Integrations
Connects to the tools you use
Set up connections through ERP.AI and Proto workflows, with every change logged.
- Container & K8s Management
- Kubernetes API
- Telemetry provider
- Deployment pipeline
- Identity provider
Container & K8s Management · Available now
Example screen with sample data
Connect cluster evidence to service decisions.
Container & K8s Management keeps a named cluster and its workload observations alongside the service records used during an incident. Platform operators record the cluster, its environment and owner, then compare desired and ready replicas for a workload.
- No per-user fees
- Manual work is free
- Configure it in Proto
OverviewProblem investigation×
Container & K8s Management · Problems
New problem
Problem investigation
Problems in one list
Problems in view
6Open
2Investigating
2Known error
2Open2
Workers restart during peak trafficMemory limit below peak working set
12 Oct
Batch pods remain pendingStorage class quota reached
12 Oct
Add
Investigating2
Ingress drops long-lived connectionsIngress timeout shorter than client retry
23 Oct
Metrics storage fills each weekendRetention job skipped old partitions
21 Oct
Add
Known error2
DNS resolution slows after scale-upNode DNS cache exhausted
01 Nov
Image pulls stall on new nodesRegistry mirror credentials expired
17 Oct
Add
Workers restart during peak trafficProblems · PRB-11598
OpenRoot causeMemory limit below peak working set
ServiceCustomer portal API
OwnerSamir Haddad
Opened at12 Oct
Approve changePRB-11598 · Workers restart during peak traffic
Change Approver decides whether the reviewed service change proceeds.
Waiting for Change Approver
ApproveReturn
- Tables, ready to use
- 12
- Saved views
- 14
- Roles with set permissions
- 6
- Reports and dashboards
- 8
Features
From cluster reading to service record
Every feature works on the same records. Pick one to see the screen your team works in.
Clusters
Keep each Kubernetes environment, provider and accountable owner identifiable.
- Register
- Observe
- Respond
- Clusters
- Deployment observations
- People
OverviewCluster register×
Container & K8s Management · Clusters
New
Cluster register
Clusters
| Cluster name | Provider | Version | Environment | Owner | |
|---|---|---|---|---|---|
| Production west cluster | Private datacenter | v1.29.8 | Production west | Nikhil Rao | |
| Production east cluster | Northfield cloud region | v1.29.9 | Production east | Leo Fischer | |
| Staging services cluster | Harborline managed hosting | v1.30.3 | Staging | Marta Kowalska | |
| Development sandbox cluster | Eastmere colocation | v1.30.4 | Development | Samir Haddad | |
| Analytics batch cluster | Westhaven cloud region | v1.31.1 | Disaster recovery | Dario Silva |
6 of 6 records
Production west clusterClusters · CLS-27960
ProviderPrivate datacenter
Versionv1.29.8
EnvironmentProduction west
OwnerNikhil Rao
Deployment Observations
Record workload readiness by cluster, namespace and observation time.
- Locate
- Measure
- Compare
- Deployment observations
- Clusters
- Incidents
- Incident updates
OverviewWorkload readiness×
Container & K8s Management · Deployment Observations
New
Workload readiness
Deployment observations
| Cluster | Namespace | Workload | Observed at | Desired replicas | Ready replicas | |
|---|---|---|---|---|---|---|
| Development sandbox cluster | customer-portal | portal-api deployment | 15 Oct | 720 | 420 | |
| Edge gateway cluster | billing-prod | billing-consumer deployment | 11 Oct | 80 | 360 | |
| Disaster recovery cluster | inventory-sync | inventory-sync cronjob | 22 Oct | 280 | 640 | |
| Production east cluster | identity-gateway | identity-proxy daemonset | 16 Oct | 500 | 590 | |
| Edge gateway cluster | reporting-jobs | report-query deployment | 16 Oct | 320 | 650 |
6 of 6 records
customer-portalDeployment observations · DO-49855
ClusterDevelopment sandbox cluster
Workloadportal-api deployment
Observed at15 Oct
Desired replicas720
Ready replicas420
Incidents
Assign service issues, keep attributed updates and record resolution.
- Intake
- Ownership
- History
- Incidents
- Incident updates
- Services
- Assets
- People
- Deployment observations
OverviewCritical incident list×
Container & K8s Management · Incidents
New
Critical incident list
Incidents
| Summary | Priority | Status | Service | Asset | Assigned to | |
|---|---|---|---|---|---|---|
| Restart west pool workers | Critical | Open | Inventory sync service | West pool worker 4 | Farah Malik | |
| Reduce ingress controller replicas | Critical | Acknowledged | Billing event processor | East rack ingress node | Marta Kowalska | |
| Portal API latency above SLO | Critical | Investigating | Order routing service | Build runner node 3 | Dario Silva | |
| Rotate expiring gateway certificate | Critical | Resolved | Order routing service | West pool worker 4 | Farah Malik | |
| Billing queue backlog is growing | Critical | Closed | Reporting query service | North rack control node 1 | Adaeze Nwosu |
6 of 6 records
Restart west pool workersIncidents · INC-87643
CriticalStatusOpen
ServiceInventory sync service
AssetWest pool worker 4
Assigned toFarah Malik
Opened at22 Oct
Changes
Review service change risk, schedule, rollback plan and outcome.
- Prepare
- Decide
- Confirm
- Changes
- Services
- Incidents
- Problems
- Deployment observations
OverviewChanges for review×
Container & K8s Management · Changes
New
Changes for review
Approval conditions
| Summary | Risk | Status | Scheduled start | Scheduled end | Rollback plan | |
|---|---|---|---|---|---|---|
| Restart west pool workers | Low | Review | 18 Oct | 27 Oct | Restore prior deployment revision | |
| Reduce ingress controller replicas | Medium | Review | 23 Oct | 27 Oct | Reapply previous ingress manifest | |
| Portal API latency above SLO | High | Review | 24 Oct | 27 Oct | Return replicas to prior count | |
| Rotate expiring gateway certificate | Low | Review | 13 Oct | 18 Oct | Restore sealed secret backup | |
| Billing queue backlog is growing | Medium | Review | 02 Nov | 08 Nov | Pin workloads to old node pool |
6 of 6 records
Restart west pool workersChanges · CHN-55528
LowStatusReview
Scheduled start18 Oct
Scheduled end27 Oct
Rollback planRestore prior deployment revision
ServiceReporting query service
Problems
Track recurring service issues, root cause work and an accountable owner.
- Identify
- Investigate
- Resolve
- Problems
- Services
- Incidents
- Incident updates
- Changes
- Deployment observations
OverviewMy incidents×
Container & K8s Management · Problems
New
My incidents
Incidents
| Summary | Status | Priority | Service | Asset | Assigned to | |
|---|---|---|---|---|---|---|
| Restart west pool workers | Open | Low | Order routing service | East rack ingress node | Leo Fischer | |
| Reduce ingress controller replicas | Acknowledged | Medium | Document rendering worker | North rack control node 2 | Ethan Brooks | |
| Portal API latency above SLO | Investigating | High | Order routing service | Metrics collector node | Ethan Brooks | |
| Rotate expiring gateway certificate | Resolved | Critical | Order routing service | North rack control node 2 | Marta Kowalska | |
| Billing queue backlog is growing | Closed | Low | Order routing service | Build runner node 3 | Nikhil Rao |
6 of 6 records
Restart west pool workersIncidents · INC-78839
OpenPriorityLow
ServiceOrder routing service
AssetEast rack ingress node
Assigned toLeo Fischer
Service Requests
Track requested service work from submission through a decision and completion.
- Submit
- Decide
- Fulfill
- Service requests
- People
- Services
- Changes
OverviewSubmitted requests×
Container & K8s Management · Service Requests
New
Submitted requests
Approval waits for the request approver
| Request type | Status | Requester | Service | Requested at | Due at | |
|---|---|---|---|---|---|---|
| Namespace access | Submitted | Samir Haddad | Notification dispatcher | 16 Oct | 22 Oct | |
| Deployment approval | Submitted | Leo Fischer | Identity gateway | 19 Oct | 28 Oct | |
| Capacity increase | Submitted | Samir Haddad | Inventory sync service | 16 Oct | 24 Oct | |
| Secret rotation | Submitted | Camila Torres | Inventory sync service | 22 Oct | 30 Oct | |
| Log retention change | Submitted | Yuki Tanaka | Billing event processor | 13 Oct | 16 Oct |
6 of 6 records
Namespace accessService requests · SR-95440
SubmittedRequesterSamir Haddad
ServiceNotification dispatcher
Requested at16 Oct
Due at22 Oct
Service Records
Maintain service ownership, assets and receipts that support operational work.
- Own
- Locate
- Reconcile
- Services
- Assets
- Integration receipts
- People
- Users
OverviewService asset register×
Container & K8s Management · Service Records
New
Service asset register
Assets
Assigned3
NODE-WEST-01North rack A3
NODE-EAST-01West rack C2
STOR-BACKUP-02East rack B1
Add
Maintenance3
NODE-WEST-02North rack A4
NODE-EAST-02West rack C3
RUNNER-BUILD-03Backup room D2
Add
Process
Follow the operating path
Observation to verified change ships ready to use. One step needs a named approval before the work moves on: Approve change (Change Approver).
Drag a card to another stage, or focus it and use the arrow keys. It updates to match the stage.
-
Record cluster
Name the environment and its owner.
Workers restart during peak trafficPRB-11598 · Platform OperatorRecord clusterBy Platform Operator -
Observe workload
Record dated desired and ready replicas.
Workers restart during peak trafficPRB-11598 · Platform OperatorObserve workloadBy Platform Operator -
Investigate incident
Assign service impact and keep attributed updates.
Workers restart during peak trafficPRB-11598 · Incident ResponderInvestigate incidentBy Incident Responder -
Approve change
Review risk, window and rollback plan.
Workers restart during peak trafficPRB-11598 · Change ApproverApprove changeBy Change ApproverApprovalChange Approver decides whether the reviewed service change proceeds. -
Verify change
Record the implemented or rolled-back outcome.
Workers restart during peak trafficPRB-11598 · Platform OperatorVerify changeBy Platform Operator -
Close incident
Confirm the service result and retain the history.
Workers restart during peak trafficPRB-11598 · Incident ResponderClose incidentBy Incident Responder
Inside a record
Connect the sources you use
Open any problem to see the roles that work on it, its process stages and its approvals on one page.
- 1Record details
Root cause, Service, Owner and Opened at.
- 2Linked records
Records linked to this problem.
- 3Who works on it
Platform Operator, Incident Responder and Service Owner, each with their own permissions.
- 4Process stages
Record problem, with the role that acts.
- 5Approvals
Approve change, with who decides.
Problem investigationWorkers restart during peak traffic×
Container & K8s Management · Problems
Log activity
Workers restart during peak traffic
Memory limit below peak working set · Open
1
Record details
Root causeMemory limit below peak working set
ServiceCustomer portal API
OwnerSamir Haddad
Opened at12 Oct
3
Who works on it
- Platform OperatorManage Clusters, deployment observations, Changes
- Incident ResponderManage Incidents, incident updates
- Service OwnerManage Services, Problems, Assets
4
Process stages
- Record problemService Owner
5
Approvals
- Approve changeChange ApproverApproved
Approvals
Decisions with a named owner
OverviewApprovals×
Container & K8s Management · Overview
Approvals
Signed in as Change Approver
Waiting for you
4 To review1 Overdue2 Decided todayOnly mine
0 selectedApproveReturn
- SR-95440Namespace accessSubmitted1dActions
- SR-62957Deployment approvalSubmitted1dActions
- SR-11196Capacity increaseSubmitted5dActions
- SR-92508Secret rotationSubmitted4dActions
Who approves what
| Decision | Approver | |
|---|---|---|
| Approve changeReview risk, window and rollback plan. | Change Approver | |
| Approve Changes | Change Approver | |
| Approve service requests | Request Approver |
3 approval rules · set per role in Proto
Roles
A workspace for each role
Everyone opens Container & K8s Management to the work that is theirs. Pick a role to see what it starts with.
Platform Operator
Maintains the cluster register and records workload observations and implementation evidence.
This is the first screen this role sees.
OverviewChange window schedule×
Change window schedule
Changes · Platform Operator
| Summary | Status | Risk | Scheduled start | Scheduled end | Service | |
|---|---|---|---|---|---|---|
| Restart west pool workers | Approved | Low | 28 Oct | 05 Nov | Identity gateway | |
| Reduce ingress controller replicas | Implementing | Medium | 14 Oct | 16 Oct | Reporting query service | |
| Portal API latency above SLO | Approved | High | 29 Oct | 01 Nov | Customer portal API | |
| Rotate expiring gateway certificate | Implementing | Low | 03 Nov | 12 Nov | Billing event processor | |
| Billing queue backlog is growing | Approved | Medium | 03 Nov | 12 Nov | Inventory sync service |
6 of 6 records
Incident Responder
Owns investigation and attributed updates for assigned service incidents.
This is the first screen this role sees.
OverviewProblem investigation×
Problem investigation
Problems · Incident Responder
Open2
Workers restart during peak trafficMemory limit below peak working set
12 Oct
Batch pods remain pendingStorage class quota reached
12 Oct
Add
Investigating2
Ingress drops long-lived connectionsIngress timeout shorter than client retry
23 Oct
Metrics storage fills each weekendRetention job skipped old partitions
21 Oct
Add
Known error2
DNS resolution slows after scale-upNode DNS cache exhausted
01 Nov
Image pulls stall on new nodesRegistry mirror credentials expired
17 Oct
Add
Service Owner
Maintains service context and owns problem investigation for that service.
This is the first screen this role sees.
OverviewChange window schedule×
Change window schedule
Changes · Service Owner
| Summary | Status | Risk | Scheduled start | Scheduled end | Service | |
|---|---|---|---|---|---|---|
| Restart west pool workers | Approved | Low | 28 Oct | 05 Nov | Identity gateway | |
| Reduce ingress controller replicas | Implementing | Medium | 14 Oct | 16 Oct | Reporting query service | |
| Portal API latency above SLO | Approved | High | 29 Oct | 01 Nov | Customer portal API | |
| Rotate expiring gateway certificate | Implementing | Low | 03 Nov | 12 Nov | Billing event processor | |
| Billing queue backlog is growing | Approved | Medium | 03 Nov | 12 Nov | Inventory sync service |
6 of 6 records
Change Approver
Reviews change risk, schedule and rollback plan before a change is implemented.
This is the first screen this role sees.
OverviewChange window schedule×
Change window schedule
Changes · Change Approver
| Summary | Status | Risk | Scheduled start | Scheduled end | Service | |
|---|---|---|---|---|---|---|
| Restart west pool workers | Approved | Low | 28 Oct | 05 Nov | Identity gateway | |
| Reduce ingress controller replicas | Implementing | Medium | 14 Oct | 16 Oct | Reporting query service | |
| Portal API latency above SLO | Approved | High | 29 Oct | 01 Nov | Customer portal API | |
| Rotate expiring gateway certificate | Implementing | Low | 03 Nov | 12 Nov | Billing event processor | |
| Billing queue backlog is growing | Approved | Medium | 03 Nov | 12 Nov | Inventory sync service |
6 of 6 records
Request Approver
Decides whether a submitted service request proceeds to fulfillment.
This is the first screen this role sees.
OverviewSubmitted requests×
Submitted requests
Service requests · Request Approver
| Request type | Status | Requester | Service | Requested at | Due at | |
|---|---|---|---|---|---|---|
| Namespace access | Submitted | Samir Haddad | Notification dispatcher | 16 Oct | 22 Oct | |
| Deployment approval | Submitted | Leo Fischer | Identity gateway | 19 Oct | 28 Oct | |
| Capacity increase | Submitted | Samir Haddad | Inventory sync service | 16 Oct | 24 Oct | |
| Secret rotation | Submitted | Camila Torres | Inventory sync service | 22 Oct | 30 Oct | |
| Log retention change | Submitted | Yuki Tanaka | Billing event processor | 13 Oct | 16 Oct |
6 of 6 records
Admin
Configures access, people mappings and event receipt handling.
- Manage Users, People, integration receipts
- View Services, Assets, Clusters, deployment observations, Incidents, Changes, Problems, service requests
Reports
Read workload and service outcomes
Review workload readiness, service incidents, changes and request outcomes.
Overview×
Container & K8s Management · Overview
Container & K8s Management dashboard
This quarter
Changes in view
6Approved
3Implementing
3Problems by status
How far records got this quarter
| Stage | Reached | Step | |
|---|---|---|---|
| Open | 33 | ||
| Investigating | 24 | 73% | |
| Known error | 12 | 50% | |
| Resolved | 6 | 50% |
Number of problems by status
Sample counts per stage
- 01Workload readiness by cluster
Compare desired replicas and ready replicas by Cluster, Namespace and Workload at each observed at.
- 02Open incidents by service and priority
Group Incidents by Service, Priority and Status to focus response.
- 03Incident resolution time
Compare opened at, acknowledged at and resolved at for resolved Incidents.
- 04Problem status by service
Group Problems by Service, Owner and Status and inspect root cause.
- 05Change schedule and outcomes
Review scheduled start, scheduled end, Risk and Status for Changes.
- 06Request completion
Compare requested at, due at and completed at by service requests Status and Service.
- 07Asset state by service
Group Assets by Service and Status with asset tag and Location.
- 08Integration receipt outcomes
Review Provider, external event ID, received at and Status for integration receipts.
Who it's for
Fit your operations team
See how Container & K8s Management fits your kind of team.
Controls
Keep decisions and intake distinct
Each stage of observation to verified change names the role that acts and the status it records, so a change is never anonymous.
OverviewObservation to verified change×
Container & K8s Management · Process
Observation to verified change
Each stage, the role that acts and the status it records
Process stages
10Approval steps
1Roles involved
4Tables touched
9Acknowledge incident
Incident ResponderStatusOpen → Acknowledged
RecordsIncidents, Changes, People
Investigate incident
Incident ResponderStatusAcknowledged → Investigating
RecordsIncident updates, Incidents, Changes
Record problem
Service OwnerStatusOpen → Investigating
RecordsProblems, Services, Changes
- StartIncidentsAcknowledge incidentOpen → Acknowledged
- StepIncident updatesInvestigate incidentAcknowledged → Investigating
- StepProblemsRecord problemOpen → Investigating
- StepChangesDraft changeDraft → Review
- ApprovalChangesApprove changeReview → Approved
AI agent
Prepare work with Proto
Ask Proto to prepare cluster evidence and service work for human review.
It can
- Platform Operator: Compare the latest desired and ready replicas for a named workload and cluster.
- Incident Responder: Gather the assigned incident's service, asset and dated updates for shift handoff.
- Service Owner: Summarize open problems and reviewed changes for a service.
- Change Approver: Prepare a list of changes in Review with risk, window and rollback plan.
Stays with a person
- A named Change Approver decides whether a reviewed change proceeds.
- The Request Approver decides submitted service work.
- The Incident Responder confirms the service outcome before closure.
- Proto can organize records and prepare a recommendation, but these decisions follow the team's configured access and review process.
For the checkout workload in the production cluster, compare the latest desired and ready replica readings.
Show the incidents assigned to me with their latest attributed updates.
List service changes in Review with their risk, scheduled window and rollback plan.
A named Change Approver decides whether a reviewed change proceeds.
Data model
Your data, in tables you can change
Every record lives in a table you can see, extend and export. Add fields or whole tables in Proto.
Browse all 12 tables12tables
- Clusters3
- Deployment Observations4
- Incidents6
- Changes5
- Problems6
- Service Requests4
- Service Records5
Implementation
Roll out one team at a time
Start with one team and one process, then add the rest once the basics are running.
Set the operating boundary
Choose the first clusters, services and workloads. Name their owners and agree which outside systems provide live readings and deployment execution.
Map identities and records
Link People to Users, Services to owners and Assets, and DeploymentObservations to Clusters. Preserve stable event identifiers during any import.
Configure the pilot
Use a production and a test cluster with a recognizable workload. Record an observation, open an incident for the affected service and assign the responder.
Rehearse decisions
Have the Change Approver judge a planned window and rollback plan. Test an approved and rejected request, a rolled-back change and a failed integration receipt.
Review and release
Compare source events and imported records, check role access, and confirm that later observations and incident updates explain the operating outcome.
Pricing
$0per user
The base account and everyday work in the ERP.AI app are free, for every user.
- No per-user or per-seat fees
- Viewing, editing, approving and exporting records in the app is free
- Pay only for machine work: API calls, imports, workflows and AI agents
- Prepaid credits from $5. Purchased credits never expire
Machine work rate card Included per $1 of credit
| Usage | Per $1 |
|---|---|
| Record reads (list, filter, get) | 20,000 |
| Search or aggregate queries | 4,000 |
| Record writes | 4,000 |
| Imported or synced records | 200,000 |
| Workflow runs | 2,000 |
| AI agent runs | 2,000 |
| Scheduled jobs or webhooks | 20,000 |
| File uploads (MB) | 2,000 |
| CSV or PDF exports | 100 |
How do I get started?
Click Get started, then use Proto to configure Container & K8s Management for your business. Pilot a named cluster, one service and a workload whose desired and ready replicas your team can check.
What does a workload observation contain?
It identifies a cluster, namespace, workload, observation time and desired and ready replica figures. Use it with the service and incident records when investigating impact.
Can I distinguish production and test environments?
Yes. Each cluster has an environment and named owner. Agree on environment labels before loading observations so similar workload names remain distinguishable.
How do incidents get an owner?
Assign the incident to a person and record the acknowledgment time. Keep later findings as dated updates with their authors.
How are changes approved?
A Change Approver reviews the service, risk, scheduled start and end, and rollback plan before recording an approval or rejection. The operator records implementation and its result separately.
Can a service request replace a change review?
A request captures the desired service work and its approval. Production intervention can have its own change record with risk, window, rollback plan and review decision.
Where do Kubernetes readings come from?
Agree on the source for cluster and workload data during implementation. A Kubernetes API or telemetry connection can supply readings once configured and reconciled with the named records.
How do we review outside events?
Integration receipts retain provider and external event identifiers, a received time, payload hash and intake outcome. Review rejected and failed receipts against the provider source.
Can we track recurring failures?
Use a problem record for the service, with a named owner, root cause findings and a status that distinguishes investigation from known error and resolution.
What drives the cost?
Manual use follows the platform pricing policy. Imports, integrations and agent activity use the published metering rules; implementation scope is agreed separately.
Set up Container & K8s Management your way.
Start free, then use Proto to change fields, steps and approvals before your team goes live.