Bridge Update Papercuts 10/02/23

[guineveresaenger]

During ops, we ran a bridge release and updated the bridge to all providers. Some updates were successful; others failed.

These are sorted by category and linked to the successful PR or failed Workflows.

Author Identity Unknown

This is a very frustrating occurrence where git cannot determine the identity of the runner in question.
Here's the issue for that: pulumi/upgrade-provider#144

• aws - Author identity unknown: https://github.com/pulumi/pulumi-aws/actions/runs/6387505142/job/17335943737#step:3:1487
• gcp - Author identity unknown: https://github.com/pulumi/pulumi-gcp/actions/runs/6387531061/job/17336012024#step:3:1240
• azuread - Author identity unknown: https://github.com/pulumi/pulumi-azuread/actions/runs/6387507006/job/17335949495#step:3:993
• artifactory - Author identity unknown: https://github.com/pulumi/pulumi-artifactory/actions/runs/6387502698/job/17335937010#step:3:905
• akamai - Author identity unknown: https://github.com/pulumi/pulumi-akamai/actions/runs/6387499107/job/17335926372#step:3:860
• confluentcloud - Author identity unknown: https://github.com/pulumi/pulumi-confluentcloud/actions/runs/6387514022/job/17335967429#step:3:859
• consul - Author identity unknown: https://github.com/pulumi/pulumi-consul/actions/runs/6387515397/job/17335971334#step:3:856
• databricks - Author identity unknown: https://github.com/pulumi/pulumi-databricks/actions/runs/6387517179/job/17335975505#step:3:856
• digitalocean - Author identity unknown: https://github.com/pulumi/pulumi-digitalocean/actions/runs/6387519955/job/17335982486#step:3:939
• f5bigip - Author identity unknown: https://github.com/pulumi/pulumi-f5bigip/actions/runs/6387528602/job/17336005355#step:3:880
• gitlab - Author identity unknown: https://github.com/pulumi/pulumi-gitlab/actions/runs/6387533580/job/17336018063#step:3:950
• hcloud Author identity unknown: https://github.com/pulumi/pulumi-hcloud/actions/runs/6387534715/job/17336020835#step:3:873
• local Author identity unknown: https://github.com/pulumi/pulumi-local/actions/runs/6387544229/job/17336043004#step:3:684
• pagerduty Author identity unknown: https://github.com/pulumi/pulumi-pagerduty/actions/runs/6387563739/job/17336090973#step:3:858
• random - Author identity unknown: https://github.com/pulumi/pulumi-random/actions/runs/6387571150/job/17336109875#step:3:970
• splunk - Author identity unknown: https://github.com/pulumi/pulumi-splunk/actions/runs/6387581463/job/17336136599#step:3:818
• mysql - Author identity unknown: https://github.com/pulumi/pulumi-mysql/actions/runs/6387549497/job/17336055127#step:3:792

No space left on device/ Runner received shutdown signal

Unsure if these are the same cause, but they're in a similar space.

• azure - No space left on device (github runner) https://github.com/pulumi/pulumi-azure/actions/runs/6387505876
• datadog - Runner received shutdown signal https://github.com/pulumi/pulumi-datadog/actions/runs/6387518686
• oci Runner shut down https://github.com/pulumi/pulumi-oci/actions/runs/6387555747/job/17336069503

make upstream failure

Filed in pulumi/upgrade-provider#172

• auth0 - make upstream failure: https://github.com/pulumi/pulumi-auth0/actions/runs/6387504018/job/17335940476
• external - make upstream failure: https://github.com/pulumi/pulumi-external/actions/runs/6387527214/job/17336002314#step:3:1178

Dirty Worktree

Not all of these have a similar diff!

• ec - PR failing with dirty worktree: https://github.com/pulumi/pulumi-ec/actions/runs/6387566744/job/17336172052?pr=154
• keycloak Dirty worktree in PR https://github.com/pulumi/pulumi-keycloak/actions/runs/6387584570/job/17336224131?pr=276
• kong Dirty worktree in PR https://github.com/pulumi/pulumi-kong/actions/runs/6387587213/job/17336223423?pr=213
• newrelic - dirty worktree on PR: https://github.com/pulumi/pulumi-newrelic/actions/runs/6387602019/job/17336250903?pr=514
• rabbitmq dirty worktree on PR: https://github.com/pulumi/pulumi-rabbitmq/actions/runs/6387604327/job/17336290791?pr=190
• sumologic - dirty worktree with a dotnet logo diff??? https://github.com/pulumi/pulumi-sumologic/actions/runs/6387646681/job/17336401366?pr=169#step:19:49
• vsphere - dirty worktree failure, seems related to wheels https://github.com/pulumi/pulumi-vsphere/actions/runs/6387634885/job/17336384057?pr=328
• cloudflare Dirty Worktree on PR: https://github.com/pulumi/pulumi-cloudflare/actions/runs/6387559200/job/17336159829?pr=503

Grab Bag

These all warrant more investigation on the provider

• azuredevops - this provider did not receive an upgrade-bridge Workflow at all
• docker - Failing lint and failing tests on PR: https://github.com/pulumi/pulumi-docker/actions/runs/6387560829/job/17336274461?pr=773
• venafi - PR open, with unrelated longstanding test failures: https://github.com/pulumi/pulumi-venafi/actions/runs/6387627243/job/17336377934?pr=181

Merged After Rerun due to flaky python deps install

Issue for this in #650

• mongodbatlas - python deps install failure on PR, rerun: https://github.com/pulumi/pulumi-mongodbatlas/actions/runs/6387600977?pr=290 - UPDATE: merged
• openstack - flaky Python deps, trigger rerun: https://github.com/pulumi/pulumi-openstack/actions/runs/6387606471?pr=398 UPDATE merged
• tailscale failure to install python deps on PR tests; rerun https://github.com/pulumi/pulumi-tailscale/actions/runs/6387645772?pr=226 UPDATE: PR needed rebase
• signalfx - failure to install Python deps on PR; triggered a rerun: https://github.com/pulumi/pulumi-signalfx/actions/runs/6387610874?pr=309 UPDATE merged
• slack - flaky python deps install on PR tests; rerun https://github.com/pulumi/pulumi-slack/actions/runs/6387634352?pr=145 UPDATE merged
• snowflake failure to install Python deps on PR; rerunning: https://github.com/pulumi/pulumi-snowflake/actions/runs/6387634581?pr=343 UPDATE: merged

Merged After A Third Rerun due to failure on setup script

• rancher2 - Python deps install failure on PR tests; rerun https://github.com/pulumi/pulumi-rancher2/actions/runs/6387641696?pr=272 UPDATE: Failure on setup script https://github.com/pulumi/pulumi-rancher2/actions/runs/6387641696/job/17370496012?pr=272; rerun: https://github.com/pulumi/pulumi-rancher2/actions/runs/6387641696?pr=272 UPDATE merged after two reruns

Merged After Rerun due to a 502 on install-pulumi-cli

Issue for this in #652

• libvirt - on PR, pulumi/action-install-pulumi-cli@v2 had a bad gateway? Re-running tests: https://github.com/pulumi/pulumi-libvirt/actions/runs/6387578369?pr=142 - UPDATE: merged
• wavefront - 502 error on installing pulumi/action-install-pulumi-cli@v2 in PR tests; rerunning jobs https://github.com/pulumi/pulumi-wavefront/actions/runs/6387633615?pr=220 UPDATE: merged

Merged

I believe a human manually merged all these.

• aiven - merged
• alicloud merged
• archive - merged
• civo merged
• cloudamqp merged
• cloudinit - merged Upgrade pulumi-terraform-bridge to v3.61.0 pulumi-cloudinit#130
• dnsimple - merged 😀
• equinix-metal merged
• fastly merged
• github merged
• http merged
• kafka merged
• linode merged
• mailgun merged
• minio merged
• nomad merged
• ns1 merged
• null merged
• okta merged 😀
• onelogin merged
• opsgenie merged
• postgresql merged
• rke merged
• spotinst merged
• tls merged

Summary

• 25 merged, manually, with no issue

• 9 merged after re-kicking flaky tests

• 9 dirty work trees

• 2 failure to find upstream go.mod

• 3 runners struggling with the workload

• 17 failed with git not accepting the identity of the runner attempting to open a PR

• 3 triggered genuine breaking tests

[t0yv0]

A lot of these fail because workflows are not merging. You wouldn't get newrelic to fail if pulumi/pulumi-newrelic#517 merged for example, but it's not auto merging. I think @mjeffryes was looking into making sure all workflows auto merge.