_types:
- name: vm_myserver
cloud_properties: { instance_type: m4.xlarge, lbtargetgroups:
[my_target_group] }

Although the balancer and target groups are already created, the newly
created VMs are not added to the target group and I have to do this
manually via the aws console.
Do you have any ideas?

Thanks,
Tsvetan

--
Danny Berger

Hi,

In my cloud config descriptor in the vm_types section I have specified
that I want the created via bosh deployment VM instances to be added to a
specific load balancer target group using the lbtargetgroups attribute,
e.g.:
vm_types:
- name: vm_myserver
cloud_properties: { instance_type: m4.xlarge, lbtargetgroups:
[my_target_group] }

Although the balancer and target groups are already created, the newly
created VMs are not added to the target group and I have to do this
manually via the aws console.
Do you have any ideas?

Thanks,
Tsvetan

pivotal deploys updated stemcells regularly to PWS. high and critical CVEs
have a 48hr goal. we catch up on lows and mediums generally approximately
once per month.

On Wed, Mar 1, 2017 at 5:48 PM, Jonathan Stockley <jstockle(a)opentext.com>
wrote:

Hi, before deploying/upgrading a stemcell in production our security
group runs vulnerability scans on our staging deployments.
The problem is that by the time we get the stemcell into staging (about a
4-6 weeks), they have updated the vulnerability database and then there
scan find new issues.

How often are people upgrading stemcells in production?
How do you handle vulnerability scanning of BOSH deployed apps?
How about run.pivotal.io? How do they address this?

Thanks,
Jo

--
Thank you,

James Bayer

Hey Leandro,

Thanks for this feedback. One of the things the CF team has been working
on is better tooling around deploying Cloud Foundry, including IaaS setup.
A lot of this work is still in flight, which is why the CF docs don't have
any information about it yet; when the work is complete, there will
definitely be an overhaul of the docs to explain how to use the new tools.

In your outline, I think I'm noticing the following concerns:
1. How do I set up an AWS account so that I can deploy a CF? What are the
requirements (for example, quotas) for my account?
2. Can we default to using the most up-to-date instances?
3. Once my account is set up, how do I deploy CF (with Diego)?
4. How do I understand what I'm deploying and how I can modify it on my
own?

Let me know if I've understood your questions correctly: I'll make sure
that we take this feedback into account when we build the docs for the new
tools.

In the meantime, I'm happy to point people to the new tools, provided that
I emphasize that* these tools are not ready for production and still
undergoing active development. You should use these at your own risk.*
- To set up IaaS for CF: Take a look a "bbl" (short for bosh-bootloader
<https://github.com/cloudfoundry/bosh-bootloader>). This is tool that
does all the work to get you to a bosh director -- it create IaaS resources
like VPCs, NAT boxes, etc., and the creates a BOSH director that you can
target.
- To deploy CF: Take a look at cf-deployment
<https://github.com/cloudfoundry/cf-deployment>. It contains a BOSH
manifest for deploying Cloud Foundry (with Diego!). It also uses new BOSH
features to make manifest generation a good deal simpler. This will be the
future for how to deploy CF when we deprecated cf-release.

Again, these aren't ready for prime time yet, but it should be good for
development, testing, and experimentation -- not to mention that we'd love
feedback on this from the community. Feel free to follow up if you have any
questions.

David Sabeti
Product Manager, CF Release Integration Team


On Thu, Feb 23, 2017 at 12:04 AM Leandro David Cacciagioni <
leandro.21.2008(a)gmail.com> wrote:

Guys,

I have had a few issues over the last weeks deploying Cloud Foundry to
AWS and I think that maybe it will be nice to:

- Update the docs to be able to deploy without using the bosh aws gem
(It's broken at least with ruby 2.3.3/rbenv & Fedora Linux 25).
- Also is not nice that this GEM deletes everything when you try
to tear down all that was created, it not only delete what was create by
it, but everything else in the account if possible.
- Update the docs specifying the minimum AWS quotas required for
minimal HA deployment.
- Update the stub sample to use m4 instances, since this has been in
AWS for quite some time and they are already in most of the AWS regions.
- Beyond AWS it would be nice a detailed step by step guide to deploy
CF with Diego enabled, since for those who have never touched CF is a
little difficult to understand all the moving parts of CF.
- In my case I was giving more than a week of training, at least
one hour per day until they catch up the concepts and understand the basics
of how CF works (Forget about make them modify a bosh deployment to match a
Diego deployment if that's not in the docs)

Hope this help us all.

Cheers, Leandro.-