| @bennetefx:matrix.org | Hi there Zuul community, facing an error in my pipeline sometimes nowadays, I am not able to understand why this behaviour is happening. Can you comment on this: | 01:47 |
|---|---|---|
| 2025-02-11 01:35:10.537141 | TASK [Ansible Task Running a Shell Script] | ||
| 2025-02-11 01:35:10.860492 | [buildserver] Waiting on logger | ||
| 2025-02-11 01:35:11.668833 | [buildserver] Waiting on logger | ||
| 2025-02-11 01:35:12.270429 | [buildserver] Waiting on logger | ||
| 2025-02-11 01:35:13.076001 | [buildserver] Waiting on logger | ||
| 2025-02-11 01:35:13.678601 | [buildserver] Waiting on logger | ||
| 2025-02-11 01:35:14.481160 | [buildserver] Waiting on logger | ||
| @bennetefx:matrix.org | * Hi there Zuul community, facing an error in my pipeline sometimes nowadays, I am not able to understand why this behaviour is happening. Can you comment on this: | 01:48 |
| ``` | ||
| 2025-02-11 01:35:10.537141 | TASK \[Ansible Task Running a Shell Script\] | ||
| 2025-02-11 01:35:10.860492 | \[node\] Waiting on logger | ||
| 2025-02-11 01:35:11.668833 | \[node\] Waiting on logger | ||
| 2025-02-11 01:35:12.270429 | \[node\] Waiting on logger | ||
| 2025-02-11 01:35:13.076001 | \[node\] Waiting on logger | ||
| 2025-02-11 01:35:13.678601 | \[node\] Waiting on logger | ||
| 2025-02-11 01:35:14.481160 | \[node\] Waiting on logger | ||
| ``` | ||
| @clarkb:matrix.org | That means the console logging service is either not being started or has died for some reason | 01:53 |
| @clarkb:matrix.org | If you reboot in the job for example you need to restart it. | 01:53 |
| @clarkb:matrix.org | Could also be that your job is blocking network access to it. I believe it uses tcp port 19885 | 01:53 |
| @bennetefx:matrix.org | Thank you very much Clark. | 02:01 |
| @fungicide:matrix.org | does the job eventually complete anyway, or die in a timeout? | 02:02 |
| @bennetefx:matrix.org | It does complete eventually. | 02:03 |
| @fungicide:matrix.org | that rules out the test node crashing or something, at least | 02:54 |
| @bennetefx:matrix.org | Yes | 03:34 |
| -@gerrit:opendev.org- Dong Zhang proposed: [zuul/zuul] 940872: Implement keystore functions for OIDC RS256 https://review.opendev.org/c/zuul/zuul/+/940872 | 08:07 | |
| -@gerrit:opendev.org- Vladimir Kozhukalov proposed: [zuul/zuul-jobs] 941516: [remove-registry-tag] Allow using in a loop https://review.opendev.org/c/zuul/zuul-jobs/+/941516 | 08:07 | |
| @cidlik:matrix.org | Hello there! | 08:30 |
| I know that the Ansible has a strategy ["free"](https://docs.ansible.com/ansible/latest/collections/ansible/builtin/free_strategy.html#free-strategy) which allows to run playbooks on multiple nodes independently. If one node any error will appear, then will proceed execution will be proceeding on others. | ||
| I would like to know, has any opportunity to get the same result at the Zuul job level? A job runs several playbooks. I want to run the job in multiple nodes simultaneously via nodeset, but I know that sometimes it may fail in one node. But I would prefer that the execution would continue on other nodes. | ||
| @cidlik:matrix.org | * Hello there! | 08:31 |
| I know that the Ansible has a strategy ["free"](https://docs.ansible.com/ansible/latest/collections/ansible/builtin/free_strategy.html#free-strategy) which allows to run playbooks on multiple nodes independently. If any error will appear in one node, then will proceed execution will be proceeding on others. | ||
| I would like to know, has any opportunity to get the same result at the Zuul job level? A job runs several playbooks. I want to run the job in multiple nodes simultaneously via nodeset, but I know that sometimes it may fail in one node. But I would prefer that the execution would continue on other nodes. | ||
| @cidlik:matrix.org | * Hello there! | 08:31 |
| I know that the Ansible has a strategy ["free"](https://docs.ansible.com/ansible/latest/collections/ansible/builtin/free_strategy.html#free-strategy) which allows to run playbooks on multiple nodes independently. If any error will appear in one node, then execution will be proceeding on others. | ||
| I would like to know, has any opportunity to get the same result at the Zuul job level? A job runs several playbooks. I want to run the job in multiple nodes simultaneously via nodeset, but I know that sometimes it may fail in one node. But I would prefer that the execution would continue on other nodes. | ||
| -@gerrit:opendev.org- Dong Zhang proposed: [zuul/zuul] 940971: Manage OIDC signing key rotation https://review.opendev.org/c/zuul/zuul/+/940971 | 08:32 | |
| -@gerrit:opendev.org- Dong Zhang proposed: [zuul/zuul] 941235: Implement command for deleting OIDC signing keys https://review.opendev.org/c/zuul/zuul/+/941235 | 08:42 | |
| -@gerrit:opendev.org- Dong Zhang proposed: [zuul/zuul] 941235: Implement command for deleting OIDC signing keys https://review.opendev.org/c/zuul/zuul/+/941235 | 08:56 | |
| -@gerrit:opendev.org- Simon Westphahl proposed: | 10:54 | |
| - [zuul/zuul] 941435: Make Gerrit event pre-processor multi-threaded https://review.opendev.org/c/zuul/zuul/+/941435 | ||
| - [zuul/zuul] 941541: wip: Avoid data raced on Gerrit change update https://review.opendev.org/c/zuul/zuul/+/941541 | ||
| @prime149:matrix.org | Hi team. Can you please review https://review.opendev.org/c/zuul/zuul-jobs/+/941516 ? | 13:27 |
| We recently merged this https://review.opendev.org/c/zuul/zuul-jobs/+/881524 which partially solved the issue, but there is still a problem with the remove_registry_tag_api_url variable. | ||
| @prime149:matrix.org | This change is backward compatible so I assume we don't need 2-weeks announcement before merging it. | 13:28 |
| @joao15130:matrix.org | Hello guys. I'm making good progress on our implementation but a few challenges remain. One of them is due to how nodepool manages resources. | 16:06 |
| For a particular storage system we have in-house, we need to install a debian package (not available in a repo but downloadable from our support website) and create a file with some parameters known in advance. | ||
| Generally these operations are done at the host level but because of nodepool creating images, we need to make sure that the image has already the package installed and the file configured before executing any tempest tests. | ||
| @joao15130:matrix.org | * Hello guys. I'm making good progress on our implementation but a few challenges remain. One of them is due to how nodepool manages resources. | 16:06 |
| For a particular storage system we have in-house, we need to install a debian package (not available in a repo but downloadable from our support website) and create a file with some parameters known in advance. | ||
| Generally these operations are done at the host level but because of nodepool creating images, we need to make sure that the image has already the package installed and the file configured before executing any tempest tests. | ||
| I was thinking of writing a new elements but I was looking for your advices on how to tackle this objective. | ||
| Any help would be highly appreciated. | ||
| @clarkb:matrix.org | Doesn't this reduce to simply running different jobs for the different independent executions? | 16:14 |
| @clarkb:matrix.org | yes you can write an element to achieve this. We do similar to ensure our images cache git repos for example. | 16:15 |
| @joao15130:matrix.org | Do you have any example on how can I upload a package into the image without a repo? | 16:22 |
| @clarkb:matrix.org | Off the top of my head no. But your element could contain a shell script that wget/curl'd the pacakge, verified its integrity with a hash check or similar then dpkg -i the file | 16:33 |
| @clarkb:matrix.org | basically hwoever you would normally install it do that process within the element | 16:33 |
| @fungicide:matrix.org | joao15130: note that you could also choose to retrieve and install the package during the job, for example using a pre-run phase playbook, rather than preinstalling it in the images themselves. that gives you the flexibility to run jobs on the same images with and without that package, or with different versions of the package | 16:41 |
| @joao15130:matrix.org | fungi: Yes that's an another solution and it was my B plan. I have more experience on ansible. But how can I tell the playbook to upload a file which resides actually on my zuul-config repo to the nodepool instance ? | 16:59 |
| -@gerrit:opendev.org- Zuul merged on behalf of Clark Boylan: [zuul/nodepool] 941294: Pin googleapis-common-protos https://review.opendev.org/c/zuul/nodepool/+/941294 | 20:19 | |
| @joao15130:matrix.org | I think I've nailed how to do it, thanks fungi | 20:34 |
Generated by irclog2html.py 2.17.3 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!