[KeyVault Keys and Secrets Hotfix 4.0.3] Commits to make sure the builds work #8782
Conversation
sadasant
force-pushed
the
hotfix/keyvault-challengeAuth-keys-secrets-4.0.3-CI-fix
branch
from
cdb3f32 to
32a0697
Compare
sadasant
force-pushed
the
hotfix/keyvault-challengeAuth-keys-secrets-4.0.3-CI-fix
branch
from
a41ced4 to
b92e102
Compare
sadasant
requested review from
bterlson,
danieljurek,
daviwil,
mikeharder,
mitchdenny,
sophiajt and
weshaggard
as code owners
There was a problem hiding this comment.
What is the reason for this casting? I understand that jws.decode() takes a string, but why is it that we get an error now, but not when the last stable version of keyvault keys was released?
There was a problem hiding this comment.
I don't know. I know that as we go back in time, stability on the commits to master is lost, due to how the environment changes. By environment I mean all of the projects we build, but even more how the CI pipelines have changed. Only recently we've been enforcing stuff more consistently across the whole project.
During last week's hotfix release for identity, there were similar issues. Not similar to this specific change, but core-lro reported to be broken on a commit that should have been green, since the commit was used to release another package. When I investigated the issue, I was able to confirm that many commits in the past, despite having a relationship with releases, either appeared to be broken on the github reports, or were broken locally when I tried to build them. I also found out that the CI logs are lost after some time in the past. I reported all of this to @KarishmaGhiya
There's also the fact that this change does absolutely no damage to the keyvault hotfix release.
There was a problem hiding this comment.
Just to avoid loose threads: The way core-lro was fixed for the identity release was to move it forward to the next stable release, only in that folder, while downgrading a couple of dependencies. Fortunately, identity didn't depend on core-lro, so that was fine.
There was a problem hiding this comment.
Ok, but as string should work just fine right? Why do we need to cast to any and then cast to string?
There was a problem hiding this comment.
Just as string was enough indeed! Nice catch!
sadasant
requested review from
HarshaNalluru,
XiaoningLiu,
chradek,
jeremymeng,
jiacfan,
richardpark-msft,
southpolesteve,
vinjiang and
xirzec
as code owners
sadasant
force-pushed
the
hotfix/keyvault-challengeAuth-keys-secrets-4.0.3-CI-fix
branch
from
cabdfa2 to
1c9e889
Compare
ramya-rao-a
removed request for
HarshaNalluru,
XiaoningLiu,
bterlson,
chradek,
danieljurek,
daviwil,
jeremymeng,
jiacfan,
richardpark-msft,
southpolesteve,
vinjiang and
xirzec
|
@KarishmaGhiya Can you review the pipeline related changes in this PR? |
sdk/identity/identity/test/node/managedIdentityCredential.spec.ts
Outdated
Show resolved
Hide resolved
|
Thanks @KarishmaGhiya @sadasant Please merge this PR and update the 3 related PRs |
|
I can merge this PR. |
sadasant
merged commit f08ac0f
into
Azure:hotfix/keyvault-challengeAuth-keys-secrets-4.0.3
sadasant
deleted the
hotfix/keyvault-challengeAuth-keys-secrets-4.0.3-CI-fix
branch
5 participants
This PR has what is needed to make the base branch for the challenge based authentication hotfix stable.
Related to: