Conversation
Bump porter-sandbox to 0.1.58 and type sandbox creation with the SDK's own SandboxSpec, so resources, snapshot_id, env_groups and internal networking are no longer hidden behind a hand-declared shape. The client is now a thin adapter over the SDK instead of a Proxy cast, with the same error-detail enrichment. Move file traffic under $HOME to Porter's volume file API: one atomic PUT per write and 8 MiB ranged GETs per read instead of 64 KiB base64 exec chunks. Import archives land in the workspace through the same path, and the export archive is now built under $HOME/.qm-export (pruned from the home export) so it reads back through the volume too. Scratch bodies, paths outside the volume, and clusters without the sandbox-volumes mount keep the exec path; a client error from the files API marks that volume as exec-only for the process, a timeout falls back for that call only. Carry files outside $HOME across body rotation with Porter filesystem snapshots. A scope's body is snapshotted before restart, egress rotation and idle reaping, and refreshed during the idle sweep when the scope has been active since; the next body starts from snapshot_id with image left empty as the docs require. One snapshot is kept per scope, superseded and destroyed ones are deleted, a snapshot from a different image is ignored, and one Porter can no longer start from is forgotten and the body comes from the image. computerStatus reports it as recovery strategy provider_snapshot. Add idle reaping: reapDeepIdle retires bodies whose scope has been inactive past the deep-idle window and has no running process session, leaving the volume and snapshot in place, so ttl_seconds is only the cost cap. The scope record (body id, last activity, snapshot pointer) lives in the durable porter_sandbox_bodies map and joins the legacy sandbox bindings. Honour Pagination.has_next_page when listing, find the live body through the home volume's attached_to before falling back to a tag scan, cache body name-to-id resolution instead of a lookup plus get per exec, raise the SDK's 30 second client timeout to 120 seconds because snapshots.create resolves only when the capture completes, and surface the failed body's log tail and the lifetime cap in computerStatus. Exec stays unbounded by the SDK and is never retried, as documented. PORTER_CPUS and PORTER_MEMORY_MB set resources.cpu and resources.memory and fill the advertised profile spec.
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
1 participant
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.
Aligns the Porter backend with the Porter sandbox documentation and SDK 0.1.58.
Docs: https://docs.porter.run/sandboxes/volumes https://docs.porter.run/sandboxes/sdk/typescript/reference https://docs.porter.run/sandboxes/cli
Needs live QA before merge.