Improve progress reporting when layers are locally present #6362
Conversation
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
With containerd snapshotter, some layers skip the "Downloading" phase
entirely and go directly from "Pulling fs layer" to "Download complete".
These layers only receive a placeholder extra={current: 1, total: 1}
instead of actual size information.
The previous fix in #6357 still required ALL layers to have extra before
calculating progress, which caused the UI to jump from 0% to 70%+ when
the first real layer reported after cached layers completed.
This change improves progress calculation by:
- Separating "real" layers (with actual size > 1 byte) from "placeholder"
layers (cached/small layers with size = 1)
- Calculating weighted progress only from real layers
- Not reporting progress until at least one real layer has size info
- Correctly tracking stage when placeholder layers are still extracting
Tested with real-world pull events from a Home Assistant Core update
that had 12 layers where 5 skipped downloading entirely.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <[email protected]>
When pulling Docker images with many layers, tiny layers that complete immediately would show inflated progress (e.g., 70%) even when most layers haven't started reporting yet. This made the UI jump to 70% quickly, then appear stuck during actual download. The fix scales progress by the fraction of layers that have reported: - If 2/25 layers report at 70%, progress shows ~5.6% instead of 70% - As more layers report, progress increases proportionally - When all layers have reported, no scaling is applied This ensures progress accurately reflects overall download status rather than being dominated by a few tiny layers that complete first. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-Authored-By: Claude <[email protected]>
Member
Author
|
Closing, in testing progress was still not quite satisfactory. PR #6371 replaces this with a simpler approach. |
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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.
Proposed change
Improves Docker pull progress reporting when using containerd snapshotter and some layers are already locally present (cached).
With containerd snapshotter, some layers skip the "Downloading" phase entirely and go directly from "Pulling fs layer" to "Download complete". These layers only receive a placeholder
extra={current: 1, total: 1}instead of actual size information.The previous fix in #6357 still required ALL layers to have
extrabefore calculating progress, which caused the UI to jump from 0% to 70%+ when the first real layer reported after cached layers completed.This change improves progress calculation by:
Before: Progress jumps 0% → 70% → 100%
After: Progress shows smooth 0% → 1% → 2% → ... → 100%
Type of change
Additional information
Checklist
ruff format supervisor tests)If API endpoints or add-on configuration are added/changed: