Description
Describe the bug
I am reporting a series of critical deployment blockers encountered while installing NemoClaw v0.1.0 on an Ubuntu 24.04 server within a restricted corporate network (using a local HTTP proxy).
Despite multiple workarounds, the system fails primarily due to the nested K3s environment ignoring host configurations and CLI command inconsistencies.
System Environment
- OS: Ubuntu 24.04 LTS (Noble Numbat)
- Docker Version: 29.3.0
- NemoClaw CLI: v0.1.0 (OpenShell 0.0.14)
- Network: Restricted Intranet (Global Proxy configured via Docker systemd)
Critical Issues Encountered
1. Nested Networking Isolation (The "I/O Timeout" Loop)
The host Docker daemon has a working proxy, but the nested K3s cluster inside the openshell-cluster-nemoclaw container completely ignores host proxy settings. It attempts to pull rancher/mirrored-pause:3.6 directly from registry-1.docker.io, resulting in an infinite i/o timeout.
2. Self-Destructive Onboarding Logic
nemoclaw onboard unconditionally destroys the gateway container before starting. This makes "air-gap" workarounds impossible because manually imported images (docker save/load or ctr images import) are wiped before the K3s initialization begins.
3. Cgroup v2 Permission Denied
On Ubuntu 24.04, the internal K3s fails with: forbidden: User system:node... cannot get resource "pods". This appears to be a known compatibility issue with Cgroup v2 that setup-spark tries to fix but fails due to mandatory phone verification for the NVIDIA API Key.
4. CLI Parsing Bugs
The CLI (v0.0.14) does not recognize its own documented commands or arguments:
nemoclaw sandbox returns Unknown command.
openshell sandbox create --file returns unexpected argument '--file' found.
openshell shell returns unrecognized subcommand 'shell'.
Steps to Reproduce
- Install NemoClaw on Ubuntu 24.04 in a network requiring a proxy.
- Run
sudo nemoclaw onboard.
- Observe the
Initializing environment hang followed by K8s namespace timeout.
Suggested Fixes
- Proxy Passthrough: Allow the nested K3s to inherit
HTTP_PROXY from the host.
- Persistence: Do not destroy existing gateway containers during
onboard if they contain pre-loaded images.
- CLI Refactor: Ensure the CLI arguments match the documentation provided in the help menus.
Reproduction Steps
- Install NemoClaw v0.1.0 on Ubuntu 24.04.5 LTS in a restricted network environment.
- Configure HTTP_PROXY/HTTPS_PROXY for the host Docker daemon.
- Run
sudo /usr/bin/nemoclaw onboard.
- The process successfully pulls the gateway image but fails at
[2/7] Starting OpenShell gateway.
- Error logs show the nested K3s cluster ignores the host proxy and attempts to pull
rancher/mirrored-pause:3.6 directly, resulting in an i/o timeout.
- Also, documented commands like
nemoclaw sandbox or openshell shell return Unknown command or unrecognized subcommand.
Environment
- OS: Ubuntu 24.04.5 LTS (Noble Numbat)
- Node.js: v20.20.1
- Docker: 29.3.0
- NemoClaw: v0.1.0 (OpenShell CLI 0.0.14)
- CPU: x86_64
- Network: Restricted Intranet with Local Proxy (Port 39949)
Debug Output
Error: × K8s namespace not ready
╰─▶ timed out waiting for namespace 'openshell' to exist: Error from server (NotFound): namespaces "openshell" not found
container logs:
E0324 09:22:29.988161 127 kuberuntime_manager.go:1558] "CreatePodSandbox for pod failed" err="rpc error: code = DeadlineExceeded desc = failed to start sandbox \"24908ae96f473cd0ee718ba48ca436048414f01592f06110067525a67885da91\": failed to get sandbox image \"rancher/
mirrored-pause:3.6\": failed to pull image \"rancher/mirrored-pause:3.6\": failed to pull and unpack image \"docker.io/rancher/mirrored-pause:3.6\": failed to resolve reference \"docker.io/rancher/mirrored-pause:3.6\": failed to do request: Head \"https://registry-1.docker.io/
v2/rancher/mirrored-pause/manifests/3.6\": dial tcp 54.234.18.200:443: i/o timeout" pod="kube-system/coredns-7566b5ff58-zmq5p"
E0324 09:22:29.988269 127 pod_workers.go:1324] "Error syncing pod, skipping" err="failed to \"CreatePodSandbox\" for \"coredns-7566b5ff58-zmq5p_kube-system(3469b982-48b4-4969-90ab-f4d0c03c425b)\" with CreatePodSandboxError: \"Failed to create sandbox for pod \\\"coredns-
7566b5ff58-zmq5p_kube-system(3469b982-48b4-4969-90ab-f4d0c03c425b)\\\": rpc error: code = DeadlineExceeded desc = failed to start sandbox \\\"24908ae96f473cd0ee718ba48ca436048414f01592f06110067525a67885da91\\\": failed to get sandbox image \\\"rancher/mirrored-pause:3.6\\\":
failed to pull image \\\"rancher/mirrored-pause:3.6\\\": failed to pull and unpack image \\\"docker.io/rancher/mirrored-pause:3.6\\\": failed to resolve reference \\\"docker.io/rancher/mirrored-pause:3.6\\\": failed to do request: Head \\\"https://registry-1.docker.io/v2/
rancher/mirrored-pause/manifests/3.6\\\": dial tcp 54.234.18.200:443: i/o timeout\"" pod="kube-system/coredns-7566b5ff58-zmq5p" podUID="3469b982-48b4-4969-90ab-f4d0c03c425b"
E0324 09:22:30.104581 127 log.go:32] "RunPodSandbox from runtime service failed" err="rpc error: code = DeadlineExceeded desc = failed to start sandbox \"75f06196b5eecc3af70ff3ec5725f813ff2ae095d01d7d53d87ae27fd4baabda\": failed to get sandbox image \"rancher/mirrored-
pause:3.6\": failed to pull image \"rancher/mirrored-pause:3.6\": failed to pull and unpack image \"docker.io/rancher/mirrored-pause:3.6\": failed to resolve reference \"docker.io/rancher/mirrored-pause:3.6\": failed to do request: Head \"https://registry-1.docker.io/v2/
rancher/mirrored-pause/manifests/3.6\": dial tcp 54.234.18.200:443: i/o timeout"
E0324 09:22:30.104631 127 kuberuntime_sandbox.go:71] "Failed to create sandbox for pod" err="rpc error: code = DeadlineExceeded desc = failed to start sandbox \"75f06196b5eecc3af70ff3ec5725f813ff2ae095d01d7d53d87ae27fd4baabda\": failed to get sandbox image \"rancher/
mirrored-pause:3.6\": failed to pull image \"rancher/mirrored-pause:3.6\": failed to pull and unpack image \"docker.io/rancher/mirrored-pause:3.6\": failed to resolve reference \"docker.io/rancher/mirrored-pause:3.6\": failed to do request: Head \"https://registry-1.docker.io/
v2/rancher/mirrored-pause/manifests/3.6\": dial tcp 54.234.18.200:443: i/o timeout" pod="kube-system/metrics-server-786d997795-c44dq"
E0324 09:22:30.104657 127 kuberuntime_manager.go:1558] "CreatePodSandbox for pod failed" err="rpc error: code = DeadlineExceeded desc = failed to start sandbox \"75f06196b5eecc3af70ff3ec5725f813ff2ae095d01d7d53d87ae27fd4baabda\": failed to get sandbox image \"rancher/
mirrored-pause:3.6\": failed to pull image \"rancher/mirrored-pause:3.6\": failed to pull and unpack image \"docker.io/rancher/mirrored-pause:3.6\": failed to resolve reference \"docker.io/rancher/mirrored-pause:3.6\": failed to do request: Head \"https://registry-1.docker.io/
v2/rancher/mirrored-pause/manifests/3.6\": dial tcp 54.234.18.200:443: i/o timeout" pod="kube-system/metrics-server-786d997795-c44dq"
E0324 09:22:30.104730 127 pod_workers.go:1324] "Error syncing pod, skipping" err="failed to \"CreatePodSandbox\" for \"metrics-server-786d997795-c44dq_kube-system(28e0d2c3-9594-4a42-9e3f-342ed4288266)\" with CreatePodSandboxError: \"Failed to create sandbox for pod \\
\"metrics-server-786d997795-c44dq_kube-system(28e0d2c3-9594-4a42-9e3f-342ed4288266)\\\": rpc error: code = DeadlineExceeded desc = failed to start sandbox \\\"75f06196b5eecc3af70ff3ec5725f813ff2ae095d01d7d53d87ae27fd4baabda\\\": failed to get sandbox image \\\"rancher/mirrored-
pause:3.6\\\": failed to pull image \\\"rancher/mirrored-pause:3.6\\\": failed to pull and unpack image \\\"docker.io/rancher/mirrored-pause:3.6\\\": failed to resolve reference \\\"docker.io/rancher/mirrored-pause:3.6\\\": failed to do request: Head \\\"https://registry-
1.docker.io/v2/rancher/mirrored-pause/manifests/3.6\\\": dial tcp 54.234.18.200:443: i/o timeout\"" pod="kube-system/metrics-server-786d997795-c44dq" podUID="28e0d2c3-9594-4a42-9e3f-342ed4288266"
E0324 09:22:33.829219 127 handler_proxy.go:143] error resolving kube-system/metrics-server: no endpoints available for service "metrics-server"
E0324 09:22:42.738127 127 resource_quota_controller.go:460] "Error during resource discovery" err="unable to retrieve the complete list of server APIs: metrics.k8s.io/v1beta1: stale GroupVersion discovery: metrics.k8s.io/v1beta1"
I0324 09:22:42.818308 127 garbagecollector.go:792] "failed to discover some groups" groups="map[\"metrics.k8s.io/v1beta1\":\"stale GroupVersion discovery: metrics.k8s.io/v1beta1\"]"
W0324 09:22:44.142825 127 handler_proxy.go:99] no RequestInfo found in the context
E0324 09:22:44.142950 127 controller.go:102] "Unhandled Error" err=<
loading OpenAPI spec for "v1beta1.metrics.k8s.io" failed with: failed to download v1beta1.metrics.k8s.io: failed to retrieve openAPI spec, http error: ResponseCode: 503, Body: service unavailable
, Header: map[Content-Type:[text/plain; charset=utf-8] X-Content-Type-Options:[nosniff]]
>
I0324 09:22:44.142976 127 controller.go:109] OpenAPI AggregationController: action for item v1beta1.metrics.k8s.io: Rate Limited Requeue.
W0324 09:22:44.143919 127 handler_proxy.go:99] no RequestInfo found in the context
E0324 09:22:44.143980 127 controller.go:113] "Unhandled Error" err="loading OpenAPI spec for \"v1beta1.metrics.k8s.io\" failed with: Error, could not get list of group versions for APIService"
I0324 09:22:44.144008 127 controller.go:126] OpenAPI AggregationController: action for item v1beta1.metrics.k8s.io: Rate Limited Requeue.
E0324 09:23:09.737905 127 log.go:32] "RunPodSandbox from runtime service failed" err="rpc error: code = DeadlineExceeded desc = failed to start sandbox \"83d003e607560678730a073a23594917d7d9cdee09d4c550b0f85e1879154df6\": failed to get sandbox image \"rancher/mirrored-
pause:3.6\": failed to pull image \"rancher/mirrored-pause:3.6\": failed to pull and unpack image \"docker.io/rancher/mirrored-pause:3.6\": failed to resolve reference \"docker.io/rancher/mirrored-pause:3.6\": failed to do request: Head \"https://registry-1.docker.io/v2/
rancher/mirrored-pause/manifests/3.6\": dial tcp 54.234.18.200:443: i/o timeout"
E0324 09:23:09.737991 127 kuberuntime_sandbox.go:71] "Failed to create sandbox for pod" err="rpc error: code = DeadlineExceeded desc = failed to start sandbox \"83d003e607560678730a073a23594917d7d9cdee09d4c550b0f85e1879154df6\": failed to get sandbox image \"rancher/
mirrored-pause:3.6\": failed to pull image \"rancher/mirrored-pause:3.6\": failed to pull and unpack image \"docker.io/rancher/mirrored-pause:3.6\": failed to resolve reference \"docker.io/rancher/mirrored-pause:3.6\": failed to do request: Head \"https://registry-1.docker.io/
v2/rancher/mirrored-pause/manifests/3.6\": dial tcp 54.234.18.200:443: i/o timeout" pod="kube-system/local-path-provisioner-6bc6568469-ln6xj"
E0324 09:23:09.738042 127 kuberuntime_manager.go:1558] "CreatePodSandbox for pod failed" err="rpc error: code = DeadlineExceeded desc = failed to start sandbox \"83d003e607560678730a073a23594917d7d9cdee09d4c550b0f85e1879154df6\": failed to get sandbox image \"rancher/
mirrored-pause:3.6\": failed to pull image \"rancher/mirrored-pause:3.6\": failed to pull and unpack image \"docker.io/rancher/mirrored-pause:3.6\": failed to resolve reference \"docker.io/rancher/mirrored-pause:3.6\": failed to do request: Head \"https://registry-1.docker.io/
v2/rancher/mirrored-pause/manifests/3.6\": dial tcp 54.234.18.200:443: i/o timeout" pod="kube-system/local-path-provisioner-6bc6568469-ln6xj"
E0324 09:23:09.738155 127 pod_workers.go:1324] "Error syncing pod, skipping" err="failed to \"CreatePodSandbox\" for \"local-path-provisioner-6bc6568469-ln6xj_kube-system(8ff1c18e-93dd-4078-9929-982b900e73d7)\" with CreatePodSandboxError: \"Failed to create sandbox for
pod \\\"local-path-provisioner-6bc6568469-ln6xj_kube-system(8ff1c18e-93dd-4078-9929-982b900e73d7)\\\": rpc error: code = DeadlineExceeded desc = failed to start sandbox \\\"83d003e607560678730a073a23594917d7d9cdee09d4c550b0f85e1879154df6\\\": failed to get sandbox image \\
\"rancher/mirrored-pause:3.6\\\": failed to pull image \\\"rancher/mirrored-pause:3.6\\\": failed to pull and unpack image \\\"docker.io/rancher/mirrored-pause:3.6\\\": failed to resolve reference \\\"docker.io/rancher/mirrored-pause:3.6\\\": failed to do request: Head \\
\"https://registry-1.docker.io/v2/rancher/mirrored-pause/manifests/3.6\\\": dial tcp 54.234.18.200:443: i/o timeout\"" pod="kube-system/local-path-provisioner-6bc6568469-ln6xj" podUID="8ff1c18e-93dd-4078-9929-982b900e73d7"
E0324 09:23:12.180068 127 log.go:32] "RunPodSandbox from runtime service failed" err="rpc error: code = Unknown desc = failed to start sandbox \"d4b5cf149ce51e968ec79c3c1f2e2d602ad6be5f2673cd1b2096818bd5acef32\": failed to get sandbox image \"rancher/mirrored-pause:3.6\":
failed to pull image \"rancher/mirrored-pause:3.6\": failed to pull and unpack image \"docker.io/rancher/mirrored-pause:3.6\": failed to resolve reference \"docker.io/rancher/mirrored-pause:3.6\": failed to do request: Head \"https://registry-1.docker.io/v2/rancher/mirrored-
pause/manifests/3.6\": dial tcp 162.125.18.133:443: i/o timeout"
E0324 09:23:12.180163 127 kuberuntime_sandbox.go:71] "Failed to create sandbox for pod" err="rpc error: code = Unknown desc = failed to start sandbox \"d4b5cf149ce51e968ec79c3c1f2e2d602ad6be5f2673cd1b2096818bd5acef32\": failed to get sandbox image \"rancher/mirrored-
pause:3.6\": failed to pull image \"rancher/mirrored-pause:3.6\": failed to pull and unpack image \"docker.io/rancher/mirrored-pause:3.6\": failed to resolve reference \"docker.io/rancher/mirrored-pause:3.6\": failed to do request: Head \"https://registry-1.docker.io/v2/
rancher/mirrored-pause/manifests/3.6\": dial tcp 162.125.18.133:443: i/o timeout" pod="kube-system/helm-install-openshell-74gwd"
E0324 09:23:12.180222 127 kuberuntime_manager.go:1558] "CreatePodSandbox for pod failed" err="rpc error: code = Unknown desc = failed to start sandbox \"d4b5cf149ce51e968ec79c3c1f2e2d602ad6be5f2673cd1b2096818bd5acef32\": failed to get sandbox image \"rancher/mirrored-
pause:3.6\": failed to pull image \"rancher/mirrored-pause:3.6\": failed to pull and unpack image \"docker.io/rancher/mirrored-pause:3.6\": failed to resolve reference \"docker.io/rancher/mirrored-pause:3.6\": failed to do request: Head \"https://registry-1.docker.io/v2/
rancher/mirrored-pause/manifests/3.6\": dial tcp 162.125.18.133:443: i/o timeout" pod="kube-system/helm-install-openshell-74gwd"
E0324 09:23:12.180348 127 pod_workers.go:1324] "Error syncing pod, skipping" err="failed to \"CreatePodSandbox\" for \"helm-install-openshell-74gwd_kube-system(2ac6b774-6551-4dff-a34b-b04aca273c10)\" with CreatePodSandboxError: \"Failed to create sandbox for pod \\\"helm-
install-openshell-74gwd_kube-system(2ac6b774-6551-4dff-a34b-b04aca273c10)\\\": rpc error: code = Unknown desc = failed to start sandbox \\\"d4b5cf149ce51e968ec79c3c1f2e2d602ad6be5f2673cd1b2096818bd5acef32\\\": failed to get sandbox image \\\"rancher/mirrored-pause:3.6\\\":
failed to pull image \\\"rancher/mirrored-pause:3.6\\\": failed to pull and unpack image \\\"docker.io/rancher/mirrored-pause:3.6\\\": failed to resolve reference \\\"docker.io/rancher/mirrored-pause:3.6\\\": failed to do request: Head \\\"https://registry-1.docker.io/v2/
rancher/mirrored-pause/manifests/3.6\\\": dial tcp 162.125.18.133:443: i/o timeout\"" pod="kube-system/helm-install-openshell-74gwd" podUID="2ac6b774-6551-4dff-a34b-b04aca273c10"
E0324 09:23:12.263299 127 log.go:32] "RunPodSandbox from runtime service failed" err="rpc error: code = DeadlineExceeded desc = failed to start sandbox \"440d80ad2d52cb938df732a85f7c595987cb5a06f2d8e681781073fcbb2974eb\": failed to get sandbox image \"rancher/mirrored-
pause:3.6\": failed to pull image \"rancher/mirrored-pause:3.6\": failed to pull and unpack image \"docker.io/rancher/mirrored-pause:3.6\": failed to resolve reference \"docker.io/rancher/mirrored-pause:3.6\": failed to do request: Head \"https://registry-1.docker.io/v2/
rancher/mirrored-pause/manifests/3.6\": dial tcp 54.234.18.200:443: i/o timeout"
E0324 09:23:12.263391 127 kuberuntime_sandbox.go:71] "Failed to create sandbox for pod" err="rpc error: code = DeadlineExceeded desc = failed to start sandbox \"440d80ad2d52cb938df732a85f7c595987cb5a06f2d8e681781073fcbb2974eb\": failed to get sandbox image \"rancher/
mirrored-pause:3.6\": failed to pull image \"rancher/mirrored-pause:3.6\": failed to pull and unpack image \"docker.io/rancher/mirrored-pause:3.6\": failed to resolve reference \"docker.io/rancher/mirrored-pause:3.6\": failed to do request: Head \"https://registry-1.docker.io/
v2/rancher/mirrored-pause/manifests/3.6\": dial tcp 54.234.18.200:443: i/o timeout" pod="agent-sandbox-system/agent-sandbox-controller-0"
E0324 09:23:12.263457 127 kuberuntime_manager.go:1558] "CreatePodSandbox for pod failed" err="rpc error: code = DeadlineExceeded desc = failed to start sandbox \"440d80ad2d52cb938df732a85f7c595987cb5a06f2d8e681781073fcbb2974eb\": failed to get sandbox image \"rancher/
mirrored-pause:3.6\": failed to pull image \"rancher/mirrored-pause:3.6\": failed to pull and unpack image \"docker.io/rancher/mirrored-pause:3.6\": failed to resolve reference \"docker.io/rancher/mirrored-pause:3.6\": failed to do request: Head \"https://registry-1.docker.io/
v2/rancher/mirrored-pause/manifests/3.6\": dial tcp 54.234.18.200:443: i/o timeout" pod="agent-sandbox-system/agent-sandbox-controller-0"
E0324 09:23:12.263580 127 pod_workers.go:1324] "Error syncing pod, skipping" err="failed to \"CreatePodSandbox\" for \"agent-sandbox-controller-0_agent-sandbox-system(7f72633d-d333-4d82-a715-94eff50ed115)\" with CreatePodSandboxError: \"Failed to create sandbox for pod \\
\"agent-sandbox-controller-0_agent-sandbox-system(7f72633d-d333-4d82-a715-94eff50ed115)\\\": rpc error: code = DeadlineExceeded desc = failed to start sandbox \\\"440d80ad2d52cb938df732a85f7c595987cb5a06f2d8e681781073fcbb2974eb\\\": failed to get sandbox image \\\"rancher/
mirrored-pause:3.6\\\": failed to pull image \\\"rancher/mirrored-pause:3.6\\\": failed to pull and unpack image \\\"docker.io/rancher/mirrored-pause:3.6\\\": failed to resolve reference \\\"docker.io/rancher/mirrored-pause:3.6\\\": failed to do request: Head \\\"https://
registry-1.docker.io/v2/rancher/mirrored-pause/manifests/3.6\\\": dial tcp 54.234.18.200:443: i/o timeout\"" pod="agent-sandbox-system/agent-sandbox-controller-0" podUID="7f72633d-d333-4d82-a715-94eff50ed115"
E0324 09:23:12.705070 127 log.go:32] "RunPodSandbox from runtime service failed" err="rpc error: code = DeadlineExceeded desc = failed to start sandbox \"b8c07f4285c6867f0b57213a9dc1997a090e4732301e3919ad01bf420759cd50\": failed to get sandbox image \"rancher/mirrored-
pause:3.6\": failed to pull image \"rancher/mirrored-pause:3.6\": failed to pull and unpack image \"docker.io/rancher/mirrored-pause:3.6\": failed to resolve reference \"docker.io/rancher/mirrored-pause:3.6\": failed to do request: Head \"https://registry-1.docker.io/v2/
rancher/mirrored-pause/manifests/3.6\": dial tcp 162.125.18.133:443: i/o timeout"
E0324 09:23:12.705129 127 kuberuntime_sandbox.go:71] "Failed to create sandbox for pod" err="rpc error: code = DeadlineExceeded desc = failed to start sandbox \"b8c07f4285c6867f0b57213a9dc1997a090e4732301e3919ad01bf420759cd50\": failed to get sandbox image \"rancher/
mirrored-pause:3.6\": failed to pull image \"rancher/mirrored-pause:3.6\": failed to pull and unpack image \"docker.io/rancher/mirrored-pause:3.6\": failed to resolve reference \"docker.io/rancher/mirrored-pause:3.6\": failed to do request: Head \"https://registry-1.docker.io/
v2/rancher/mirrored-pause/manifests/3.6\": dial tcp 162.125.18.133:443: i/o timeout" pod="kube-system/coredns-7566b5ff58-zmq5p"
E0324 09:23:12.705151 127 kuberuntime_manager.go:1558] "CreatePodSandbox for pod failed" err="rpc error: code = DeadlineExceeded desc = failed to start sandbox \"b8c07f4285c6867f0b57213a9dc1997a090e4732301e3919ad01bf420759cd50\": failed to get sandbox image \"rancher/
mirrored-pause:3.6\": failed to pull image \"rancher/mirrored-pause:3.6\": failed to pull and unpack image \"docker.io/rancher/mirrored-pause:3.6\": failed to resolve reference \"docker.io/rancher/mirrored-pause:3.6\": failed to do request: Head \"https://registry-1.docker.io/
v2/rancher/mirrored-pause/manifests/3.6\": dial tcp 162.125.18.133:443: i/o timeout" pod="kube-system/coredns-7566b5ff58-zmq5p"
E0324 09:23:12.705221 127 pod_workers.go:1324] "Error syncing pod, skipping" err="failed to \"CreatePodSandbox\" for \"coredns-7566b5ff58-zmq5p_kube-system(3469b982-48b4-4969-90ab-f4d0c03c425b)\" with CreatePodSandboxError: \"Failed to create sandbox for pod \\\"coredns-
7566b5ff58-zmq5p_kube-system(3469b982-48b4-4969-90ab-f4d0c03c425b)\\\": rpc error: code = DeadlineExceeded desc = failed to start sandbox \\\"b8c07f4285c6867f0b57213a9dc1997a090e4732301e3919ad01bf420759cd50\\\": failed to get sandbox image \\\"rancher/mirrored-pause:3.6\\\":
failed to pull image \\\"rancher/mirrored-pause:3.6\\\": failed to pull and unpack image \\\"docker.io/rancher/mirrored-pause:3.6\\\": failed to resolve reference \\\"docker.io/rancher/mirrored-pause:3.6\\\": failed to do request: Head \\\"https://registry-1.docker.io/v2/
rancher/mirrored-pause/manifests/3.6\\\": dial tcp 162.125.18.133:443: i/o timeout\"" pod="kube-system/coredns-7566b5ff58-zmq5p" podUID="3469b982-48b4-4969-90ab-f4d0c03c425b"
E0324 09:23:12.744807 127 resource_quota_controller.go:460] "Error during resource discovery" err="unable to retrieve the complete list of server APIs: metrics.k8s.io/v1beta1: stale GroupVersion discovery: metrics.k8s.io/v1beta1"
I0324 09:23:12.829412 127 garbagecollector.go:792] "failed to discover some groups" groups="map[\"metrics.k8s.io/v1beta1\":\"stale GroupVersion discovery: metrics.k8s.io/v1beta1\"]"
E0324 09:23:15.705699 127 log.go:32] "RunPodSandbox from runtime service failed" err="rpc error: code = DeadlineExceeded desc = failed to start sandbox \"6375472c6b9bcbdb8a4ee5ef2dc5ea50c20ede7a81df997b55a9a2d41d11db27\": failed to get sandbox image \"rancher/mirrored-
pause:3.6\": failed to pull image \"rancher/mirrored-pause:3.6\": failed to pull and unpack image \"docker.io/rancher/mirrored-pause:3.6\": failed to resolve reference \"docker.io/rancher/mirrored-pause:3.6\": failed to do request: Head \"https://registry-1.docker.io/v2/
rancher/mirrored-pause/manifests/3.6\": dial tcp 59.24.3.173:443: i/o timeout"
E0324 09:23:15.705782 127 kuberuntime_sandbox.go:71] "Failed to create sandbox for pod" err="rpc error: code = DeadlineExceeded desc = failed to start sandbox \"6375472c6b9bcbdb8a4ee5ef2dc5ea50c20ede7a81df997b55a9a2d41d11db27\": failed to get sandbox image \"rancher/
mirrored-pause:3.6\": failed to pull image \"rancher/mirrored-pause:3.6\": failed to pull and unpack image \"docker.io/rancher/mirrored-pause:3.6\": failed to resolve reference \"docker.io/rancher/mirrored-pause:3.6\": failed to do request: Head \"https://registry-1.docker.io/
v2/rancher/mirrored-pause/manifests/3.6\": dial tcp 59.24.3.173:443: i/o timeout" pod="kube-system/metrics-server-786d997795-c44dq"
E0324 09:23:15.705833 127 kuberuntime_manager.go:1558] "CreatePodSandbox for pod failed" err="rpc error: code = DeadlineExceeded desc = failed to start sandbox \"6375472c6b9bcbdb8a4ee5ef2dc5ea50c20ede7a81df997b55a9a2d41d11db27\": failed to get sandbox image \"rancher/
mirrored-pause:3.6\": failed to pull image \"rancher/mirrored-pause:3.6\": failed to pull and unpack image \"docker.io/rancher/mirrored-pause:3.6\": failed to resolve reference \"docker.io/rancher/mirrored-pause:3.6\": failed to do request: Head \"https://registry-1.docker.io/
v2/rancher/mirrored-pause/manifests/3.6\": dial tcp 59.24.3.173:443: i/o timeout" pod="kube-system/metrics-server-786d997795-c44dq"
E0324 09:23:15.705963 127 pod_workers.go:1324] "Error syncing pod, skipping" err="failed to \"CreatePodSandbox\" for \"metrics-server-786d997795-c44dq_kube-system(28e0d2c3-9594-4a42-9e3f-342ed4288266)\" with CreatePodSandboxError: \"Failed to create sandbox for pod \\
\"metrics-server-786d997795-c44dq_kube-system(28e0d2c3-9594-4a42-9e3f-342ed4288266)\\\": rpc error: code = DeadlineExceeded desc = failed to start sandbox \\\"6375472c6b9bcbdb8a4ee5ef2dc5ea50c20ede7a81df997b55a9a2d41d11db27\\\": failed to get sandbox image \\\"rancher/mirrored-
pause:3.6\\\": failed to pull image \\\"rancher/mirrored-pause:3.6\\\": failed to pull and unpack image \\\"docker.io/rancher/mirrored-pause:3.6\\\": failed to resolve reference \\\"docker.io/rancher/mirrored-pause:3.6\\\": failed to do request: Head \\\"https://registry-
1.docker.io/v2/rancher/mirrored-pause/manifests/3.6\\\": dial tcp 59.24.3.173:443: i/o timeout\"" pod="kube-system/metrics-server-786d997795-c44dq" podUID="28e0d2c3-9594-4a42-9e3f-342ed4288266"
Command failed (exit 1): openshell gateway start --name nemoclaw
Logs
Checklist
Description
Describe the bug
I am reporting a series of critical deployment blockers encountered while installing NemoClaw v0.1.0 on an Ubuntu 24.04 server within a restricted corporate network (using a local HTTP proxy).
Despite multiple workarounds, the system fails primarily due to the nested K3s environment ignoring host configurations and CLI command inconsistencies.
System Environment
Critical Issues Encountered
1. Nested Networking Isolation (The "I/O Timeout" Loop)
The host Docker daemon has a working proxy, but the nested K3s cluster inside the
openshell-cluster-nemoclawcontainer completely ignores host proxy settings. It attempts to pullrancher/mirrored-pause:3.6directly fromregistry-1.docker.io, resulting in an infinitei/o timeout.2. Self-Destructive Onboarding Logic
nemoclaw onboardunconditionally destroys the gateway container before starting. This makes "air-gap" workarounds impossible because manually imported images (docker save/loadorctr images import) are wiped before the K3s initialization begins.3. Cgroup v2 Permission Denied
On Ubuntu 24.04, the internal K3s fails with:
forbidden: User system:node... cannot get resource "pods". This appears to be a known compatibility issue with Cgroup v2 thatsetup-sparktries to fix but fails due to mandatory phone verification for the NVIDIA API Key.4. CLI Parsing Bugs
The CLI (v0.0.14) does not recognize its own documented commands or arguments:
nemoclaw sandboxreturnsUnknown command.openshell sandbox create --filereturnsunexpected argument '--file' found.openshell shellreturnsunrecognized subcommand 'shell'.Steps to Reproduce
sudo nemoclaw onboard.Initializing environmenthang followed by K8s namespace timeout.Suggested Fixes
HTTP_PROXYfrom the host.onboardif they contain pre-loaded images.Reproduction Steps
sudo /usr/bin/nemoclaw onboard.[2/7] Starting OpenShell gateway.rancher/mirrored-pause:3.6directly, resulting in ani/o timeout.nemoclaw sandboxoropenshell shellreturnUnknown commandorunrecognized subcommand.Environment
Debug Output
Logs
Checklist