do not include customData in AzureMachinePool hash calculation #1197
Conversation
k8s-ci-robot
added
kind/bug
k8s-ci-robot
added
area/provider/azure
devigned
force-pushed
the
customdata-hash
branch
from
8f9ad57
to
7634b05
Compare
k8s-ci-robot
added
size/XS
|
Pass 1, retry |
|
Pass1 was successful Pass 2, retry |
|
Pass 2 was successful Pass 3, retry |
|
Pass 3 was successful Pass 4, retry |
|
Pass 4 was successful Pass 5, retry |
|
Pass 5 was partially successful with Windows failing with an unrelated error: Pass 6, retry |
devigned
force-pushed
the
customdata-hash
branch
from
7634b05
to
db46bad
Compare
|
Pass 6 was partially successful with Windows with VMSS error. Rebased on master to get the latest from v1alpha4 and hopefully cluster data from test clusters that failed to come up. |
|
Pass 7 was successful Pass 8, retry |
|
Pass 8 was successful Pass 9, retry |
|
Pass 9 failed with e[1]: *** [Makefile:249: /home/prow/go/src/sigs.k8s.io/cluster-api-provider-azure/hack/tools/bin/kubectl-v1.19.2] Error 56 again. Looking that looks like it is related to curl having a network issue. created #1202 Pass 10, retry |
|
/retest |
|
Pass 11, retry |
|
Pass 11 success Pass 12, retry |
|
Pass 12 success Pass 13, retry |
|
Pass 18 success. Pass 19, retry |
|
Pass 19 Fail; IPv6 Pass 20, retry |
|
Pass 20 success Pass 21, retry |
|
Pass 21 success Pass 22, retry |
devigned
force-pushed
the
customdata-hash
branch
from
11b61d4
to
9210284
Compare
|
Pass 22 success @nader-ziada, @CecileRobertMichon I think this is looking pretty solid. wdyt? |
|
looks good to go! passed most of the times /lgtm |
k8s-ci-robot
added
the
lgtm
devigned
force-pushed
the
customdata-hash
branch
from
9210284
to
3aeef19
Compare
k8s-ci-robot
removed
the
lgtm
config/manager/manager.yaml
Outdated
| @@ -24,6 +24,7 @@ spec: | |||
| - --leader-elect | |||
| - "--metrics-bind-addr=127.0.0.1:8080" | |||
| - "--feature-gates=MachinePool=${EXP_MACHINE_POOL:=false},AKS=${EXP_AKS:=false}" | |||
| - "--v=2" | |||
There was a problem hiding this comment.
was this checked in by mistake?
There was a problem hiding this comment.
I increased verbosity of the test logging. To make it easier, I add an explicit verbosity to the controller.
There was a problem hiding this comment.
isn't 1 the default though? if so this increases verbosity for all default clusterctl deployments
There was a problem hiding this comment.
Looks like the default is 0. I did not intend on increasing the default. I'll change it.
devigned
force-pushed
the
customdata-hash
branch
from
3aeef19
to
410a36d
Compare
test/e2e/config/azure-dev.yaml
Outdated
| @@ -75,6 +75,9 @@ providers: | |||
| targetName: "cluster-template-windows.yaml" | |||
| - sourcePath: "${PWD}/templates/test/cluster-template-prow-machine-pool-windows.yaml" | |||
| targetName: "cluster-template-machine-pool-windows.yaml" | |||
| replacements: | |||
| - old: "--v=2" | |||
There was a problem hiding this comment.
| - old: "--v=2" | |
| - old: "--v=0" |
There was a problem hiding this comment.
Great catch!! Fixed
devigned
force-pushed
the
customdata-hash
branch
from
410a36d
to
9d2dd13
Compare
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: CecileRobertMichon The full list of commands accepted by this bot can be found here. The pull request process is described here
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
k8s-ci-robot
added
lgtm
What type of PR is this?
/kind bug
What this PR does / why we need it:
AzureMachinePool will occasionally enter into a cycle where the hash tag calculated from the last model update will be different than the hash calculated for a newly reconciled AzureMachinePool since the bootstrap data secret has been updated since the previous reconcile causing the customData of the VMSS to change, thus changing the hash for the entire model.
This PR removes the customData from the hash calculation and increases test controller log verbosity.
Which issue(s) this PR fixes (optional, in
fixes #<issue number>(, fixes #<issue_number>, ...)format, will close the issue(s) when PR gets merged):Related: #1190, #1186, #1182
TODOs:
Release note: