Skip to content

feat: add alert types for config mismatch, erasure set health, KMS, and storage capacity#528

Merged
Praveenrajmani merged 1 commit intominio:mainfrom
Praveenrajmani:alerts
Apr 1, 2026
Merged

feat: add alert types for config mismatch, erasure set health, KMS, and storage capacity#528
Praveenrajmani merged 1 commit intominio:mainfrom
Praveenrajmani:alerts

Conversation

@Praveenrajmani
Copy link
Copy Markdown
Contributor

@Praveenrajmani Praveenrajmani commented Jan 29, 2026 •
edited
Loading

Introduces a new GetAlerts admin API (POST /admin/alerts) that streams alert events from the MinIO server as newline-delimited JSON, using Go 1.23's iter.Seq2 for idiomatic range-over-function iteration.

New types:

  • Alert — the top-level alert event with type, timestamp, title, message, dedup key, and deployment/cluster metadata
  • AlertDetails — strongly-typed union struct; only one field populated per alert type
  • AlertLogOpts — query options (filter by type, check interval, max per node)

Alert types and their detail structs:

  • license-expiry (LicenseExpiryDetails) — license ID, expiry time, state
  • certificate-expiry (CertificateExpiryDetails) — common name, serial, DNS names, days until expiry
  • config-mismatch (ConfigMismatchDetails) — peer address, mismatch type (env/args/endpoints), differing values
  • erasure-set-health (ErasureSetHealthDetails) — pool/set ID, online/total drives, read+write quorum, condition
  • kms-unavailable (KMSUnavailableDetails) — KMS type
  • storage-capacity (StorageCapacityDetails) — free/total bytes, free percent

Also adds AlertInternalSubSys, AlertWebhookSubSys, and AlertKafkaSubSys subsystem constants to parse-config.go for alert routing configuration.

klauspost
klauspost reviewed Jan 29, 2026
View reviewed changes
Copy link
Copy Markdown
Contributor

@klauspost klauspost left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Some clarification needed.

@Praveenrajmani Praveenrajmani self-assigned this Feb 2, 2026
@Praveenrajmani Praveenrajmani force-pushed the alerts branch 3 times, most recently from 5a93af7 to 7895a80 Compare February 8, 2026 16:35
@Praveenrajmani Praveenrajmani force-pushed the alerts branch 2 times, most recently from 63d71f9 to 68bf62f Compare February 18, 2026 09:59
@Praveenrajmani Praveenrajmani changed the title feat: add alerting API and types feat: add alert types for config mismatch, erasure set health, KMS, and storage capacity Mar 25, 2026
@Praveenrajmani Praveenrajmani marked this pull request as ready for review March 25, 2026 15:24
shtripat
shtripat reviewed Mar 25, 2026
View reviewed changes
shtripat
shtripat approved these changes Mar 26, 2026
View reviewed changes
Copy link
Copy Markdown
Contributor

@shtripat shtripat left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

looks good overall

@ramondeklein
Copy link
Copy Markdown
Contributor

ramondeklein commented Mar 27, 2026

@Praveenrajmani Can you please update the description why this is needed? I guess it's being used in AIStor code to send out alerts?

@klauspost
Copy link
Copy Markdown
Contributor

klauspost commented Mar 30, 2026

Honestly it seems overengineered that each alert type must have its own data structure.


// Alert represents a single alert event in the system.
// It captures alert information with contextual metadata including deployment and cluster information.
type Alert struct {
	Type         AlertType    `json:"type"`
	Timestamp    time.Time    `json:"timestamp"`
	Title        string       `json:"title"`
	Message      string       `json:"message"`
	Custom      map[string]string `json:"custom"`
	DeploymentID string       `json:"deploymentId"`
	ClusterName  string       `json:"clusterName"`
	DedupKey     string       `json:"dedupKey,omitempty"`
}

I know you can't extract specific values that easily, but honestly I don't see a need in 99% of cases.

feat: add alert types for config mismatch, erasure set health, KMS, a…
fd1a4f9 
…nd storage capacity

Extends the alerting system with four new AlertType constants and their
corresponding detail structs: ConfigMismatchDetails (peer address, mismatch
type, env/args/endpoint diffs), ErasureSetHealthDetails (drive counts,
write quorum, condition), KMSUnavailableDetails (KMS type), and
StorageCapacityDetails (free/total bytes and percentage).
@Praveenrajmani
Copy link
Copy Markdown
Contributor Author

Praveenrajmani commented Mar 30, 2026

Yeah, agree @klauspost we dont want individual types. Have simplified it now. PTAL.

@Praveenrajmani Praveenrajmani merged commit 48fa5ea into minio:main Apr 1, 2026
5 checks passed
@Praveenrajmani Praveenrajmani deleted the alerts branch April 1, 2026 08:36
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@shtripat shtripat shtripat approved these changes

@dhananjaykrutika dhananjaykrutika dhananjaykrutika approved these changes

@csumisha csumisha csumisha approved these changes

@klauspost klauspost Awaiting requested review from klauspost

@harshavardhana harshavardhana Awaiting requested review from harshavardhana

@vadmeste vadmeste Awaiting requested review from vadmeste

@ramondeklein ramondeklein Awaiting requested review from ramondeklein

@allanrogerr allanrogerr Awaiting requested review from allanrogerr

@kanagarajkm kanagarajkm Awaiting requested review from kanagarajkm

Assignees

@Praveenrajmani Praveenrajmani

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

6 participants

@Praveenrajmani @ramondeklein @klauspost @shtripat @dhananjaykrutika @csumisha