Friday, 2025-02-14

« Thursday, 2025-02-13 Index Saturday, 2025-02-15 »
@bennetefx:matrix.orgHi Clark, it seems like this issue is only there for one of my non-ephemeral static node. I guess the service has died on this node and subsequent jobs that are being run on this node does not restart this service (it's supposed to right?).02:16
-@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/zuul] 941620: Use change networks in gerrit driver queries https://review.opendev.org/c/zuul/zuul/+/94162003:20
@jim:acmegating.combennetefx: put it in the first pre-run playbook of your base job so that every job will start it if it isn't already running.  it noops if it's already running.03:21
@bennetefx:matrix.orgYeah I did, added the **start_zuul_console** role as a pre-run of my base job and it is working fine now. Thank you.03:38
-@gerrit:opendev.org- Simon Westphahl proposed: [zuul/zuul] 941626: Fix Gerrit change query retry exceeded https://review.opendev.org/c/zuul/zuul/+/94162606:38
-@gerrit:opendev.org- Simon Westphahl proposed: [zuul/zuul] 941626: Fix Gerrit change query retry exceeded https://review.opendev.org/c/zuul/zuul/+/94162606:42
-@gerrit:opendev.org- Simon Westphahl proposed: [zuul/zuul] 941626: Fix Gerrit change query retry exceeded https://review.opendev.org/c/zuul/zuul/+/94162606:52
@cidlik:matrix.orgYeap, partially. But key words are "different jobs" here. I wouldn't want to breed similar jobs which have different names and nodeset only. It's hard to support. I would prefer to have one job for multiple nodes. Or to run one job several times but for different nodes in one buildset.07:13
@joao15130:matrix.orgHello, new day new problem.09:26
We have observed some out-of-sync between nodepool and our openstack env.
From time to time nodepool list images while openstack nova doesn't show anything. The images remain available in nodepool and zuul picks something which doesn't really exist.
Even worse, we can see instances created by nodepool deleted without any reason, and the instances remain visible from nodepool. I wasn't able to see any errors in openstack logs.
Our env is based upon a devstack AIO installation and we have registered this cloud as a provider.
images-dir: /etc/nodepool/dib
build-log-dir: /etc/nodepool/dib_log
elements-dir: /etc/nodepool/elements
build-log-retention: 7
webapp:
port: 8005
listen_address: 0.0.0.0
zookeeper-servers:
- host: zk
port: 2281
zookeeper-tls:
cert: /var/certs/certs/client.pem
key: /var/certs/keys/clientkey.pem
ca: /var/certs/certs/cacert.pem
labels:
- name: nodepool-jammy
min-ready: 1
diskimages:
- name: nodepool-jammy
elements:
- ubuntu-minimal
- vm
- simple-init
- growroot
- cache-devstack
- openstack-repos
- nodepool-base
- infra-package-needs
- zuul-worker
release: jammy
env-vars:
GIT_BASE: http://opendev.org
DIB_DEBIAN_COMPONENTS: 'main,universe'
DIB_APT_LOCAL_CACHE: '0'
DIB_TMP: '/opt/dib_tmp'
DIB_DISABLE_APT_CLEANUP: '1'
DIB_DEBOOTSTRAP_EXTRA_ARGS: '--no-check-gpg'
TMPDIR: /root/
DIB_CHECKSUM: '1'
DIB_IMAGE_CACHE: /opt/dib_cache
DIB_GRUB_TIMEOUT: '0'
DIB_SHOW_IMAGE_USAGE: '1'
ZUUL_USER_SSH_PUBLIC_KEY: |
ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIKGWg2C9M/BkwRawt6143wKiO4or18fW42hkXkZ6xuaQ root@9cab794c8bc6
providers:
- name: ci-cloud
driver: openstack
cloud: ci-cloud
boot-timeout: 600
launch-timeout: 3600
launch-retries: 3
clean-floating-ips: true
diskimages:
- name: nodepool-jammy
config-drive: true
pools:
- name: dell-openstack-ci
auto-floating-ip: true
max-servers: 4
security-groups:
- nodepool-secgroup
networks:
- nodepool
labels:
- name: nodepool-jammy
diskimage: nodepool-jammy
flavor-name: m1.large
```
images-dir: /etc/nodepool/dib
build-log-dir: /etc/nodepool/dib_log
elements-dir: /etc/nodepool/elements
build-log-retention: 7
webapp:
port: 8005
listen_address: 0.0.0.0
zookeeper-servers:
- host: zk
port: 2281
zookeeper-tls:
cert: /var/certs/certs/client.pem
key: /var/certs/keys/clientkey.pem
ca: /var/certs/certs/cacert.pem
labels:
- name: nodepool-jammy
min-ready: 1
diskimages:
- name: nodepool-jammy
elements:
- ubuntu-minimal
- vm
- simple-init
- growroot
- cache-devstack
- openstack-repos
- nodepool-base
- infra-package-needs
- zuul-worker
release: jammy
env-vars:
GIT_BASE: http://opendev.org
DIB_DEBIAN_COMPONENTS: 'main,universe'
DIB_APT_LOCAL_CACHE: '0'
DIB_TMP: '/opt/dib_tmp'
DIB_DISABLE_APT_CLEANUP: '1'
DIB_DEBOOTSTRAP_EXTRA_ARGS: '--no-check-gpg'
TMPDIR: /root/
DIB_CHECKSUM: '1'
DIB_IMAGE_CACHE: /opt/dib_cache
DIB_GRUB_TIMEOUT: '0'
DIB_SHOW_IMAGE_USAGE: '1'
ZUUL_USER_SSH_PUBLIC_KEY: |
ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIKGWg2C9M/BkwRawt6143wKiO4or18fW42hkXkZ6xuaQ root@9cab794c8bc6
providers:
- name: ci-cloud
driver: openstack
cloud: ci-cloud
boot-timeout: 600
launch-timeout: 3600
launch-retries: 3
clean-floating-ips: true
diskimages:
- name: nodepool-jammy
config-drive: true
pools:
- name: dell-openstack-ci
auto-floating-ip: true
max-servers: 4
security-groups:
- nodepool-secgroup
networks:
- nodepool
labels:
- name: nodepool-jammy
diskimage: nodepool-jammy
flavor-name: m1.large
```
@joao15130:matrix.org* Hello, new day new problem.09:27
We have observed some out-of-sync between nodepool and our openstack env.
From time to time nodepool list images while openstack nova doesn't show anything. The images remain available in nodepool and zuul picks something which doesn't really exist.
Even worse, we can see instances created by nodepool deleted without any reason, and the instances remain visible from nodepool. I wasn't able to see any errors in openstack logs.
Our env is based upon a devstack AIO installation and we have registered this cloud as a provider.
images-dir: /etc/nodepool/dib
build-log-dir: /etc/nodepool/dib\_log
elements-dir: /etc/nodepool/elements
build-log-retention: 7
webapp:
port: 8005
listen\_address: 0.0.0.0
zookeeper-servers:
- host: zk
port: 2281
zookeeper-tls:
cert: /var/certs/certs/client.pem
key: /var/certs/keys/clientkey.pem
ca: /var/certs/certs/cacert.pem
labels:
- name: nodepool-jammy
min-ready: 1
diskimages:
- name: nodepool-jammy
elements:
- ubuntu-minimal
- vm
- simple-init
- growroot
- cache-devstack
- openstack-repos
- nodepool-base
- infra-package-needs
- zuul-worker
release: jammy
env-vars:
GIT\_BASE: http://opendev.org
DIB\_DEBIAN\_COMPONENTS: 'main,universe'
DIB\_APT\_LOCAL\_CACHE: '0'
DIB\_TMP: '/opt/dib\_tmp'
DIB\_DISABLE\_APT\_CLEANUP: '1'
DIB\_DEBOOTSTRAP\_EXTRA\_ARGS: '--no-check-gpg'
TMPDIR: /root/
DIB\_CHECKSUM: '1'
DIB\_IMAGE\_CACHE: /opt/dib\_cache
DIB\_GRUB\_TIMEOUT: '0'
DIB\_SHOW\_IMAGE\_USAGE: '1'
ZUUL\_USER\_SSH\_PUBLIC\_KEY: |
ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIKGWg2C9M/BkwRawt6143wKiO4or18fW42hkXkZ6xuaQ root@9cab794c8bc6
providers:
- name: ci-cloud
driver: openstack
cloud: ci-cloud
boot-timeout: 600
launch-timeout: 3600
launch-retries: 3
clean-floating-ips: true
diskimages:
- name: nodepool-jammy
config-drive: true
pools:
- name: dell-openstack-ci
auto-floating-ip: true
max-servers: 4
security-groups:
- nodepool-secgroup
networks:
- nodepool
labels:
- name: nodepool-jammy
diskimage: nodepool-jammy
flavor-name: m1.large
```
@joao15130:matrix.org* Hello, new day new problem.09:27
We have observed some out-of-sync between nodepool and our openstack env.
From time to time nodepool list images while openstack nova doesn't show anything. The images remain available in nodepool and zuul picks something which doesn't really exist.
Even worse, we can see instances created by nodepool deleted without any reason, and the instances remain visible from nodepool. I wasn't able to see any errors in openstack logs.
Our env is based upon a devstack AIO installation and we have registered this cloud as a provider.
```
images-dir: /etc/nodepool/dib
build-log-dir: /etc/nodepool/dib\_log
elements-dir: /etc/nodepool/elements
build-log-retention: 7
webapp:
port: 8005
listen\_address: 0.0.0.0
zookeeper-servers:
- host: zk
port: 2281
zookeeper-tls:
cert: /var/certs/certs/client.pem
key: /var/certs/keys/clientkey.pem
ca: /var/certs/certs/cacert.pem
labels:
- name: nodepool-jammy
min-ready: 1
diskimages:
- name: nodepool-jammy
elements:
- ubuntu-minimal
- vm
- simple-init
- growroot
- cache-devstack
- openstack-repos
- nodepool-base
- infra-package-needs
- zuul-worker
release: jammy
env-vars:
GIT\_BASE: http://opendev.org
DIB\_DEBIAN\_COMPONENTS: 'main,universe'
DIB\_APT\_LOCAL\_CACHE: '0'
DIB\_TMP: '/opt/dib\_tmp'
DIB\_DISABLE\_APT\_CLEANUP: '1'
DIB\_DEBOOTSTRAP\_EXTRA\_ARGS: '--no-check-gpg'
TMPDIR: /root/
DIB\_CHECKSUM: '1'
DIB\_IMAGE\_CACHE: /opt/dib\_cache
DIB\_GRUB\_TIMEOUT: '0'
DIB\_SHOW\_IMAGE\_USAGE: '1'
ZUUL\_USER\_SSH\_PUBLIC\_KEY: |
ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIKGWg2C9M/BkwRawt6143wKiO4or18fW42hkXkZ6xuaQ root@9cab794c8bc6
providers:
- name: ci-cloud
driver: openstack
cloud: ci-cloud
boot-timeout: 600
launch-timeout: 3600
launch-retries: 3
clean-floating-ips: true
diskimages:
- name: nodepool-jammy
config-drive: true
pools:
- name: dell-openstack-ci
auto-floating-ip: true
max-servers: 4
security-groups:
- nodepool-secgroup
networks:
- nodepool
labels:
- name: nodepool-jammy
diskimage: nodepool-jammy
flavor-name: m1.large
```
```
@joao15130:matrix.orgHello, new day new problem.09:28
We have observed some out-of-sync between nodepool and our openstack env.
From time to time nodepool list images while openstack nova doesn't show anything. The images remain available in nodepool and zuul picks something which doesn't really exist.
Even worse, we can see instances created by nodepool deleted without any reason, and the instances remain visible from nodepool. I wasn't able to see any errors in openstack logs.
Our env is based upon a devstack AIO installation and we have registered this cloud as a provider.
```
images-dir: /etc/nodepool/dib
build-log-dir: /etc/nodepool/dib\_log
elements-dir: /etc/nodepool/elements
build-log-retention: 7
webapp:
port: 8005
listen\_address: 0.0.0.0
zookeeper-servers:
- host: zk
port: 2281
zookeeper-tls:
cert: /var/certs/certs/client.pem
key: /var/certs/keys/clientkey.pem
ca: /var/certs/certs/cacert.pem
labels:
- name: nodepool-jammy
min-ready: 1
diskimages:
- name: nodepool-jammy
elements:
- ubuntu-minimal
- vm
- simple-init
- growroot
- cache-devstack
- openstack-repos
- nodepool-base
- infra-package-needs
- zuul-worker
release: jammy
env-vars:
GIT\_BASE: http://opendev.org
DIB\_DEBIAN\_COMPONENTS: 'main,universe'
DIB\_APT\_LOCAL\_CACHE: '0'
DIB\_TMP: '/opt/dib\_tmp'
DIB\_DISABLE\_APT\_CLEANUP: '1'
DIB\_DEBOOTSTRAP\_EXTRA\_ARGS: '--no-check-gpg'
TMPDIR: /root/
DIB\_CHECKSUM: '1'
DIB\_IMAGE\_CACHE: /opt/dib\_cache
DIB\_GRUB\_TIMEOUT: '0'
DIB\_SHOW\_IMAGE\_USAGE: '1'
ZUUL\_USER\_SSH\_PUBLIC\_KEY: |
ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIKGWg2C9M/BkwRawt6143wKiO4or18fW42hkXkZ6xuaQ root@9cab794c8bc6
providers:
- name: ci-cloud
driver: openstack
cloud: ci-cloud
boot-timeout: 600
launch-timeout: 3600
launch-retries: 3
clean-floating-ips: true
diskimages:
- name: nodepool-jammy
config-drive: true
pools:
- name: dell-openstack-ci
auto-floating-ip: true
max-servers: 4
security-groups:
- nodepool-secgroup
networks:
- nodepool
labels:
- name: nodepool-jammy
diskimage: nodepool-jammy
flavor-name: m1.large
```
@joao15130:matrix.org* Hello, new day new problem.09:28
We have observed some out-of-sync between nodepool and our openstack env.
From time to time nodepool list images while openstack nova doesn't show anything. The images remain available in nodepool and zuul picks something which doesn't really exist.
Even worse, we can see instances created by nodepool deleted without any reason, and the instances remain visible from nodepool. I wasn't able to see any errors in openstack logs.
Our env is based upon a devstack AIO installation and we have registered this cloud as a provider.
```
images-dir: /etc/nodepool/dib
build-log-dir: /etc/nodepool/dib_log
elements-dir: /etc/nodepool/elements
build-log-retention: 7
webapp:
port: 8005
listen_address: 0.0.0.0
zookeeper-servers:
- host: zk
port: 2281
zookeeper-tls:
cert: /var/certs/certs/client.pem
key: /var/certs/keys/clientkey.pem
ca: /var/certs/certs/cacert.pem
labels:
- name: nodepool-jammy
min-ready: 1
diskimages:
- name: nodepool-jammy
elements:
- ubuntu-minimal
- vm
- simple-init
- growroot
- cache-devstack
- openstack-repos
- nodepool-base
- infra-package-needs
- zuul-worker
release: jammy
env-vars:
GIT_BASE: http://opendev.org
DIB_DEBIAN_COMPONENTS: 'main,universe'
DIB_APT_LOCAL_CACHE: '0'
DIB_TMP: '/opt/dib_tmp'
DIB_DISABLE_APT_CLEANUP: '1'
DIB_DEBOOTSTRAP_EXTRA_ARGS: '--no-check-gpg'
TMPDIR: /root/
DIB_CHECKSUM: '1'
DIB_IMAGE_CACHE: /opt/dib_cache
DIB_GRUB_TIMEOUT: '0'
DIB_SHOW_IMAGE_USAGE: '1'
ZUUL_USER_SSH_PUBLIC_KEY: |
ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIKGWg2C9M/BkwRawt6143wKiO4or18fW42hkXkZ6xuaQ root@9cab794c8bc6
providers:
- name: ci-cloud
driver: openstack
cloud: ci-cloud
boot-timeout: 600
launch-timeout: 3600
launch-retries: 3
clean-floating-ips: true
diskimages:
- name: nodepool-jammy
config-drive: true
pools:
- name: dell-openstack-ci
auto-floating-ip: true
max-servers: 4
security-groups:
- nodepool-secgroup
networks:
- nodepool
labels:
- name: nodepool-jammy
diskimage: nodepool-jammy
flavor-name: m1.large
```
-@gerrit:opendev.org- Dong Zhang proposed:09:35
- [zuul/zuul] 940872: Implement keystore functions for OIDC RS256 https://review.opendev.org/c/zuul/zuul/+/940872
- [zuul/zuul] 941629: Use ZuulTreeCache for OIDC signing keys https://review.opendev.org/c/zuul/zuul/+/941629
-@gerrit:opendev.org- Dong Zhang proposed:09:38
- [zuul/zuul] 940872: Implement keystore functions for OIDC RS256 https://review.opendev.org/c/zuul/zuul/+/940872
- [zuul/zuul] 940971: Manage OIDC signing key rotation https://review.opendev.org/c/zuul/zuul/+/940971
- [zuul/zuul] 941235: Implement command for deleting OIDC signing keys https://review.opendev.org/c/zuul/zuul/+/941235
-@gerrit:opendev.org- Dong Zhang proposed: [zuul/zuul] 941235: Implement command for deleting OIDC signing keys https://review.opendev.org/c/zuul/zuul/+/94123509:43
-@gerrit:opendev.org- Dong Zhang proposed: [zuul/zuul] 941629: Use ZuulTreeCache for OIDC signing keys https://review.opendev.org/c/zuul/zuul/+/94162909:45
-@gerrit:opendev.org- Dong Zhang proposed: [zuul/zuul] 941235: Implement command for deleting OIDC signing keys https://review.opendev.org/c/zuul/zuul/+/94123509:49
@joao15130:matrix.orgHere's an example:12:58
```
@joao15130:matrix.orgHere's an example:12:59
```
root@4ee2d4dc4696:/# nodepool list
+------------+----------+----------------+--------------------------------------+----------------+------+-------+-------------+----------+
| ID | Provider | Label | Server ID | Public IPv4 | IPv6 | State | Age | Locked |
+------------+----------+----------------+--------------------------------------+----------------+------+-------+-------------+----------+
| 0000000009 | ci-cloud | nodepool-jammy | 1c0420cf-08f3-4900-a23e-c436eb8cf150 | 10.225.110.105 | | ready | 00:00:03:39 | unlocked |
+------------+----------+----------------+--------------------------------------+----------------+------+-------+-------------+----------+
```
But in nova:
```
stack@ci-cloud-provider:~$ openstack server list --all
```
-@gerrit:opendev.org- Matthieu Huin https://matrix.to/#/@mhuin:matrix.org proposed: [zuul/zuul] 938463: sqlreporter: ensure build end data is stored even if log URL is too long https://review.opendev.org/c/zuul/zuul/+/93846313:27
-@gerrit:opendev.org- Matthieu Huin https://matrix.to/#/@mhuin:matrix.org proposed on behalf of Tristan Cacqueray https://matrix.to/#/@tristanc_:matrix.org: [zuul/zuul-jobs] 927582: Update the set-zuul-log-path-fact scheme to prevent huge url https://review.opendev.org/c/zuul/zuul-jobs/+/92758213:28
-@gerrit:opendev.org- Matthieu Huin https://matrix.to/#/@mhuin:matrix.org proposed: [zuul/zuul] 938128: autohold REST API: add ref filter validation https://review.opendev.org/c/zuul/zuul/+/93812813:29
-@gerrit:opendev.org- Matthieu Huin https://matrix.to/#/@mhuin:matrix.org proposed on behalf of Tristan Cacqueray https://matrix.to/#/@tristanc_:matrix.org: [zuul/zuul-jobs] 927600: Fix the upload-logs-s3 test playbook https://review.opendev.org/c/zuul/zuul-jobs/+/92760013:54
@fungicide:matrix.orgjoao15130: nodepool tracks the image and node states in zookeeper, so it's possible for things to get deleted directly in the provider without nodepool being certain they're really gone. i would start with trying to debug on the cloud side what's causing them to disappear15:08
@joao15130:matrix.orgfungi: The only thing I see from the cloud side is some DELETE requests that goes to nova:15:16
```
ci-cloud-provider devstack@n-api.service[1842]: INFO nova.api.openstack.requestlog [None req-b66e7d61-1147-48da-926d-98512234d3cb nodepool nodepool] 10.228.237.17 "DELETE /compute/v2.1/servers/0a7b9cd9-8829-4ab1-a5e1-59303a662db7" status: 204 len: 0 microversion: 2.96 time: 0.204060Feb 14 13:08
ci-cloud-provider devstack@n-api.service[1840]: INFO nova.api.openstack.requestlog [None req-0c20666d-5a27-44f5-9c82-7a4725e745ce nodepool nodepool] 10.228.237.17 "DELETE /compute/v2.1/servers/4adbf124-1aa6-45bf-886f-44b3eefc3ca0" status: 204 len: 0 microversion: 2.96 time: 0.207188Feb 14 13:08
```
@joao15130:matrix.orgweird thing is that it seems we got rid of the problem by renaming the cloud provider15:40
-@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed:16:03
- [zuul/zuul] 941620: Use change networks in gerrit driver queries https://review.opendev.org/c/zuul/zuul/+/941620
- [zuul/zuul] 941683: Change network future collision resolution https://review.opendev.org/c/zuul/zuul/+/941683
- [zuul/zuul] 941684: Copy query cache from winning network https://review.opendev.org/c/zuul/zuul/+/941684
@fungicide:matrix.orgjoao15130: well, that'll leave a ton of cruft behind. the safe way to rename a provider in nodepool is to evacuate it of resources (set max-servers to 0, clear out the list of images to be uploaded by nodepool, wait for everything to get cleanly deleted normally in the provider, then remove the provider definition in nodepool config and add your new one)16:10
@joao15130:matrix.orgIndeed, thanks for the procedure, but we just tried this as a workaround. Surprisingly, no instances are deleted now...16:14
@fungicide:matrix.orgyeah, i suspect you have a bunch of orphaned data in zookeeper instead, but anyway when you see the problem crop up again check on the cloud side to figure out why things disappeared out from under nodepool or errored when initially asked by nodepool to delete them16:25
@joao15130:matrix.orgfungi: We did some investigations but never found anything useful except the log from the API. No other errors.16:28
@mnaser:matrix.orgIs there a decent workaround for a scenario when I am using a third-party role that I want to use a secret against.  In my case it's connecting to Tailscale, so if I do:18:58
```
vars:
go_version: 1.23.3
tailscale_authkey: "{{ tailscale_config_authkey.secret }}"
tailscale_tags: tag:tailscale-config
tailscale_oauth_preauthorized: true
secrets:
- name: tailscale_config_authkey
secret: zuul-config-internal-tailscale-config
```
That doesn't work, because the secret has !!unsafe, so Zuul is protecting me here, alternatively, I tried:
```
vars:
go_version: 1.23.3
tailscale_authkey: "{{ hostvars['localhost'].tailscale_config_authkey.secret }}"
tailscale_tags: tag:tailscale-config
tailscale_oauth_preauthorized: true
secrets:
- name: tailscale_config_authkey
secret: zuul-config-internal-tailscale-config
```
That didn't do the trick either, it still shows as undefined.. it would be nice to be able to use this pre-existing role, so how would I tackle this sort of issue? I can't define a secret where data is just a plain string.
@mnaser:matrix.orgFWIW, the role is github.com/artis3n/ansible-role-tailscale but I don't think it's relevant in this case, this is just the case when you need to use a flat variable18:59
@mnaser:matrix.orgI'm trying an actual set_fact in the pre-tasks .. maybe that might do it19:06
@mnaser:matrix.orgOh, that seems to have done it.. I hope these channels are logged so someone someday ctrl+f and finds this19:07
@fungicide:matrix.orglooking at opendev's base job for an example, we seem to supply log upload credentials as secrets and those are plumbed through a post-run playbook into an included role from zuul/zuul-jobs which uses them19:10
@mnaser:matrix.orghttps://opendev.org/zuul/zuul-jobs/src/branch/master/roles/add-build-sshkey/README.rst22:44
Did I miss something or all those overrides are basically not usable because they're in vars/ ?
@clarkb:matrix.orgthere are 7 types of variable that have higher precedence than role vars according to https://docs.ansible.com/ansible/latest/playbook_guide/playbooks_variables.html#understanding-variable-precedence so I guess it depends on how you want to use it22:51
@clarkb:matrix.orgI'm not sure that was an intentional choice though as this was likely one of the very first roles ever written to bootstrap zuul with ansible22:51
@clarkb:matrix.orgits possible that defaults is the better choice it just wasn't known at the time22:51
@clarkb:matrix.orgya the code originates from https://opendev.org/openstack/openstack-zuul-roles/commit/d002b51c1735341c2042040238208aa2f978dd8a22:53
@clarkb:matrix.orglooking at that i highly doubt it was ever intentional to not use defaults22:54
-@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed:23:23
- [zuul/zuul] 941620: Use change networks in gerrit driver queries https://review.opendev.org/c/zuul/zuul/+/941620
- [zuul/zuul] 941683: Change network future collision resolution https://review.opendev.org/c/zuul/zuul/+/941683
- [zuul/zuul] 941684: Copy query cache from winning network https://review.opendev.org/c/zuul/zuul/+/941684
-@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed on behalf of Simon Westphahl: [zuul/zuul] 941435: Make Gerrit event pre-processor multi-threaded https://review.opendev.org/c/zuul/zuul/+/94143523:23
@mnaser:matrix.orgCool. I’ll try and work on clearing that up23:29
-@gerrit:opendev.org- James E. Blair https://matrix.to/#/@jim:acmegating.com proposed: [zuul/zuul] 920212: Use latest patchset with Gerrit cherry-pick https://review.opendev.org/c/zuul/zuul/+/92021223:33
« Thursday, 2025-02-13 Index Saturday, 2025-02-15 »

Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!