Age | Commit message (Collapse) | Author | Files | Lines |
|
This PS:
1. Build kolla image in VM
2. Use sha512sum to check kolla image
3. Delete suffix dot when there is no ext_tag
Change-Id: I94d99e3ff69c47cc247051fae812bfdb29ff0246
Signed-off-by: Zhijiang Hu <hu.zhijiang@zte.com.cn>
(cherry picked from commit 377cae861b2c04d8934786acc4d07b41ba98acc5)
|
|
Change-Id: I6ce45197e01d278a41b9472d0bb92f037ab7d0c6
Signed-off-by: Zhijiang Hu <hu.zhijiang@zte.com.cn>
(cherry picked from commit b47b3753c63ba9d28d5a11efb1bb4ce3932d3f1a)
|
|
Otherwise, we encountered:
File target/el7/noarch/opnfv-2017-10-11_09-58-21.bin is larger than 4GiB-1.
-allow-limited-size was not specified. There is no way do represent this file size. Aborting.
Change-Id: Ied43c9e0c0e635ff52740c18b3b3001a4e6a3604
Signed-off-by: Zhijiang Hu <hu.zhijiang@zte.com.cn>
(cherry picked from commit 3580a4c2bf8596dc94ab92619b6185757725f566)
|
|
Change-Id: I6ddfe8e291ea8f6d673bd82c4df310d63f61a130
Signed-off-by: Zhijiang Hu <hu.zhijiang@zte.com.cn>
(cherry picked from commit 47d0995307685d7d3903fd909e6b1e7ad4311649)
|
|
|
|
stable/euphrates
|
|
This is a bug fix of 02e1f1c8ab3f3857ca2370f98575ccbf41f12a01
Change-Id: I628c5e33a76bc28e0af139e338b29a7393700525
Signed-off-by: Zhijiang Hu <hu.zhijiang@zte.com.cn>
(cherry picked from commit 15405530cd0b27f7e2f27d5c889b561569057057)
|
|
|
|
1. Eliminate hard code about pxe bridge
2. Replace the bridge in daisy VM template with the pxe bridge
provided by jenkins
Change-Id: I7a8463181847417d10705ce9ff4db6e68d505c6d
Signed-off-by: Alex Yang <yangyang1@zte.com.cn>
(cherry picked from commit 56975272e92efa3f5f6f719028dc147c1d90e9ce)
|
|
Change-Id: I5caadb42b3890858e7642a5fd260392370dbbfc8
Signed-off-by: Zhijiang Hu <hu.zhijiang@zte.com.cn>
(cherry picked from commit 02e1f1c8ab3f3857ca2370f98575ccbf41f12a01)
|
|
Sometimes the VMs do not reboot when using "virsh reboot".
The virtual depoyment CI jobs failed recently.
Change-Id: Id8b3ef638a7ba8f584c6e4c03d11e77d2919a35b
Signed-off-by: Alex Yang <yangyang1@zte.com.cn>
(cherry picked from commit 799e1323c05853e83ba01a54ac42818b4309c591)
|
|
Change-Id: I642d130566137dff727b7227658afe6ace9473d6
Signed-off-by: Alex Yang <yangyang1@zte.com.cn>
(cherry picked from commit 94d3957c5aba2db70d06929ac9538280a71c6641)
|
|
Change-Id: I6514292b16c784ced0dcb82d639fe10b425d63e2
Signed-off-by: Zhijiang Hu <hu.zhijiang@zte.com.cn>
|
|
Kolla offcial image tag is, for example, 4.0.2. Daisy images can
be names as 4.0.2.n And Kolla supports upgrading not only from
4.0.2 to 4.0.3, but also from 4.0.2 to 4.0.2.n. It is a convenient
way for Daisy to publish its own small changes into images for
upgrading.
Change-Id: I9d283e1ffa452ac3e04046ab6c96d018d42e0bf9
Signed-off-by: Zhijiang Hu <hu.zhijiang@zte.com.cn>
|
|
Change-Id: I023a637e793bae845feabed0d7b80072d3f9f0b7
Signed-off-by: Zhijiang Hu <hu.zhijiang@zte.com.cn>
|
|
This can be seen as a code clean up before we implementing
running deploy.sh without recreate daisy vm and networks.
Change-Id: I9bb67fdc7220cc70e2e876cc05750c22630d5ac8
Signed-off-by: Zhijiang Hu <hu.zhijiang@zte.com.cn>
|
|
JIRA: DAISY-42
Change-Id: I0fd709bb0dbee42cdc73076773cb635be6ba02cd
Signed-off-by: Alex Yang <yangyang1@zte.com.cn>
|
|
|
|
|
|
Rename all scenarios with "odl_l3" in their name in Euphrates to
just "odl".
Daisy will keep "odl_l3" and "odl_l2" (in code, not for user)
for further reference.
Change-Id: Ib762dd808d4f9467b0e6827b8bbed6d9df7e0e0e
Signed-off-by: Zhijiang Hu <hu.zhijiang@zte.com.cn>
|
|
JIRA: DAISY-42
JIRA: DAISY-56
In bare metal deployment, we can use PDF to get MAC addresses of nodes
https://gerrit.opnfv.org/gerrit/#/c/38387/.
Then we can use the MACs to help to distinguish the discovered nodes
and assign roles to them, like virtual deployment in the link
https://gerrit.opnfv.org/gerrit/#/c/38381/.
Change-Id: Ib0f1a60b8935f528a828f716ccc916b767cfa6f9
Signed-off-by: Alex Yang <yangyang1@zte.com.cn>
|
|
|
|
There are some error messages in virtual deployment, such as
"Domain not found" or "network is already active".
The script forgets to destroy the keep-alived network, and should
not destroy a non-existent VM or network.
Change-Id: I8d9dce9d70f732bd6942b293e407e1845d81fc39
Signed-off-by: Alex Yang <yangyang1@zte.com.cn>
|
|
|
|
Change-Id: I9764612171ef3bf2cdfc652420a2b162fcbfab43
Signed-off-by: zhouya <zhou.ya@zte.com.cn>
|
|
JIRA: DAISY-56
controller01 -> $name
Change-Id: Icd959ca55079a6ac0bfbd181ff134d7decfb89f5
Signed-off-by: Alex Yang <yangyang1@zte.com.cn>
|
|
JIRA: DAISY-56
Now the roles are assigned to nodes randomly, because the function
'add_hosts_interface' in tempest.py just uses zip to map the
host's name to the host.
libvirt_utils.py: get mac addresses from VM
environment.py: save the mac addresses
daisy_server.py: write a new deploy.yml which contains the mac addresses
and copy the file to daisy server
get_conf.py: get mac addresses from the new deploy.yml
tempest.py: assigned roles to nodes when the mac addresses matched
controller.xml: increase the RAM to make difference with computer nodes
deploy.sh: apply this change to bash script
Change-Id: Ia61b60d39d319c5d01e3505727fafc63a0585858
Signed-off-by: Alex Yang <yangyang1@zte.com.cn>
|
|
just move the code from get_para_from_deploy to get_conf file
no functionality change.
Change-Id: I86aa1325ff37cb2ae0784c9487e62e95cc23f644
Signed-off-by: zhouya <zhou.ya@zte.com.cn>
|
|
|
|
From [1], We got an error from docker: "No such container: daisy"
while issuing "docker rm -v -f daisy", this OK because we do
docker run --rm before this. So it is safe to add "|| true" after
"docker rm -v -f daisy".
[1] https://build.opnfv.org/ci/job/daisy-build-daily-master/500/console
Change-Id: I3d17595156f1b6181a84d9a03e2cd6ddff275eb3
Signed-off-by: Zhijiang Hu <hu.zhijiang@zte.com.cn>
|
|
Switch to upstream daisy's stable/ocata branch
Change-Id: I5ff0b0a28a8d2f76f0cb813af8f8241175bb6054
Signed-off-by: Zhijiang Hu <hu.zhijiang@zte.com.cn>
|
|
Enable HA scenario options due to support HA function in upsteam.
Change-Id: Ie1889afc1a149f171a9b324eb284fd650baed397
Signed-off-by: zhongjun <zhong.jun@zte.com.cn>
|
|
Change-Id: I0157bf8b6fa9be254c61bb384065f80107ab3dda
Signed-off-by: root <zhou.ya@zte.com.cn>
|
|
Project job deployment job uses os-nosdn-nofeature-ha currently,
so we need to treat os-nosdn-nofeature-ha as a valid scenario
to let the job continue to work.
Change-Id: Ib9311ada9b043b1f695f43edb51adbb3714d3356
Signed-off-by: Zhijiang Hu <hu.zhijiang@zte.com.cn>
|
|
Add the scenario argument valid check in deploy.sh.
Change-Id: Ifeeed3882b22ba379975c2356d761e0536c5c2c9
Signed-off-by: zhongjun <zhong.jun@zte.com.cn>
|
|
Change-Id: I759d864efa524c0d564b9d93aa480e155149adaa
Signed-off-by: zhouya <zhou.ya@zte.com.cn>
|
|
|
|
|
|
Change-Id: I0660c016e18491395c7253e5576f8fa1c8aa051e
Signed-off-by: Zhijiang Hu <hu.zhijiang@zte.com.cn>
|
|
Default is os-nosdn-nofeature-noha
Change-Id: I12e70552c426884269c2c7f1bfa05e1db5658bea
Signed-off-by: Zhijiang Hu <hu.zhijiang@zte.com.cn>
|
|
There are error messages "iptables: No chain/target/match by that name"
in baremetal deployment:
https://build.opnfv.org/ci/job/daisy-deploy-daily-master/298/console
Change-Id: I7e2940222fd0a99ee42823a08a285bdd93892fe6
Signed-off-by: Alex Yang <yangyang1@zte.com.cn>
|
|
Change-Id: I654a27042ff9807a502773d48df1c26d12302de7
Signed-off-by: zhouya <zhou.ya@zte.com.cn>
|
|
Change-Id: I179b5aab51958c0cd4653e65bbd74df5cfd7c53e
Signed-off-by: SerenaFeng <feng.xiaowei@zte.com.cn>
|
|
|
|
Change-Id: I3026c4dd83cc19246c19ab568005c4c0b37ff9d9
Signed-off-by: SerenaFeng <feng.xiaowei@zte.com.cn>
|
|
Change-Id: Ib80710c784d384ebc27eb0f51fcb4384017eecca
Signed-off-by: Alex Yang <yangyang1@zte.com.cn>
|
|
The upstream openstack/daisycloud-core supports ocata openstack now.
https://review.openstack.org/#/c/465410/
Change-Id: I14825c80cdd2297e5b0df3680f30fa5c32de3cc4
Signed-off-by: Alex Yang <yangyang1@zte.com.cn>
|
|
|
|
1) This mainly increase PACKETS_PER_BUFFER to 65536 to reduce the
frequency of TCP client acks.
2) Also kills TCP_BUFF_SIZE and define each buffer size in a
more intuitive way.
3) Free more unused memory to prevent being killed by oom-killer
after enlarged PACKETS_PER_BUFFER.
4) Increase client's select() timeout to 20 secs, since we encountered
timeout due to CPU busy in the same BM but with 20 VMs.
Tested this PS in a 10 VM node env, and it can multicast a 2.7G
file to 10 VMs in 6 minutes, while unicast needs 30+ minutes.
Change-Id: Iaf862fb1f1259cc770f720ccdd95dcc281aef262
Signed-off-by: Zhijiang Hu <hu.zhijiang@zte.com.cn>
|
|
the physical network device of external network is hardcode to
'physnet1',change it to obtain from network.yml configuration
file.
Change-Id: Id2e45ac488619db2247e73cc3fed5706db31d9e9
Signed-off-by: zhongjun <zhong.jun@zte.com.cn>
|