Age | Commit message (Collapse) | Author | Files | Lines |
|
Change-Id: If646ce606f96d0bcd6681ada193d60e6b59f93ff
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
Issue where jinja 2 templates were missing from spec file. Also adds
error checking to make sure the nic templates were correctly generated.
There was an error with dependencies missing on the build server - this
patch includes a fix to make sure it is installed before trying to make.
Change-Id: I7989ccf297a4e0d81326cccc06b2132018e7d8ae
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
|
|
|
|
Add support for CPU pinning of libvirt guests and also
restricting host processes to selected CPUs via isolcpus.
Hugepage support is added using the same mechanism as
isolcpus, along with a perf scenario where all 3 performance
options are enabled.
Deploy options are now parsed in python
JIRA: APEX-127
JIRA: APEX-105
JIRA: APEX-106
Change-Id: I438e80fb88e596cc017595d43bc1efda1001325c
opnfv-tht-pr: 8
Signed-off-by: Michael Chapman <woppin@gmail.com>
|
|
brbm bridge is not created during deployment because of existing random
bridge containg text brbm in it.
JIRA: APEX-153
Change-Id: I48520c2865a4de0de849c5147337dafeeb5fb51f
Signed-off-by: Juha Haapavirta <juha.haapavirta@nokia.com>
|
|
Heat uses sub-process forking for increasing number of workers to handle
tasks. This patch limits number of workers in undercloud and overcloud
for virtual deployments to help battle resource contention.
Change-Id: I70edff708b7dcaf114b285046af3138a67167150
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
JIRA: APEX-114
opnfv-tht-pr: 14
Change-Id: I328058b9f4503ccf7ec0a8d587292504b93cd17f
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
Testing this to see if it improves stability of deploys
Change-Id: I1bd2539e87230c080e2e652a1d74681d95a293fe
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
|
|
Changes Include:
- All VMs (undercloud and overcloud) now default to 4 vCPUs
- Allow specifying number of vCPUs per overcloud VM with --virtual-cpus <number>
- Allow specifying amount of RAM per overcloud VM with --virtual-ram <size in GB>
- Undercloud now uses 10GiB of RAM to avoid OOM
JIRA: APEX-111
Change-Id: I541c669b869a90134bc8af25786fe3b1a557376b
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
Change-Id: I561c636ac8f0cde9a23a3a74906bd88662f6dd59
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
Convert compute and controller NIC templates to jinja2 format.
Added support for IPv6 for those NIC templates. Existing IPv4
template generation behavior does not change.
The .template files are kept in tree for now, they will be
removed after VLAN changes are made.
Change-Id: I2ffc7c403af3659db780ece2bd9195cd62748f0c
Signed-off-by: Feng Pan <fpan@redhat.com>
|
|
|
|
Changes:
- Implements network_settings.yaml file parsing in python.
- Adds support for both IPv4 and IPv6 in network_settings.yaml
- Adds support for api_network in network_settings.yaml
- Removes bash library functions for network related functions.
- Adds dependency to python34-yaml for apex-common package.
Note that support for ipv6 and api_network is not complete yet.
Proper configuriration of network environment and nic template
files will be added later.
Change-Id: I087f725dabedfef109c9de1f58ce2611da647e87
Signed-off-by: Feng Pan <fpan@redhat.com>
|
|
Purpose of this commit is to test opnfv-tht changes are not breaking
nosdn deployment.
opnfv-tht-pr: 10
Change-Id: I85da61a9591999d1249d55700241d71256019a81
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
- first stage in standarizing how custom config
values are passed in by moving
to using ENV vars to pass in custom params
- next stage is to remove the -r and -c
options in deploy.sh
Moving this direction will reduce managment of
custom parameters and rely on ENV vars to make
the customizations for development
Change-Id: Ibbc171d5e6801b27c1b62ca65c2a9f7830483b97
Signed-off-by: Dan Radez <dradez@redhat.com>
|
|
Using pool-refresh should be sufficient to clean up storage pool.
Change-Id: I10a58340c4c24b429a2d46d8c9f5a469fd4e18f2
Signed-off-by: Feng Pan <fpan@redhat.com>
|
|
Changes include:
- IP utility library in python 3 that supports both IPv4 and IPv6
address generation. This library currently includes a single
function of generating IP ranges or single IP for a given CIDR.
More functionality will be added at a later time to support
features such as IP address calculation.
- Updated common-function.sh to use python library to generate IP
ranges. All existing bash functions are preserved, so any
callers will get identical IP ranges as before.
- Add dependency to python3 for opnfv-apex-common package.
- Add python dependency to build.sh
No change is made to interface related functions.
Change-Id: Idc6998754f9f3c7a3868ec5b5768f3bb5f78cd90
Signed-off-by: Feng Pan <fpan@redhat.com>
|
|
JIRA: APEX-75
Change-Id: Iec4abc319db98c32b94a0603811401fd7f9a59db
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
- ensures the storage vols are cleaned up properly
Change-Id: I55c0dbcbbc884658b9e37e34aa85d00e29e99b57
Signed-off-by: Dan Radez <dradez@redhat.com>
|
|
Patch adds helpful utility functions and automatic parsing of heat
output into a readable format indicating where a failure is in a
deployment. Example: http://fpaste.org/354210/
New opnfv-util function allows a user to interact with a deployment
easily:
- opnfv-util undercloud <user> connects user to undercloud VM
- opnfv-util debug-stack parses the heat failure output
Above arguments also accept partial matches.
JIRA: APEX-75
Change-Id: I5ccfee64ee2958de0d00a3b25cd9b29de60c9e20
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
New param introduced into THT that exposes the image to use for the
Controller. The new var is defaulted to use overcloud-control and is
causing master to fail.
Changes Include:
- Adds opnfv-environment yaml file to use for common parameters
required in all deployments
- Defaults controllerImage to be overcloud-full
Change-Id: I5da8921d9c9efd167748a3d6f42866d18f46d2b7
Signed-off-by: Tim Rozet <trozet@redhat.com>
|
|
- fix the undercloud dependancy chain so the tarball gets rolled now
- adding rpm-check jobs for each of the rpms, this runs the build but doesn't spit out the rpm file
- added conditionals for rpm-check commands, only take the time to run them if the Makefile or specs have changed
Change-Id: I9731b278d06d92deb3f69066ba4468e64647582d
Signed-off-by: Dan Radez <dradez@redhat.com>
|
|
- Only support for NOSDN for this patch
- Other SDN controllers will not be tested for this patch
- AODH is now upstream and not needed to be carried in Apex
- puppet-cinder-quota-fix.patch is now upstream and not needed to be carried in Apex
- puppet-neutron-force-metadata.patch is now upstream and not needed to be carried in Apex
- removing the catalog of nic files and creating a templatization for them
JIRA: APEX-108
JIRA: APEX-93
JIRA: APEX-92
Change-Id: Ic1a24baec7ef295e9dc1b8b72ac36ca5b05578ea
Signed-off-by: Dan Radez <dradez@redhat.com>
|
|
- storage mgmt and storage were being update by the storage sed commands
- ext net had a double Ports:: in the line
Change-Id: I95cfdbced1e817e6e69805708b6b4e504abd8c43
Signed-off-by: Dan Radez <dradez@redhat.com>
|
|
This commit is not backwards compatible with cleaning.
VMs named Instack will need to be manually removed.
Change-Id: I6829119e43a0a572da07e6d9757c5208c7fe91d4
|
|
Removes --remove-all-storage option in ci/clean.sh
JIRA: APEX-121
Change-Id: Ia8063112333d12d09cfb16dd1a02e14551dae7d7
Signed-off-by: Feng Pan <fpan@redhat.com>
|
|
Changed the number of compute nodes to no longer be
hard coded to 1 and now calculate to be total_nodes -1
for non-HA deployments.
Rebased to new build.
Change-Id: I8dc135876a8b436714806b79d4193d225761534d
Signed-off-by: randyl <r.levensalor@cablelabs.com>
|
|
- replacing brbm with logical names br-netname
- replacing instack-virt-setup with tripleo scripts
JIRA: APEX-90, APEX-80
Change-Id: I58a15dee8de882e034c8af8a3368ca0647741b13
Signed-off-by: Dan Radez <dradez@redhat.com>
|
|
JIRA: APEX-84
Change-Id: Ic806118c3c3032abff6cebe5c93ad62a6a0a23c7
Signed-off-by: Dan Radez <dradez@redhat.com>
|
|
This patch fixes bug in ci/build.sh that cause the following error
message:
/bin/opnfv-deploy: line 197: [: too many arguments
Change-Id: I92419473262a51806baee849669af9b84b94045c
Signed-off-by: Ryota MIBU <r-mibu@cq.jp.nec.com>
|
|
Change-Id: Id2d6d46a53129d4ae2f93fad835536da067e42df
JIRA: APEX-51
Signed-off-by: Dan Radez <dradez@redhat.com>
|
|
- reverting the patched growfs bug kernel
- updating ODL to RC 3.1
- syntax fix on error inventory syntax check
Change-Id: I4c8f2bd648a2caa1caa2d33650bf1b73824f7179
Signed-off-by: Dan Radez <dradez@redhat.com>
|
|
Change-Id: I38416967c72b03ae443d3ffa9e24fa23ad24ff74
|
|
Change-Id: I618a6d6e762a464751d61c070571df400a3c8eda
|
|
|
|
Change-Id: I515ac692a1a2251f16f4b9e06f94c6a626374285
JIRA: APEX-81, APEX-69
|
|
JIRA: APEX-95
pm_type is already in the sample instackenv.json but wasn't previously
being read.
Change-Id: Ia82c5010481d77d142d4fa3ca6e738558e8e9e4e
|
|
|
|
Change-Id: I034bd9712ed06992dca59e2b362ef535cf3ab4d2
JIRA: APEX-96
|
|
JIRA: APEX-82
Change-Id: I0f2164a67b425ae5c99cd6bfee2182b23867b4cf
|
|
|
|
JIRA: APEX-74
/var/log/messages for each node will be slurped into a new folder in the
stack user's home directory. If debug was passed, it will also be printed
out along with the nova list for that node.
Change-Id: I726da67b076bdf1fb38ba1805456cbd7c7d9e096
|
|
Change-Id: I3474118e58ae4946abfe9e74bb73084aad3634b2
JIRA: APEX-88
|
|
- Forcing no-ha to 1 compute node
- updating install-packages to allow detached deploy
- updating kernel to one with a growfs bug fix
- removing service script with growfs work around
- adding rootpw to overcloud when in debug mode
JIRA: APEX-85, APEX-76, APEX-73
Change-Id: I202275c9cbeb864a044b119a9e440d59a1955c40
|
|
This will persist networks across reboots instead of creating a
transient network.
Change-Id: I98b8e0c577c145197dd7195274b27ded2cc96297
|
|
|
|
|
|
Change-Id: I6c8bd9a949b1b818bedd4994f490da4a703ee6a5
Signed-off-by: Ryota MIBU <r-mibu@cq.jp.nec.com>
|