fix(llm): route OpenClaw through LLMSpy gateway#139
Draft
bussyjd wants to merge 4 commits intofeat/appkit-nanobotfrom
Draft
fix(llm): route OpenClaw through LLMSpy gateway#139bussyjd wants to merge 4 commits intofeat/appkit-nanobotfrom
bussyjd wants to merge 4 commits intofeat/appkit-nanobotfrom
Conversation
LLMSpy was scaled to 0 due to a missing providers.json in the deployed init container. The source llm.yaml already had the fix but the cluster was stale. This commit also updates OpenClaw's default Ollama baseUrl to point to LLMSpy (port 8000) instead of directly to Ollama, and fixes the default model name from "glm4" to "glm-4.7-flash".
Revert Ollama limits to cpu: 2000m, memory: 4Gi to match the previous defaults and avoid over-provisioning on local dev clusters.
This reverts commit bae5bd9.
Replace the in-cluster Ollama Deployment/PVC/Service with an
ExternalName Service that routes ollama.llm.svc.cluster.local to the
host machine. LLMSpy and all consumers use the stable cluster-internal
DNS name; the ExternalName target is resolved at stack init time:
k3d → host.k3d.internal
k3s → node gateway IP (future)
CopyDefaults now accepts a replacements map so stack init can resolve
{{OLLAMA_HOST}} (and future placeholders) in infrastructure templates.
bussyjd
changed the base branch from
feat/appkit-nanobot
to
integration-okr-1
bussyjd
changed the base branch from
integration-okr-1
to
feat/appkit-nanobot
Contributor
|
Is it just that you had ollama on your host previously? or is the a reason it can't go in the stack? imo without good reason i see this as a con for the efficacy of the stack if it has to be on host. I'd much rather we put it in the stack and as an easy opt in for like |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
llms.jsonandproviders.json(root cause of the crash / scale-to-0)glm4toglm-4.7-flashin the LLMSpy ConfigMapllmspy.llm.svc:8000) instead of directly to Ollama (ollama.llm.svc:11434)agentModelinto the Helm chart's generatedopenclaw.jsonTest plan
llms.jsonandproviders.jsonpresent in/home/llms/.llms//v1/chat/completionsreturns valid OpenAI-format JSON from Ollamaopenclaw.json) showsbaseUrlpointing to LLMSpy