[fdio] Bump compute RAM defaults for virtual PODs 04/67804/4
authorAlexandru Avadanii <Alexandru.Avadanii@enea.com>
Wed, 8 May 2019 20:17:23 +0000 (22:17 +0200)
committerAlexandru Avadanii <Alexandru.Avadanii@enea.com>
Wed, 8 May 2019 22:40:57 +0000 (00:40 +0200)
commitb381277ae274473ae4e05a1aa9dd171dbab461d6
treeb9f9006603f3d72cfabcc1d509b3a82a2ea2da22
parente2bbc14187ddc65b1b54762a4cebcb48cba38bf0
[fdio] Bump compute RAM defaults for virtual PODs

Hugepage count has been recently bumped for virtual PODs via IDF
changes in Pharos, so align our FDio scenarios with the new RAM
requirements.

While at it, fix wrong pod_config template evaluation by moving it
after the templated scenario files are expanded, since pod_config
relies on scenario node definition.

Also, configure VPP to use decimal interface names by default to
align with Pharos macro for the VPP interface name string.

Change-Id: Ib3a89c294a3a2755567fdbe07e3be2b8ca1a5714
Signed-off-by: Alexandru Avadanii <Alexandru.Avadanii@enea.com>
mcp/config/scenario/os-nosdn-fdio-ha.yaml
mcp/config/scenario/os-nosdn-fdio-noha.yaml
mcp/patches/salt-formula-linux/0002-network-Bring-in-basic-VPP-support.patch
mcp/reclass/classes/cluster/mcp-common-noha/openstack_compute_pdf.yml.j2
mcp/reclass/classes/cluster/mcp-common-noha/openstack_gateway_pdf.yml.j2
mcp/scripts/lib_template.sh