UPSTREAM: <carry>: openshift: return no nodegroup when scaling size is zero #42
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
openshift-ci-robot
added
the
size/S
ingvagabund
reviewed
Feb 21, 2019
| @@ -415,6 +421,12 @@ func (c *machineController) nodeGroupForNode(node *apiv1.Node) (cloudprovider.No | |||
| return nil, fmt.Errorf("failed to build nodegroup for node %q: %v", node.Name, err) | |||
| } | |||
|
|
|||
| // We don't scale from 0 so nodes must belong to a nodegroup | |||
| // that has a scale size of at least 1. | |||
| if nodegroup.MaxSize()-nodegroup.MinSize() < 1 { | |||
There was a problem hiding this comment.
Can you also add a testcase where you set both sizes to be equal and verify you get nil node group?
There was a problem hiding this comment.
Test cases now covered in TestControllerNodeGroupForNodeLookup() - 56bc385
|
/test e2e-aws-operator |
frobware
force-pushed
the
fix-bz-1656334
branch
from
dd031b3 to
eec7f96
Compare
openshift-ci-robot
added
size/L
frobware
force-pushed
the
fix-bz-1656334
branch
4 times, most recently
from
ce7f209 to
084f847
Compare
These will be used/shared by the controller tests.
frobware
force-pushed
the
fix-bz-1656334
branch
from
084f847 to
efc492b
Compare
|
/lgtm |
|
/hold |
openshift-ci-robot
added
the
do-not-merge/hold
frobware
force-pushed
the
fix-bz-1656334
branch
from
efc492b to
90f408b
Compare
|
/hold cancel |
openshift-ci-robot
removed
the
do-not-merge/hold
frobware
force-pushed
the
fix-bz-1656334
branch
from
90f408b to
f313586
Compare
frobware
changed the title
|
/hold |
openshift-ci-robot
added
the
do-not-merge/hold
|
/hold cancel |
openshift-ci-robot
removed
the
do-not-merge/hold
… == 0 Change nodeGroupForNode() to return nil when the underlying node group has no scaling bounds. The underlying MachineSet in these cases is either missing the following annotations: ``` annotations: machine.openshift.io/cluster-api-autoscaler-node-group-min-size: "1" machine.openshift.io/cluster-api-autoscaler-node-group-max-size: "6" ``` or they are present but the result of `max-size - min-size` is zero. This change helps to remove a lot of spurious error messages from the core of the autoscaler logs at runtime. Without this change we see the following printed every time the autoscaler scans the state of the cluster (which defaults to 10s) and can be very misleading. ```console $ oc logs -f cluster-autoscaler-default-77f884d74b-527bq E1204 08:26:01.996252 1 utils.go:467] Error while checking node group size worker-us-east-2a: group size not found E1204 08:26:01.996545 1 utils.go:467] Error while checking node group size worker-us-east-2b: group size not found E1204 08:26:01.996678 1 utils.go:467] Error while checking node group size worker-us-east-2c: group size not found ``` Those unit tests that were previously relying on nodeGroupForNode() to always return a node group when no bounds had been specified have been modified so that they now have a scaling bound >= 1. Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1656334
These are now covered by TestControllerNodeGroupForNodeLookup
frobware
force-pushed
the
fix-bz-1656334
branch
from
f313586 to
d64b32a
Compare
|
/lgtm Thanks for the improvements |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Change nodeGroupForNode() to return nil when the underlying node group
has no scaling bounds. The underlying MachineSet in these cases is
either missing the following annotations:
or they are present but the result of
max-size - min-sizeis zero.This change helps to remove a lot of spurious error messages from the
core of the autoscaler logs at runtime. Without this change we see the
following printed every time the autoscaler scans the state of the
cluster (which defaults to 10s) and can be very misleading.
Those unit tests that were previously relying on nodeGroupForNode() to
always return a node group when no bounds had been specified have been
modified so that they now have a scaling bound >= 1.
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1656334