Skip to content
Closed
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion kubernetes/apps/ai/llmkube/models/qwen36-27b-sglang.yaml
Original file line number Diff line number Diff line change
Expand Up @@ -77,7 +77,7 @@ spec:
runtime: sglang
# Operator default binds :: (IPv6-only on this IPv4 cluster); override to v4.
bindAddress: 0.0.0.0
image: ghcr.io/tanguille/sglang-rdna4:v0.5.16-gfx1201@sha256:615cb4265400f5468da4c32221f0086df075b126ce8dbfa78d998e844a5e424d
image: ghcr.io/tanguille/sglang-rdna4:v0.5.16-gfx1201@sha256:51ececd579cec3785357c07e318c2badf9e333ea07c7fd47a0e19fcb4a7c2b24

Copy link
Copy Markdown

Choose a reason for hiding this comment

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

🗄️ Data Integrity & Integration | 🟠 Major | ⚡ Quick win

Bump the HiCache directory suffix with the new image digest.

The manifest explicitly requires this suffix to change whenever the image digest changes. Leaving sglang-v0.5.16_awq-f541031d unchanged allows KV pages produced by the previous image to be reused by the new image, potentially serving stale or incompatible cache data.

Also applies to: 124-127

🤖 Prompt for AI Agents
Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@kubernetes/apps/ai/llmkube/models/qwen36-27b-sglang.yaml` at line 80, The
HiCache directory suffix must match the updated image digest. Update the suffix
in the manifest’s HiCache configuration, including the corresponding entries
around the additional referenced locations, from sglang-v0.5.16_awq-f541031d to
the suffix derived from digest
51ececd579cec3785357c07e318c2badf9e333ea07c7fd47a0e19fcb4a7c2b24, so caches from
the previous image are not reused.

containerPort: 30000
endpoint:
# v0.9.8 maps endpoint.port to both Service port and targetPort; use the
Expand Down