feat: configure shared ChunkDB capacity in node deployments - #74
Merged
Merged
Conversation
tianyuzhou95
marked this pull request as ready for review
September 20, 2026 03:41
hhhhsdxxxx
approved these changes
Sep 22, 2026
Add optional capacity settings for standalone, Helm, and both cloud Terraform modules, including guided profile generation. Preserve the existing distill-fs default when no override is supplied and validate input syntax before rendering the node configuration. Document consistent node-wide configuration, supported size units, and the cache replacement required when changing capacity. Custom templates retain a marker so overrides cannot silently miss the image configuration. Signed-off-by: Tianyu Zhou <albert.zty@antgroup.com>
Consume the published stripped static release with configurable ChunkDB capacity. Pin its release URL and verified archive SHA-256 together. Verify provenance, binary hash, static linking, and stripped symbols, exercise installer rejection cases, and validate custom-capacity mounts, maintenance, and recovery with sandboxd on the Linux test node. Signed-off-by: Tianyu Zhou <albert.zty@antgroup.com>
tianyuzhou95
force-pushed
the
codex/chunk-db-capacity
branch
from
September 22, 2026 02:49
e6953f7 to
374c919
Compare
tianyuzhou95
commented
Sep 22, 2026
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Expose shared ChunkDB capacity through standalone
AKERNEL_CHUNK_DB_SIZE, Helmcore.node.config.sandboxd.chunkDbSize, Terraformchunk_db_sizefor Alibaba/Huawei Cloud, and guidedmake config CHUNK_DB_SIZE=64GiB. Omitted settings preserve the current Linux 100 GiB default. Input is validated before rendering the shared sandboxd[plugin.image].chunk_db_sizesetting in every deployment path; changing the Helm setting also changes the node Pod configuration checksum.Pin the published distill-fs v0.1.2 archive and verified SHA-256. The release contains the optimized, stripped, static Linux/amd64 executable.
The current main already pins sandboxd
e67b258, which descends from the merged ChunkDB support in inclusionAI/sandboxd#59 and additionally contains the gVisor runtime-bundle update. No separate gitlink commit is needed after rebasing.Validation:
make deploy-script-check, Terraform formatting, Helm lint, and deployment rendering checks passed. The distill-fs release archive and its static binary contract were verified, and sandboxd CI passed its real FUSE integration against that published binary.