We have stable/wallaby branch as latest stable and
d-g test only latest stable branch to make sure the compatibility
of legacy jobs. Hope this is last stable we need to update/support
in d-g.
This commit moves the legacy stable job testing to stable/wallaby.
Depends-On: https://review.opendev.org/c/openstack/devstack/+/784974
Change-Id: I5542dca0f8050a6e21aff773f49eb4799ec26828
We have stable/victoria branch as latest stable and
d-g test only latest stable branch to make sure the compatibility
of legacy jobs.
This commit moves the legacy stable job testing to stable/victoria.
Change-Id: I78a0f55c360df2804200cc801109741276248550
We have stable/ussuri branch as latest stable and
d-g test only latest stable branch to make sure the compatibility
of legacy jobs.
This commit moves the legacy stable job testing to stable/ussuri.
Depends-On: https://review.opendev.org/#/c/723605/
Change-Id: Id57e5590a5b24c2d821381f52e1c445cbbd1105b
Now we have stable/train branch ready for devstack
and so does for all service projects.
This commit adds the stable/train job running for
d-g gate.
Change-Id: Ibf278cd21d1a0921f7122deb944e1ae18b7c07c1
Now we have stable/stein branch ready for devstack
and so does for all service projects.
This commit adds the stable/stein job running for
d-g gate.
NOTE: Adding legacy job vesion of stable/stein because
only legacy jobs use devstack-gate which is what we need
to test.
This commit removes the all stable zuulv3 native jobs
which does not use devstack-gate and so does not required
to be tested here.
Change-Id: I546038fa4164a3085b33fa0ef30a51ec237bdc3f
This is a mechanically generated change to replace openstack.org
git:// URLs with https:// equivalents.
This is in aid of a planned future move of the git hosting
infrastructure to a self-hosted instance of gitea (https://gitea.io),
which does not support the git wire protocol at this stage.
This update should result in no functional change.
For more information see the thread at
http://lists.openstack.org/pipermail/openstack-discuss/2019-March/003825.html
Change-Id: I59312b9ee70b4a3a7e9912a011de770d8df0f5a4
This is the same version run by zuul.
Update devstack callback for ansible 2.5.
Change-Id: Ic0d2559346570e60025838166b9bd0081eee331a
Signed-off-by: Paul Belanger <pabelanger@redhat.com>
This job is used only by devstack-gate, import it.
Rename it to devstack-gate-hooks following our naming policy.
Change-Id: Id01d7cda5cb28b6371bdcf8b85208135caa9e9b1
Some clouds we run on no longer have MTUs of 1500 on their interfaces
and are smaller. We dynamically determine the MTU we should use on the
overlay networks. List all "real" interfaces with a permament mac addr
then use the smallest MTU less 50 bytes for vxlan overhead.
We only consider "real" interfaces so that we can run this multiple
times for separate overlays without perpetually decreasing the MTU due
to finding the lower MTU we just configured.
Change-Id: If163bf211570e690e255cc3ef6888d9e0cba19b5
We added it to openstack-zuul-jobs so we can use it as a base job in
the migration.
Use the results of the migrated
gate-tempest-dsvm-neutron-full-ubuntu-xenial to test that the base job
will work for migration purposes.
Depends-On: I3ff2b486702fed9342097cdcd30ffa0c043bdecc
Change-Id: I34514c2c38d2121c6b6fbca7e83bd273f2ec49f0
Move the "fix_disk_layout" role from a big shell script to ansible
steps for the native v3 devstack job.
This change is based on Icd4e6d8ab84471ad06e69c3e0f9bac92776efc78.
(Requires Ansible 2.3 for the parted module)
Change-Id: Ia6dcb13803a549390cad736849a26c92494a914c
Depends-On: I2a59e22a38918ae45e0518b0af85ad6e8bc9423c
Add a role to run devstack, and add it to the main devstack playbook
for the next-gen devstack job.
Change-Id: If84d764e9b2f8dbb6beb1990abc11cbb9f1156b0
This writes the contents of a variable dictionary as a devstack
localrc file. It sorts the output so that references to other
variables may be resolved.
Change-Id: Ib9716317333aecffca627f5a38a95d86fc4b320b
Note, the stack user home directory is different than current, however,
these roles are now diverging from the legacy roles. This matches
the current default devstack behavior.
Change-Id: I9a19579dd159dff38750afa3a0f9c642f2056408
It looks like we may be encountering issues running the second or
third of these commands; combine them into one to reduce system
raciness.
Change-Id: Iacb93868c678161d84bb5c06a8180b8eb12b5fe7
This job attempts to run devstack as close as possible to the manner
in Zuul v2.
Also, don't write the devstack-local conf file if no variables were
supplied.
Change-Id: I308504bf5a48652c62d4d05572c5c077a7dfe368
We've seen at least one occurence of a
"Device or resource busy while setting up superblock" error while
setting up xvde2 which, as far as we know, is a new error.
This problem was not happening before we took out sync/partprobe so
let's put that back in for safe measures.
Related-bug: #1706103
Change-Id: Iad1a75aa5fb4ae5a2f4a9cad6373091124917c24
This re-introduces the Ansible role to set up the disk layout.
The initial introduction increased dramatically the occurrence of the
race condition involving the failure when configuring swap on /dev/xvde1 [1].
This (re)revert substitutes the sync and partprobe commands by a
udevadm settle command in order to wait until the device and it's partitions
are ready for use.
[1]: https://bugs.launchpad.net/openstack-gate/+bug/1706103
This reverts commit 71aa1bd6cf.
Change-Id: Iac0efdb6077ab18c2e1991820da4fcea7642f708
This seems to have increased the rate at which the missing partition issue is occurring, let's revert for now while we figure this issue out.
This reverts commit 279ba7c1d5.
Change-Id: Idbfa6ec321fe8dd96b68b9d7213aab530e0f5381
We're moving it (almost) verbatim to a shell task for the time being
and we can iterate to separate it into smaller proper Ansible tasks
later.
Change-Id: Iddc5433922f61f399e342ac459456b9178028ada
Add some introspection to the devstack playbook so we can see the
effects as we construct the job.
Change-Id: Iec5017a708f239859d74617cfee6896a2258244d
There are apparently cases where the ansible uri module won't have a
status item in its result dict. Check that status is there before
comparing against its value when doing network sanity checks.
Without this the following issue can occur:
2017-04-17 03:33:34.536 | localhost | FAILED! => {
"failed": true,
"msg": "The conditional check 'uri_result['status'] == 200' failed. The error was: error while evaluating conditional (uri_result['status'] == 200): 'dict object' has no attribute 'status'"
}
Change-Id: I5c848bfbb955b7201abcef71c2cbba2ec0644b1a
Simplify the network sanity check using the new /etc/ci/mirror_info.sh
script. If thats present we use it to check pings to the mirror host and
HTTP against the pypi mirror. If it isn't present we don't bother. Note
this makes the sanity check fail the job again if it doesn't work which
should be fine at this point because /etc/ci/mirror_info.sh should only
exist if you have a mirror to provide information about.
Change-Id: Ifddb7c5b7ae3e446005338c9679dd5441bc57062
It is possible for 3rdparty CI not to have these settings. It was
suggested we simply ignore errors for now on this.
Change-Id: I188684b34e322d99b52c8fbdab826c6e11eca81f
Signed-off-by: Paul Belanger <pabelanger@redhat.com>
Many CI systems run pypi mirrors on nonstandard ports and the
"ping" sanity check fails to ping when a port number is present
in the pypi url.
Change-Id: I411848aa8aa4713a4c649c54f7f3490b29fd36fc
This information was once collected for gate jobs, but was removed by
I916e075d8ed3ee6079e4c4621f6eb10f1d30bf95 rework. Turned out that the
info collected by ansible itself does not provide some crucial data,
like cpu flags supported by the machine that may become handy when
debugging some performance related gate issues.
This patch restores collecting the info, using ansible.
Change-Id: Icd2bc68eb723ba2f3fc0d8a113fb247bc478121a
Replace the network sanity check bash logic for the
network_sanity_check role.
This is the last step of replacing the setup_host bash function,
thus removing also the function altogether.
Change-Id: I8814cf284e572dda66455eea7a10f61037979a9b
Ghanshyam Mann