Troubleshoot GKE TPU slice creation blocked by an unstable MIG
This skill helps diagnose GKE TPU Dynamic Slice creation failures and JobSet
scheduling delays when the Compute Engine managed instance groups (MIGs) that
back the TPU node pool aren't stable (status.isStable: false).
A SliceCreationFailed status reason means that slice creation failed because
prerequisites weren't met (for example, required Compute Engine resources are
missing) or preflight checks failed (see
Use dynamic slicing with a custom scheduler).
A MIG that is still creating, updating, or repairing instances is one cause to
check. This skill checks the slice status first, then the stability of the
backing MIGs.
Attribution
MUST READ: Before executing any gcloud, REST API, or MCP command, you
MUST read Usage Attribution. For commands executed
directly in the terminal (never in generated scripts, Terraform, or user code),
prefix gcloud with
CLOUDSDK_METRICS_REQUEST_ATTRIBUTION="agent-skills/1.0 (skill:gke-ai-troubleshooting-tpu-slice-mig-stuck-updating)"
and append agent-skills/1.0 (skill:gke-ai-troubleshooting-tpu-slice-mig-stuck-updating) to User-Agent headers.
Prerequisites
- Google Cloud SDK & Tools:
- For installing and initializing
gcloud (project selection and
authentication) and enabling the GKE API, follow
Before you begin.
- For installing
kubectl and configuring cluster credentials, follow the
"Install kubectl", "Install required plugins", and "Store cluster
information for kubectl" sections of
Install kubectl and configure cluster access.
- API Enablement & Billing:
- Verify that the project has an active Cloud Billing account.
- Enable APIs:
container.googleapis.com, compute.googleapis.com, and
logging.googleapis.com.
- Required IAM Roles:
- Logs Viewer (
roles/logging.viewer): Inspect Cloud Audit Logs for MIG
errors.
- Compute Viewer (
roles/compute.viewer): Check MIG stability (isStable)
and instance progress.
- Kubernetes Engine Viewer (
roles/container.viewer): Inspect Slice custom
resources (slices.accelerator.gke.io), nodes, and cluster operations.
- Documentation:
- Use dynamic slicing with a custom scheduler
(Sections: "Before you begin", "Create a Slice custom resource", "Monitor
the status of the Slice custom resource", "Verify the status of the nodes
and the partitions", "Upgrade your node pools")
- Install kubectl and configure cluster access
(Sections: "Install kubectl", "Install required plugins", "Store cluster
information for kubectl")
- Add and manage node pools
(Sections: "View node pools in a Standard cluster")
- View information about MIGs and managed instances
(Sections: "Check whether a MIG is stable", "Check the status of managed
instances", "List instance errors")
- Automatically apply VM configuration updates in a MIG
(Sections: "Verify whether an update rollout is complete")
- Manually upgrade a cluster's control plane or node pools
(Sections: "Check node pool upgrade status", "Cancel a node pool upgrade",
"Resume a node pool upgrade", "Roll back a node pool upgrade")
- Auto-repair nodes
(Sections: "Get information about recent automated repair events")
- Troubleshoot GKE Standard node pools
(Sections: "Issue: Node pool creation fails due to insufficient resources")
- Troubleshoot TPUs in GKE
(Sections: "Insufficient quota to satisfy the TPU request", "Insufficient
TPU resources to satisfy the TPU request")
- REST Resource: projects.locations.clusters.nodePools
(Sections: "Resource: NodePool")
- Schedule dynamic slices with Kueue and TAS
(Sections: "Monitor the status of the slice")
- About TPUs in GKE
- Get support
(Sections: "Before you contact support")
Read-only rule: Run read-only diagnostic commands only. Never drain,
delete, or re-create nodes, or run any other command that changes the cluster.
Give the user any fix to apply themselves.
When you recommend a fix, link the doc section that describes it.
Diagnostic workflow
Step 0: Collect context and set the investigation window [Low Risk]
Collect the necessary contextual parameters from the user prompt or environment:
{project_id}, {cluster_name}, {location}, {nodepool_name}, and
{issue_time}.
{mig_name} and {zone}: Name and zone of the managed instance group that
backs {nodepool_name}, from the node pool's instanceGroupUrls field (see
View node pools in a Standard cluster
and
Resource: NodePool).
If {nodepool_name} or {mig_name} isn't already known from a failing
Slice, find the affected partition, nodes, and backing MIG by following
Verify the status of the nodes and the partitions.
- Time window: By default, query a 60-minute window
[{issue_time} - 30m, {issue_time} + 30m] ({start_time} to {end_time}).
Step 1: Detect slice creation failures and MIG errors [Low Risk]
- Inspect Slice status and partition nodes: Check the
Slice custom
resource Status.Conditions for a SliceCreationFailed status reason and
identify the affected partitions, nodes, and {mig_name} by following the
"Monitor the status of the Slice custom resource" and "Verify the status of
the nodes and the partitions" sections of
Use dynamic slicing with a custom scheduler,
and
Monitor the status of the slice.
- Query Cloud Logging for MIG errors: Once
{mig_name} is identified (or
by narrowing the results to {mig_name} once known), run this Cloud Logging
query to detect errors on the backing instance group manager across
[{start_time}, {end_time}]:resource.type="gce_instance_group_manager"
resource.labels.instance_group_manager_name="{mig_name}"
severity>=ERROR
timestamp >= "{start_time}" AND timestamp <= "{end_time}"
- Findings:
- If the
Slice status reports SliceCreationFailed or the logs show
instance group errors, proceed to Step 2.
- If no error logs match and the TPU
Slice status is ACTIVE, rule out MIG
instability and check workload configuration.
Step 2: Check MIG stability (status.isStable) [Low Risk]
Follow
Check whether a MIG is stable
to inspect the stability (status.isStable) and version target status
(status.versionTarget.isReached) of {mig_name}:
- What the fields mean:
status.isStable: false: Changes are active or pending, or the MIG itself
is being modified. Proceed to Step 3.
status.versionTarget.isReached: false: At least one instance isn't running
the target version yet (see
Automatically apply VM configuration updates in a MIG).
Proceed to Step 3.
status.isStable: true and status.versionTarget.isReached: true: No
instances are undergoing or scheduled for changes (currentAction is NONE
for all instances), and the MIG itself isn't being modified. This rules out
MIG instability. If the Slice is still in SliceCreationFailed, hand off
to gke-ai-troubleshooting-tpu-dynamic-slices-monitoring, or follow the
"Create a Slice custom resource" and "Verify the status of the nodes and the
partitions" sections of
Use dynamic slicing with a custom scheduler.
Step 3: Find failed or stalled instance operations [Low Risk]
Direct the user to the "Check the status of managed instances" and "List
instance errors" sections of
View information about MIGs and managed instances
(include both section names and the document title) to inspect individual
instance states (currentAction and instanceStatus) and list the most recent
errors in {mig_name}:
- If the errors show capacity or quota failures such as
ZONE_RESOURCE_POOL_EXHAUSTED_WITH_DETAILS or QUOTA_EXCEEDED, go to Step 5,
item 2.
- Otherwise, proceed to Step 4 to check for node pool upgrade or auto-repair
operations.
Step 4: Check node pool operations [Low Risk]
A MIG that is still changing instances can be the result of a node pool upgrade
or repair:
- Check node pool upgrades: Follow
Check node pool upgrade status
to list the cluster operations and find upgrade operations targeting
{nodepool_name} in [{start_time}, {end_time}]. Also review
Upgrade your node pools
for dynamic slicing upgrade requirements (required surge settings and
deleting active super-slices before upgrading).
- If an upgrade operation is still running, go to Step 5, item 1 or item 3.
- If an upgrade operation was cancelled, failed, or is incomplete, go to Step
5, item 4.
- Check automated node repairs: Follow
Get information about recent automated repair events
to check for repair operations on
{nodepool_name}.
- If a node auto-repair operation is running, go to Step 5, item 1.
- If no operation or error explains the unstable MIG: Go to Step 5, item 5
to contact Cloud Customer Care.
Step 5: Resolution [High Risk]
Guardrail: Don't modify GKE-managed instance groups or VMs directly
through Compute Engine. Manage node pools through GKE.
- Wait for an ongoing update or auto-repair to finish: If a node pool
upgrade or automated repair is still making progress, the user can wait until
the MIG is stable (
status.isStable: true) by following
Check whether a MIG is stable.
- Resolve capacity or quota errors: When Step 3 shows
ZONE_RESOURCE_POOL_EXHAUSTED_WITH_DETAILS or QUOTA_EXCEEDED, rolling back
an upgrade won't resolve the shortage. Follow
Issue: Node pool creation fails due to insufficient resources
and the "Insufficient quota to satisfy the TPU request" and "Insufficient TPU
resources to satisfy the TPU request" sections of
Troubleshoot TPUs in GKE.
- Cancel an in-progress node pool upgrade: If the user wants to stop an
upgrade that is still running, follow
Cancel a node pool upgrade.
For dynamic slicing node pools, also check
Upgrade your node pools
for the required surge settings and deleting active super-slices before
upgrading.
- Resume or roll back a cancelled, failed, or incomplete node pool upgrade:
- Follow the "Resume a node pool upgrade" or "Roll back a node pool upgrade"
sections of
Manually upgrade a cluster's control plane or node pools.
As documented there, rollback applies if an in-progress upgrade was
cancelled, the upgrade failed, or the upgrade is incomplete due to a
maintenance window timing out (you can't roll back node pools once they
have been successfully upgraded); alternatively, a cancelled or failed,
partially completed upgrade can be resumed.
- Afterwards, verify that the MIG is stable by following
Check whether a MIG is stable.
- Escalate unexplained MIG instability: If no node pool upgrade, repair
operation, or capacity/quota error explains why
{mig_name} remains
unstable, direct the user to Cloud Customer Care by following
Get support.
- Verify node pool and slice recovery: After the instance group is stable
(
status.isStable: true), follow the "Monitor the status of the Slice custom
resource" and "Create a Slice custom resource" sections of
Use dynamic slicing with a custom scheduler
to verify that the Slice reaches the ACTIVE state. If the Slice stays
in SliceCreationFailed after the MIG stabilizes, check whether the retry
annotation documented in "Create a Slice custom resource" is set on the
Slice.
1---2name: gke-ai-troubleshooting-tpu-slice-mig-stuck-updating3description: >-4 Diagnose and resolve GKE TPU Dynamic Slice creation and JobSet scheduling5 failures blocked by underlying Compute Engine managed instance groups (MIGs)6 stuck in an updating or unstable state (isStable=False). Use when TPU slice7 creation fails with a SliceCreationFailed status reason, when JobSets remain8 stuck in pending, or when MIG status reflects isStable=False or9 versionTarget.isReached=False during TPU slice formation. Don't use for10 SliceCreationFailed causes unrelated to MIG stability (use11 gke-ai-troubleshooting-tpu-dynamic-slices-monitoring) or general pod12 scheduling failures unrelated to TPU slices.13metadata:14 source: https://skillmd.com/skills/google/gke-ai-troubleshooting-tpu-slice-mig-stuck-updating15 version: "1.0.1"16 category: Containers17---1819# Troubleshoot GKE TPU slice creation blocked by an unstable MIG2021This skill helps diagnose GKE TPU Dynamic Slice creation failures and JobSet22scheduling delays when the Compute Engine managed instance groups (MIGs) that23back the TPU node pool aren't stable (`status.isStable: false`).2425A `SliceCreationFailed` status reason means that slice creation failed because26prerequisites weren't met (for example, required Compute Engine resources are27missing) or preflight checks failed (see28[Use dynamic slicing with a custom scheduler](https://docs.cloud.google.com/kubernetes-engine/docs/how-to/create-dynamic-slices.md.txt)).29A MIG that is still creating, updating, or repairing instances is one cause to30check. This skill checks the slice status first, then the stability of the31backing MIGs.3233## Attribution3435**MUST READ**: Before executing any `gcloud`, REST API, or MCP command, you36MUST read [Usage Attribution](references/attribution.md). For commands executed37directly in the terminal (never in generated scripts, Terraform, or user code),38prefix `gcloud` with39`CLOUDSDK_METRICS_REQUEST_ATTRIBUTION="agent-skills/1.0 (skill:gke-ai-troubleshooting-tpu-slice-mig-stuck-updating)"`40and append `agent-skills/1.0 (skill:gke-ai-troubleshooting-tpu-slice-mig-stuck-updating)` to `User-Agent` headers.4142## Prerequisites4344- **Google Cloud SDK & Tools**:45 - For installing and initializing `gcloud` (project selection and46 authentication) and enabling the GKE API, follow47 [Before you begin](https://docs.cloud.google.com/kubernetes-engine/docs/how-to/create-dynamic-slices.md.txt).48 - For installing `kubectl` and configuring cluster credentials, follow the49 "Install kubectl", "Install required plugins", and "Store cluster50 information for kubectl" sections of51 [Install kubectl and configure cluster access](https://docs.cloud.google.com/kubernetes-engine/docs/how-to/cluster-access-for-kubectl.md.txt).52- **API Enablement & Billing**:53 - Verify that the project has an active Cloud Billing account.54 - Enable APIs: `container.googleapis.com`, `compute.googleapis.com`, and55 `logging.googleapis.com`.56- **Required IAM Roles**:57 - Logs Viewer (`roles/logging.viewer`): Inspect Cloud Audit Logs for MIG58 errors.59 - Compute Viewer (`roles/compute.viewer`): Check MIG stability (`isStable`)60 and instance progress.61 - Kubernetes Engine Viewer (`roles/container.viewer`): Inspect Slice custom62 resources (`slices.accelerator.gke.io`), nodes, and cluster operations.63- **Documentation**:64 - [Use dynamic slicing with a custom scheduler](https://docs.cloud.google.com/kubernetes-engine/docs/how-to/create-dynamic-slices.md.txt)65 (Sections: "Before you begin", "Create a Slice custom resource", "Monitor66 the status of the Slice custom resource", "Verify the status of the nodes67 and the partitions", "Upgrade your node pools")68 - [Install kubectl and configure cluster access](https://docs.cloud.google.com/kubernetes-engine/docs/how-to/cluster-access-for-kubectl.md.txt)69 (Sections: "Install kubectl", "Install required plugins", "Store cluster70 information for kubectl")71 - [Add and manage node pools](https://docs.cloud.google.com/kubernetes-engine/docs/how-to/node-pools.md.txt)72 (Sections: "View node pools in a Standard cluster")73 - [View information about MIGs and managed instances](https://docs.cloud.google.com/compute/docs/instance-groups/getting-info-about-migs.md.txt)74 (Sections: "Check whether a MIG is stable", "Check the status of managed75 instances", "List instance errors")76 - [Automatically apply VM configuration updates in a MIG](https://docs.cloud.google.com/compute/docs/instance-groups/rolling-out-updates-to-managed-instance-groups.md.txt)77 (Sections: "Verify whether an update rollout is complete")78 - [Manually upgrade a cluster's control plane or node pools](https://docs.cloud.google.com/kubernetes-engine/docs/how-to/upgrading-a-cluster.md.txt)79 (Sections: "Check node pool upgrade status", "Cancel a node pool upgrade",80 "Resume a node pool upgrade", "Roll back a node pool upgrade")81 - [Auto-repair nodes](https://docs.cloud.google.com/kubernetes-engine/docs/how-to/node-auto-repair.md.txt)82 (Sections: "Get information about recent automated repair events")83 - [Troubleshoot GKE Standard node pools](https://docs.cloud.google.com/kubernetes-engine/docs/troubleshooting/node-pools.md.txt)84 (Sections: "Issue: Node pool creation fails due to insufficient resources")85 - [Troubleshoot TPUs in GKE](https://docs.cloud.google.com/kubernetes-engine/docs/troubleshooting/tpus.md.txt)86 (Sections: "Insufficient quota to satisfy the TPU request", "Insufficient87 TPU resources to satisfy the TPU request")88 - [REST Resource: projects.locations.clusters.nodePools](https://docs.cloud.google.com/kubernetes-engine/docs/reference/rest/v1/projects.locations.clusters.nodePools.md.txt)89 (Sections: "Resource: NodePool")90 - [Schedule dynamic slices with Kueue and TAS](https://docs.cloud.google.com/kubernetes-engine/docs/how-to/use-gke-dynamic-slicing.md.txt)91 (Sections: "Monitor the status of the slice")92 - [About TPUs in GKE](https://docs.cloud.google.com/kubernetes-engine/docs/concepts/tpus.md.txt)93 - [Get support](https://docs.cloud.google.com/kubernetes-engine/docs/getting-support.md.txt)94 (Sections: "Before you contact support")9596> **Read-only rule**: Run read-only diagnostic commands only. Never drain,97> delete, or re-create nodes, or run any other command that changes the cluster.98> Give the user any fix to apply themselves.99100When you recommend a fix, link the doc section that describes it.101102---103104## Diagnostic workflow105106### Step 0: Collect context and set the investigation window `[Low Risk]`107108Collect the necessary contextual parameters from the user prompt or environment:109- `{project_id}`, `{cluster_name}`, `{location}`, `{nodepool_name}`, and110 `{issue_time}`.111- `{mig_name}` and `{zone}`: Name and zone of the managed instance group that112 backs `{nodepool_name}`, from the node pool's `instanceGroupUrls` field (see113 [View node pools in a Standard cluster](https://docs.cloud.google.com/kubernetes-engine/docs/how-to/node-pools.md.txt)114 and115 [Resource: NodePool](https://docs.cloud.google.com/kubernetes-engine/docs/reference/rest/v1/projects.locations.clusters.nodePools.md.txt)).116 If `{nodepool_name}` or `{mig_name}` isn't already known from a failing117 `Slice`, find the affected partition, nodes, and backing MIG by following118 [Verify the status of the nodes and the partitions](https://docs.cloud.google.com/kubernetes-engine/docs/how-to/create-dynamic-slices.md.txt).119- **Time window**: By default, query a 60-minute window `[{issue_time} - 30m,120 {issue_time} + 30m]` (`{start_time}` to `{end_time}`).121122---123124### Step 1: Detect slice creation failures and MIG errors `[Low Risk]`1251261. **Inspect Slice status and partition nodes**: Check the `Slice` custom127 resource `Status.Conditions` for a `SliceCreationFailed` status reason and128 identify the affected partitions, nodes, and `{mig_name}` by following the129 "Monitor the status of the Slice custom resource" and "Verify the status of130 the nodes and the partitions" sections of131 [Use dynamic slicing with a custom scheduler](https://docs.cloud.google.com/kubernetes-engine/docs/how-to/create-dynamic-slices.md.txt),132 and133 [Monitor the status of the slice](https://docs.cloud.google.com/kubernetes-engine/docs/how-to/use-gke-dynamic-slicing.md.txt).1342. **Query Cloud Logging for MIG errors**: Once `{mig_name}` is identified (or135 by narrowing the results to `{mig_name}` once known), run this Cloud Logging136 query to detect errors on the backing instance group manager across137 `[{start_time}, {end_time}]`:138 ```139 resource.type="gce_instance_group_manager"140 resource.labels.instance_group_manager_name="{mig_name}"141 severity>=ERROR142 timestamp >= "{start_time}" AND timestamp <= "{end_time}"143 ```144- **Findings**:145 - If the `Slice` status reports `SliceCreationFailed` or the logs show146 instance group errors, proceed to Step 2.147 - If no error logs match and the TPU `Slice` status is `ACTIVE`, rule out MIG148 instability and check workload configuration.149150---151152### Step 2: Check MIG stability (`status.isStable`) `[Low Risk]`153154Follow155[Check whether a MIG is stable](https://docs.cloud.google.com/compute/docs/instance-groups/getting-info-about-migs.md.txt)156to inspect the stability (`status.isStable`) and version target status157(`status.versionTarget.isReached`) of `{mig_name}`:158159- **What the fields mean**:160 - `status.isStable: false`: Changes are active or pending, or the MIG itself161 is being modified. Proceed to Step 3.162 - `status.versionTarget.isReached: false`: At least one instance isn't running163 the target version yet (see164 [Automatically apply VM configuration updates in a MIG](https://docs.cloud.google.com/compute/docs/instance-groups/rolling-out-updates-to-managed-instance-groups.md.txt)).165 Proceed to Step 3.166 - `status.isStable: true` and `status.versionTarget.isReached: true`: No167 instances are undergoing or scheduled for changes (`currentAction` is `NONE`168 for all instances), and the MIG itself isn't being modified. This rules out169 MIG instability. If the `Slice` is still in `SliceCreationFailed`, hand off170 to `gke-ai-troubleshooting-tpu-dynamic-slices-monitoring`, or follow the171 "Create a Slice custom resource" and "Verify the status of the nodes and the172 partitions" sections of173 [Use dynamic slicing with a custom scheduler](https://docs.cloud.google.com/kubernetes-engine/docs/how-to/create-dynamic-slices.md.txt).174175---176177### Step 3: Find failed or stalled instance operations `[Low Risk]`178179Direct the user to the "Check the status of managed instances" and "List180instance errors" sections of181[View information about MIGs and managed instances](https://docs.cloud.google.com/compute/docs/instance-groups/getting-info-about-migs.md.txt)182(include both section names and the document title) to inspect individual183instance states (`currentAction` and `instanceStatus`) and list the most recent184errors in `{mig_name}`:185186- If the errors show capacity or quota failures such as187 `ZONE_RESOURCE_POOL_EXHAUSTED_WITH_DETAILS` or `QUOTA_EXCEEDED`, go to Step 5,188 item 2.189- Otherwise, proceed to Step 4 to check for node pool upgrade or auto-repair190 operations.191192---193194### Step 4: Check node pool operations `[Low Risk]`195196A MIG that is still changing instances can be the result of a node pool upgrade197or repair:1981991. **Check node pool upgrades**: Follow200 [Check node pool upgrade status](https://docs.cloud.google.com/kubernetes-engine/docs/how-to/upgrading-a-cluster.md.txt)201 to list the cluster operations and find upgrade operations targeting202 `{nodepool_name}` in `[{start_time}, {end_time}]`. Also review203 [Upgrade your node pools](https://docs.cloud.google.com/kubernetes-engine/docs/how-to/create-dynamic-slices.md.txt)204 for dynamic slicing upgrade requirements (required surge settings and205 deleting active super-slices before upgrading).206 - If an upgrade operation is still running, go to Step 5, item 1 or item 3.207 - If an upgrade operation was cancelled, failed, or is incomplete, go to Step208 5, item 4.2092. **Check automated node repairs**: Follow210 [Get information about recent automated repair events](https://docs.cloud.google.com/kubernetes-engine/docs/how-to/node-auto-repair.md.txt)211 to check for repair operations on `{nodepool_name}`.212 - If a node auto-repair operation is running, go to Step 5, item 1.2133. **If no operation or error explains the unstable MIG**: Go to Step 5, item 5214 to contact Cloud Customer Care.215216---217218### Step 5: Resolution `[High Risk]`219220> **Guardrail**: Don't modify GKE-managed instance groups or VMs directly221> through Compute Engine. Manage node pools through GKE.2222231. **Wait for an ongoing update or auto-repair to finish**: If a node pool224 upgrade or automated repair is still making progress, the user can wait until225 the MIG is stable (`status.isStable: true`) by following226 [Check whether a MIG is stable](https://docs.cloud.google.com/compute/docs/instance-groups/getting-info-about-migs.md.txt).2272. **Resolve capacity or quota errors**: When Step 3 shows228 `ZONE_RESOURCE_POOL_EXHAUSTED_WITH_DETAILS` or `QUOTA_EXCEEDED`, rolling back229 an upgrade won't resolve the shortage. Follow230 [Issue: Node pool creation fails due to insufficient resources](https://docs.cloud.google.com/kubernetes-engine/docs/troubleshooting/node-pools.md.txt)231 and the "Insufficient quota to satisfy the TPU request" and "Insufficient TPU232 resources to satisfy the TPU request" sections of233 [Troubleshoot TPUs in GKE](https://docs.cloud.google.com/kubernetes-engine/docs/troubleshooting/tpus.md.txt).2343. **Cancel an in-progress node pool upgrade**: If the user wants to stop an235 upgrade that is still running, follow236 [Cancel a node pool upgrade](https://docs.cloud.google.com/kubernetes-engine/docs/how-to/upgrading-a-cluster.md.txt).237 For dynamic slicing node pools, also check238 [Upgrade your node pools](https://docs.cloud.google.com/kubernetes-engine/docs/how-to/create-dynamic-slices.md.txt)239 for the required surge settings and deleting active super-slices before240 upgrading.2414. **Resume or roll back a cancelled, failed, or incomplete node pool upgrade**:242 - Follow the "Resume a node pool upgrade" or "Roll back a node pool upgrade"243 sections of244 [Manually upgrade a cluster's control plane or node pools](https://docs.cloud.google.com/kubernetes-engine/docs/how-to/upgrading-a-cluster.md.txt).245 As documented there, rollback applies if an in-progress upgrade was246 cancelled, the upgrade failed, or the upgrade is incomplete due to a247 maintenance window timing out (you can't roll back node pools once they248 have been successfully upgraded); alternatively, a cancelled or failed,249 partially completed upgrade can be resumed.250 - Afterwards, verify that the MIG is stable by following251 [Check whether a MIG is stable](https://docs.cloud.google.com/compute/docs/instance-groups/getting-info-about-migs.md.txt).2525. **Escalate unexplained MIG instability**: If no node pool upgrade, repair253 operation, or capacity/quota error explains why `{mig_name}` remains254 unstable, direct the user to Cloud Customer Care by following255 [Get support](https://docs.cloud.google.com/kubernetes-engine/docs/getting-support.md.txt).2566. **Verify node pool and slice recovery**: After the instance group is stable257 (`status.isStable: true`), follow the "Monitor the status of the Slice custom258 resource" and "Create a Slice custom resource" sections of259 [Use dynamic slicing with a custom scheduler](https://docs.cloud.google.com/kubernetes-engine/docs/how-to/create-dynamic-slices.md.txt)260 to verify that the `Slice` reaches the `ACTIVE` state. If the `Slice` stays261 in `SliceCreationFailed` after the MIG stabilizes, check whether the retry262 annotation documented in "Create a Slice custom resource" is set on the263 `Slice`.