Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Bug 1916379: Change metric for errors to gauge #24

Merged
merged 1 commit into from Jan 14, 2021
Merged

Bug 1916379: Change metric for errors to gauge #24

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Jump to
Jump to file
Failed to load files.
Diff view
Diff view
1 change: 1 addition & 0 deletions .gitignore
View file
Open in desktop
Expand Up @@ -13,3 +13,4 @@

# Dependency directories (remove the comment below to include it)
# vendor/
/vsphere-problem-detector
16 changes: 8 additions & 8 deletions pkg/operator/metrics.go
View file
Open in desktop
Expand Up @@ -23,10 +23,10 @@ var (
[]string{checkNameLabel},
)

clusterCheckErrrorMetric = metrics.NewCounterVec(
&metrics.CounterOpts{
Name: "vsphere_cluster_check_errors_total",
Help: "Number of failed vSphere cluster-level checks performed by vsphere-problem-detector.",
clusterCheckErrrorMetric = metrics.NewGaugeVec(
&metrics.GaugeOpts{
Name: "vsphere_cluster_check_errors",
Help: "Indicates failing vSphere cluster-level checks performed by vsphere-problem-detector. Value of 1 means - a particular check is failing.",
StabilityLevel: metrics.ALPHA,
},
[]string{checkNameLabel},
Expand All @@ -41,10 +41,10 @@ var (
[]string{checkNameLabel, nodeNameLabel},
)

nodeCheckErrrorMetric = metrics.NewCounterVec(
&metrics.CounterOpts{
Name: "vsphere_node_check_errors_total",
Help: "Number of failed vSphere node-level checks performed by vsphere-problem-detector.",
nodeCheckErrrorMetric = metrics.NewGaugeVec(
&metrics.GaugeOpts{
Name: "vsphere_node_check_errors",
Help: "Indicates failing vSphere node-level checks performed by vsphere-problem-detector. Value of 1 means - a particular check is failing on a node.",
StabilityLevel: metrics.ALPHA,
},
[]string{checkNameLabel, nodeNameLabel},
Expand Down
6 changes: 4 additions & 2 deletions pkg/operator/operator.go
View file
Open in desktop
Expand Up @@ -203,9 +203,10 @@ func (c *vSphereProblemDetectorController) runSingleClusterCheck(checkContext *c
err := checkFunc(checkContext)
if err != nil {
res.Error = err
clusterCheckErrrorMetric.WithLabelValues(name).Inc()
clusterCheckErrrorMetric.WithLabelValues(name).Set(1)
klog.V(2).Infof("%s failed: %s", name, err)
} else {
clusterCheckErrrorMetric.WithLabelValues(name).Set(0)
klog.V(2).Infof("%s passed", name)
}
clusterCheckTotalMetric.WithLabelValues(name).Inc()
Expand Down Expand Up @@ -265,9 +266,10 @@ func (c *vSphereProblemDetectorController) runSingleNodeSingleCheck(checkContext
err := check.CheckNode(checkContext, node, vm)
if err != nil {
res.Error = err
nodeCheckErrrorMetric.WithLabelValues(name, node.Name).Inc()
nodeCheckErrrorMetric.WithLabelValues(name, node.Name).Set(1)
klog.V(2).Infof("%s:%s failed: %s", name, node.Name, err)
} else {
nodeCheckErrrorMetric.WithLabelValues(name, node.Name).Set(0)
klog.V(2).Infof("%s:%s passed", name, node.Name)
}
nodeCheckTotalMetric.WithLabelValues(name, node.Name).Inc()
Expand Down