Age | Commit message (Collapse) | Author | Files | Lines |
|
|
|
|
|
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>
|
|
1. deploy.py: control the work flow of the deployment
2. daisy_server.py: maintain the ssh connection with daisy server
and execute ssh commands
3. environment.py: create/find/delete the nodes/vms, install operating system
and openstack on nodes
4. libvirt_utils.py: deal with the vm templates and call virsh commands
5. utils.py: some common functions such as file/directory/bash operation
Change-Id: I1caa4b0b3118665e15410e8f02bcb6473e5a530b
Signed-off-by: Alex Yang <yangyang1@zte.com.cn>
|
|
|
|
This print out error message of docker commands to find out the
reason of error like[1]
[1] https://build.opnfv.org/ci/job/daisy-build-daily-master/239/consoleFull
Change-Id: Ic1bd85d999dbe584764bc9a05d22579835e55516
Signed-off-by: Zhijiang Hu <hu.zhijiang@zte.com.cn>
|
|
1.add multicast flag in daisy.conf
2.sleep40 for trustme.sh execute ok
3.modify cpu mode of xml file to make it much more general
4.fix check_openstack_progress parameter error
Change-Id: Ic150698ede448b7651e95d129aeb7d97a8f34309
Signed-off-by: zhouya <zhou.ya@zte.com.cn>
|
|
|
|
Now the daisycloud-core is a big repo because there are some big files
in the git history. Use "--depth 1" to reduce the amount to download.
Change-Id: I8ce0dc6675d2239a126bcf558300f1ad45cd3fb3
Signed-off-by: Alex Yang <yangyang1@zte.com.cn>
|
|
put the kolla nova-compute.conf file update handle to prepare/
execute.py for unify the all relate kolla config in py way.
Change-Id: Iac585df97d2855038a83f9bfdadfb9e449660c9c
Signed-off-by: zhongjun <zhong.jun@zte.com.cn>
|
|
This add "losetup" before "kpartx -l" command, otherwise, we encountered
many "Buffer I/O error on dev loop0p1, logical block 104854018,
async page read" in dmesg. But they are seems no harm.
No need to do "losetup -d" after "losetup", because "kpartx -dv"
will do that for us.
Change-Id: I326ef35a241d8b09f0f3f3fc90ef749df3f4768f
Signed-off-by: Zhijiang Hu <hu.zhijiang@zte.com.cn>
|
|
|
|
|
|
|
|
There's an mistake in patch 33327.
See https://gerrit.opnfv.org/gerrit/#/c/33327/
The cleanup function was called twice, so the image file was deleted.
Change-Id: I5fe6a90b46739d1c6d8b66365d2c8e55350cde00
Signed-off-by: Alex Yang <yangyang1@zte.com.cn>
|
|
1. Move an log message to the correct location(now at line 331).
In baremetal deployment, the virtual-deploy message is printed.
2. Modify all the "===" in the log messages to the same number of "="
3. Remove the TODO list since the work finished
Change-Id: I34325c522036caf9d1aa58c9cbf30eb77843fdfc
Signed-off-by: Alex Yang <yangyang1@zte.com.cn>
|
|
Since we can not tolerate any kind of error which may result in
multicasting failed to run. So even the usage error shall be
treated as error too.
Change-Id: I5483bb2cebf437ed33ebb79eeeac640f8728a73f
Signed-off-by: Zhijiang Hu <hu.zhijiang@zte.com.cn>
|
|
|
|
the daisy install tmp workdir can be configurated as an input
argument in deploy.sh,and it also should be used in subshell
daisy-img-modify.sh,so add the input argument -w workdir
in daisy-img-modify.sh.
1.add an input argument -w workdir in daisy-img-modify.sh.
2.modify the WORKDIR default value to /tmp/workdir/daisy.
3.code refactor to put the centos7.qcow2 file clean statement
from the deploy.sh to daisy-img-modify.sh.
Change-Id: Id375a15ad2839c209329e644c5e032d044604e7d
Signed-off-by: zhongjun <zhong.jun@zte.com.cn>
|
|
|
|
Daisy4NFV project."
|
|
Alex has been contributing the majority of code to Daisy4NFV since the
last season[1][2].
I have had a talk with him on IRC meeting [3] and he expressed willing to
contribute more to Daisy4NFV project and I believe he will make a great
addition to the core review team.
All Daisy4NFV committers, please vote +1 or +2 if you support this nomination [4].
This PS also propose a committer removal. Jian Yang (yangjianyjy@chinamobile.com)
has been inactive in the project for a long time. He
- was not present in the project meetings for more than 6 months
(see participant lists in the minutes or IRC); AND
- did not contribute any patch for the project for more than 6 months; AND
- did not participate in the email discussion of the project for more than 6 month AND
- We failed to contact him offline.
[1] http://stackalytics.com/?project_type=opnfv-group&metric=commits&release=all&module=daisy
[2] https://gerrit.opnfv.org/gerrit/#/q/owner:yangyang1%2540zte.com.cn+status:merged+project:daisy
[3] http://eavesdrop.openstack.org/meetings/daisycloud/2017/daisycloud.2017-03-24-07.58.log.html
[4] https://wiki.opnfv.org/display/DEV/Committer+Promotions
Change-Id: Ibd416d93d01c7fe769dfc0c5809baca4cdefca46
Signed-off-by: Zhijiang Hu <hu.zhijiang@zte.com.cn>
|
|
Change-Id: Ic9e19d4e120fc53d96d0794239cd6e421f25ea27
Signed-off-by: zhouya <zhou.ya@zte.com.cn>
|
|
Change-Id: I50d5750e985b258cef7874c88902b8fae0539e25
Signed-off-by: Zhijiang Hu <hu.zhijiang@zte.com.cn>
|
|
|
|
Also add code/jasmine/.tarball-version to set version manually,
otherwise we need more software such as git to get the version and
also we may encounter rpmbuild error due to rpmbuild does not
support dash character in git version.
Change-Id: If9b7ef75cbc0a5e4987658454b47abf02b451989
Signed-off-by: Zhijiang Hu <hu.zhijiang@zte.com.cn>
|
|
Change-Id: Ic33381ef3339bdcec82cba7afbfc2d32f6e88be4
Signed-off-by: zhouya <zhou.ya@zte.com.cn>
|
|
|
|
refactor the deploy.sh variable name
1.unify the glabs var name to upper case.
2.refactor the variable of target_node_net/daisy_server_net
to add prefix VMDEPLOY for code readablity.
3.refactor the usage() to function usage for uniform the
code style.
Change-Id: Ibd7ea91ac8b19cd7147e3a7d97b3359880cec59c
Signed-off-by: zhongjun <zhong.jun@zte.com.cn>
|
|
Change-Id: I7a543019c8d92314ef549bf72369b7276f39577d
Signed-off-by: Zhijiang Hu <hu.zhijiang@zte.com.cn>
|
|
|
|
Change-Id: I3f9b9167864126b7b455761799dc79c40c394854
Signed-off-by: Zhijiang Hu <hu.zhijiang@zte.com.cn>
|
|
|
|
when we deploy the baremetal,do not have to use pxe_only
and skip_pxe_ipmi parameters,or we can't call 'daisy install'
command
Change-Id: Ief5131c47c16412b8dacec50b5a7858b65baa7a2
Signed-off-by: zhouya <zhou.ya@zte.com.cn>
|
|
A column named 'Tecs_version_id' is added in the upstream patch
https://review.openstack.org/#/c/450056/.
So the column numbers of 'Role_progress' and 'Role_status' should
be modified in check_openstack_progress.sh
Change-Id: I47607bfe3ab89195569f785b3f815e57cee0406d
Signed-off-by: Alex Yang <yangyang1@zte.com.cn>
|
|
|
|
Change-Id: I71255872e6bcf485f8ab72c105db931d96233d03
Signed-off-by: zhouya <zhou.ya@zte.com.cn>
|
|
Change-Id: If783d387c29dad7fa472578ead2bcd5a6d42ee70
Signed-off-by: SerenaFeng <feng.xiaowei@zte.com.cn>
|
|
|
|
|
|
In prepare_install function of tempest.py:
1. get_config.py needs dha and network parameters
2. install_os_for_vm_step2 function needs cluster_id parameter
Change-Id: Idb13f71ced76f0d99dcbe818cdac3d3f2eb7d5df
Signed-off-by: Alex Yang <yangyang1@zte.com.cn>
|
|
Change-Id: Idd960d7faf0f24ae8caac13562858ccaa1c67f7b
Signed-off-by: SerenaFeng <feng.xiaowei@zte.com.cn>
|
|
Change-Id: If3a1d1a8b74e11c5861c016bd722b78f4cf9b475
Signed-off-by: SerenaFeng <feng.xiaowei@zte.com.cn>
|
|
|
|
according to https://review.openstack.org/#/c/446820/
Change-Id: Ie5ece043ac27e207e232cb05b8b007fa6c1b04c5
Signed-off-by: Yao Lu <lu.yao135@zte.com.cn>
|
|
Change-Id: I2a911fc15c1456b409db840b9ae76c04a23d449d
Signed-off-by: SerenaFeng <feng.xiaowei@zte.com.cn>
|
|
|
|
|
|
|
|
|