-
Notifications
You must be signed in to change notification settings - Fork 224
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Feature: Blob filter #1595
Feature: Blob filter #1595
Conversation
Should we support filtering for directories for HNS accounts as well? Right now, filtering is only applied on blob items and all directories are visible which doesn't entirely restrict view as per user's request. |
We are not filtering directories because directories do not have tags, types, size and lmt populated so only name based filter can work on them. User might give name based filter only on the matching blob names hence I ignored the directories from the filtering logic. |
* feat: support workload identity token (#1556) * feat: support workload identity token * Create block pool only once in child process (#1581) * create block pool in child only * Update golang.org/x/crypto to v0.31.0 (#1594) * Update golang.org/x/crypto to v0.31.0 * sync with main (#1603) * updated year in copyright message (#1601) * Use ListBlob for hns accounts (#1555) * Optimize HNS listing * Added statfs for block-cache (#1470) * Added statfs for block_cache * Add strong consistency check for data on disk (#1604) * Add strong consistency check for data on disk * bug in block cache open call (#1580) * current implementation of open file when opened in O_WRONLY will truncate the file to zero. This is incorrect behaviour. We don't see it in the normal scenario as write-back cache is on by default. Hence all the open calls with O_WRONLY will be redirected O_RDWR. To simulate this turn of the write-back cache and then open file in O_WRONLY. * Feature: Blob filter (#1595) * Integrating blob filter in azstorage * Serve getAttr call for destination file after the Copy finishes from the cache * Cleanup on start shall be set to cleanup temp cache (#1613) * Add Tests * Refactor the code and refresh the cache after copying the attributes * Automate blobfuse2 setup for new VM (#1575) added script for blobfuse setup and azsecpack setup in VM * * Update the Unit tests. * Refactor the Code * Update Changelog * do go fmt on src * Downgrade go version to 1.22.7 due to memory issues in 1.23 (#1619) * Enable ETAG based validation on every block download to provide higher consistency (#1608) * Make etag validation a defualt option * BUG#31069208: Fixed Prefix filtering from File Path (#1618) * Fixed the logic to filter out folder prefix from path * Added/Updated/Removed test case --------- Co-authored-by: weizhi <[email protected]> Co-authored-by: Sourav Gupta <[email protected]> Co-authored-by: Jan Jagusch <[email protected]> Co-authored-by: ashruti-msft <[email protected]> Co-authored-by: syeleti-msft <[email protected]> Co-authored-by: jainakanksha-msft <[email protected]>
✅ What
Blob filter integration to filter out blob based on various parameters.
Filter library can be found here
🤔 Why
When user mounts blobfuse, all blob are visible and user is free to work on any blob. There are scenarios where customer do not want to have visibility of all the blobs that exists in container and rather have a constrained view based on blob names, size, LMT etc. This feature will allow that limited scope to be created.
👩🔬 How to validate if applicable
Mount with filter option and check all the blobs are visible or only the one matching the given filter
🔖 Related links