Remove kube-burner download and re-try download tasks

[mcornea]

In CI we hit an issue while downloading kube-burner release. This PR is removing the kube-burner release download since this is covered outside jetlag and adds retries to the tasks which download pakages.

TASK [bastion-install : Get CoreDNS, OpenShift clients, opm, kube-burner, grpcurl, minio client, and yq] ***
Tuesday 24 February 2026 08:21:03 +0000 (0:00:00.033) 0:00:20.133 ****** 
ok: [XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX] => (item={'url': 'https://github.com/coredns/coredns/releases/download/v1.10.1/coredns_1.10.1_linux_amd64.tgz', 'dest': '/root/mno/coredns_linux_amd64.tgz'})
failed: [XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX] (item={'url': 'https://github.com/kube-burner/kube-burner/releases/download/v1.15.0/kube-burner-V1.15.0-linux-x86_64.tar.gz', 'dest': '/root/mno/kube-burner-linux.tar.gz'}) => {"ansible_loop_var": "item", "changed": false, "dest": "/root/mno/kube-burner-linux.tar.gz", "elapsed": 0, "gid": 0, "group": "root", "item": {"dest": "/root/mno/kube-burner-linux.tar.gz", "url": "https://github.com/kube-burner/kube-burner/releases/download/v1.15.0/kube-burner-V1.15.0-linux-x86_64.tar.gz"}, "mode": "0644", "msg": "Request failed", "owner": "root", "response": "HTTP Error 502: Bad Gateway or Proxy Error", "secontext": "system_u:object_r:admin_home_t:s0", "size": 36486634, "state": "file", "status_code": 502, "uid": 0, "url": "https://github.com/kube-burner/kube-burner/releases/download/v1.15.0/kube-burner-V1.15.0-linux-x86_64.tar.gz"}
ok: [XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX] => (item={'url': 'https://github.com/fullstorydev/grpcurl/releases/download/v1.8.1/grpcurl_1.8.1_linux_x86_64.tar.gz', 'dest': '/root/mno/grpcurl-linux.tar.gz'})
ok: [XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX] => (item={'url': 'https://dl.min.io/client/mc/release/linux-amd64/mc', 'dest': '/usr/local/bin/mc', 'mode': '0700'})
ok: [XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX] => (item={'url': 'https://github.com/mikefarah/yq/releases/download/v4.40.3/yq_linux_amd64.tar.gz', 'dest': '/root/mno/yq_linux_amd64.tar.gz'})

[josecastillolema]

+1

[rsevilla87]

I wonder if we need other binaries downloaded by jetlag apart from kb

[akrzos]

In CI we hit an issue while downloading kube-burner release. This PR is removing the kube-burner release download since this is covered outside jetlag and adds retries to the tasks which download pakages.

TASK [bastion-install : Get CoreDNS, OpenShift clients, opm, kube-burner, grpcurl, minio client, and yq] ***
Tuesday 24 February 2026 08:21:03 +0000 (0:00:00.033) 0:00:20.133 ****** 
ok: [XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX] => (item={'url': 'https://github.com/coredns/coredns/releases/download/v1.10.1/coredns_1.10.1_linux_amd64.tgz', 'dest': '/root/mno/coredns_linux_amd64.tgz'})
failed: [XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX] (item={'url': 'https://github.com/kube-burner/kube-burner/releases/download/v1.15.0/kube-burner-V1.15.0-linux-x86_64.tar.gz', 'dest': '/root/mno/kube-burner-linux.tar.gz'}) => {"ansible_loop_var": "item", "changed": false, "dest": "/root/mno/kube-burner-linux.tar.gz", "elapsed": 0, "gid": 0, "group": "root", "item": {"dest": "/root/mno/kube-burner-linux.tar.gz", "url": "https://github.com/kube-burner/kube-burner/releases/download/v1.15.0/kube-burner-V1.15.0-linux-x86_64.tar.gz"}, "mode": "0644", "msg": "Request failed", "owner": "root", "response": "HTTP Error 502: Bad Gateway or Proxy Error", "secontext": "system_u:object_r:admin_home_t:s0", "size": 36486634, "state": "file", "status_code": 502, "uid": 0, "url": "https://github.com/kube-burner/kube-burner/releases/download/v1.15.0/kube-burner-V1.15.0-linux-x86_64.tar.gz"}
ok: [XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX] => (item={'url': 'https://github.com/fullstorydev/grpcurl/releases/download/v1.8.1/grpcurl_1.8.1_linux_x86_64.tar.gz', 'dest': '/root/mno/grpcurl-linux.tar.gz'})
ok: [XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX] => (item={'url': 'https://dl.min.io/client/mc/release/linux-amd64/mc', 'dest': '/usr/local/bin/mc', 'mode': '0700'})
ok: [XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX] => (item={'url': 'https://github.com/mikefarah/yq/releases/download/v4.40.3/yq_linux_amd64.tar.gz', 'dest': '/root/mno/yq_linux_amd64.tar.gz'})

I would be curious to know if the other place that kube-burner is downloaded from is more robust or just the preferred place/method to download kube-burner. Perhaps it allows different versions and that make it more robust?

We could remove it and a few other tools that are downloaded for convenience or put it behind a var such that CI doesn't repeat actions but allows other users of the tooling in the lab the same convenience as always.

[mcornea]

I would be curious to know if the other place that kube-burner is downloaded from is more robust or just the preferred place/method to download kube-burner. Perhaps it allows different versions and that make it more robust?

It looks like it uses the same github release downloads so I'm guessing this was a temporary github issue.

We could remove it and a few other tools that are downloaded for convenience or put it behind a var such that CI doesn't repeat actions but allows other users of the tooling in the lab the same convenience as always.

We could also remove the force: true from get_url tasks to make the tasks idempotent so they don't download the binaries with every run, wdyt?

[akrzos]

I would be curious to know if the other place that kube-burner is downloaded from is more robust or just the preferred place/method to download kube-burner. Perhaps it allows different versions and that make it more robust?

It looks like it uses the same github release downloads so I'm guessing this was a temporary github issue.

We could remove it and a few other tools that are downloaded for convenience or put it behind a var such that CI doesn't repeat actions but allows other users of the tooling in the lab the same convenience as always.

We could also remove the force: true from get_url tasks to make the tasks idempotent so they don't download the binaries with every run, wdyt?

I looked back in git blame and I believe the force: true was originally for oc cli client download which is no longer even handled by this task so we should be fine to remove the force to prevent additional download/retires on tools that have already been downloaded. (If your curious as to why there was a force, it was because we always downloaded the latest client and wanted every run to include the latest version of the client)

[mcornea]

I looked back in git blame and I believe the force: true was originally for oc cli client download which is no longer even handled by this task so we should be fine to remove the force to prevent additional download/retires on tools that have already been downloaded. (If your curious as to why there was a force, it was because we always downloaded the latest client and wanted every run to include the latest version of the client)

Cool, I removed the force from the task and added retries to cover intermittent issues.

[akrzos]

/test ?

[akrzos]

/test deploy-mno

[openshift-ci]

[APPROVALNOTIFIER] This PR is APPROVED

This pull-request has been approved by: akrzos

The full list of commands accepted by this bot can be found here.

The pull request process is described here

Details

Needs approval from an approver in each of these files:

• OWNERS [akrzos]

Approvers can indicate their approval by writing /approve in a comment
Approvers can cancel approval by writing /approve cancel in a comment