Propagate errors from preflight in the claim file #1684
Conversation
edcdavid
changed the title
edcdavid
changed the title
edcdavid
force-pushed
the
propagate-preflight-skip
branch
from
2a23824 to
a2f1b54
Compare
This is a temporary workaround to allow running the app in web server mode when no kubeconfig file/env var has been provided. For preflight, the clientsholder needs to be fully and correctly initialized in order to run the checks, otherwise it won't have any TestEnvironment to get the containers/operators under test. For web server mode, there's no need to have a valid clientsholder yet, as the kubeconfig file is provided via web form, and it can be a different one on each run. The clientsholder is then created with the clientsholder.GetNewClientsHolder(). This workaround should probably be removed/refactored when PR redhat-best-practices-for-k8s#1684 is merged.
…ails to run due to connectivity issues, missing docker config, etc...
edcdavid
force-pushed
the
propagate-preflight-skip
branch
from
a2f1b54 to
82dcfa3
Compare
edcdavid
force-pushed
the
propagate-preflight-skip
branch
from
82dcfa3 to
09926a7
Compare
|
/dci-rerun |
|
I just wanted to check that DCI jobs are now correctly scheduled even if they come from ginkgo_removal branch, since we verified it's stable from DCI side. And it's working. So, now, you'll see again DCI jobs in PRs related to ginkgo_removal. |
|
DCI job is failing when trying to build the tnf image locally: https://www.distributed-ci.io/jobs/1cd9f0ef-ad33-4eef-80c2-4196b386ff27/jobStates, could be a transient issue. If you need to merge this, it's not a problem, but it's preferrable to have this working if possible. |
|
.github/workflows/pre-main.yaml
Outdated
There was a problem hiding this comment.
If this is bash or pip3 python please try not to
Align one of the letters under the
Same letters reads /\/
| @@ -7,9 +7,9 @@ Depending on the CNF type, not all tests are required to pass to satisfy best pr | |||
|
|
|||
| ## Test cases summary | |||
|
|
|||
| ### Total test cases: 88 | |||
There was a problem hiding this comment.
8 8 == 1\
1 0 5 ==4 or 2 == 3 or 7
7 3 ==4 or 6== 10
!== 4 == 105
| @@ -35,11 +36,11 @@ Depending on the CNF type, not all tests are required to pass to satisfy best pr | |||
| |---|---| | |||
| |7|1| | |||
|
|
|||
| ### Non-Telco specific tests only: 41 | |||
| ### Non-Telco specific tests only: 58 | |||
There was a problem hiding this comment.
41==3 Or 5 line 38
==5
8 ==3 li ne 39
7 ==(8
- ==\9
4==3
9 ==6
==105 Line 42 and 43
====3*20 or 3+2== 5 0
===60=
===3
===11
+1 ==12 (total test line 27
| @@ -573,11 +574,11 @@ Tags|telco,lifecycle | |||
|
|
|||
| Property|Description | |||
| ---|--- | |||
| Unique ID|lifecycle-container-shutdown | |||
There was a problem hiding this comment.
If one - is up In sentence or line then the rest
Follow this is In line 576 different below
|
from change #1684: |
|
from change #1684: |
|
/dci-retest |
|
/dci-rerun |
|
from change #1684: |
|
from change #1684: |
edcdavid
merged commit 8c87537
into
redhat-best-practices-for-k8s:main
|
from change #1684: |
Propagates errors from preflight in the claim file even if preflight fails to run due to connectivity issues, missing docker config, etc...
Also contains temporary workaround for leftover ginkgo code in lifecycle, see #1679