You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on May 3, 2024. It is now read-only.
This file contains 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
In exiting code, decision about group is loaded vs not loaded is taken based on
grp->bgi_extents pointer value.
On restart if free fragment count = 0 then bgi_extents will remain NULL.
Later when any fragment is freed up then corresponding entry in extent list
and extent btree will be added by creating new extent using lext_create()
api i.e le_is_alloc = true.
The subsequent call to m0_balloc_load_extents() will find bgi_extents == NULL
and try to load extent allocated using M0_ALLOC_ARR where le_is_alloc = false,
but with same start/end values.
This ends up in having overlapping entry in extent list and mismatch in fragment
count in group info vs actual total entries in list.
If one of the overlapping extents goes off w.r. to value stored in btree due subsequent
free/alloc operations, then extent btree update operation fails with -2 (ENOENT) error,
Design
For Bug, Describe the fix here.
This PR is to add separate flag to decide about group loaded vs not loaded
status, and will avoid overlapping entries scenario mentioned above.
Coding
Checklist for Author
[ x ] Coding conventions are followed and code is consistent
Testing
Checklist for Author
Unit and System Tests are added
Test Cases cover Happy Path, Non-Happy Path and Scalability
Testing was performed with RPM
Impact Analysis
Checklist for Author/Reviewer/GateKeeper
Interface change (if any) are documented
Side effects on other features (deployment/upgrade)
Dependencies on other component(s)
Review Checklist
Checklist for Author
[ x ] JIRA number/GitHub Issue added to PR
[ x ] PR is self reviewed
[ x ] Jira and state/status is updated and JIRA is updated with PR link
[ x ] Check if the description is clear and explained
Documentation
Checklist for Author
Changes done to WIKI / Confluence page / Quick Start Guide
In exiting code, decision about group is loaded or not is taken based on
grp->bgi_extents pointer value. On restart if free fragment count = 0 then
bgi_extents will remain NULL. Later when any fragment is freeup then
corresponding entry in extent list and extent btree will be added by creating
new extent using lext_create() api i.e le_is_alloc = true. The subsequent call
to m0_balloc_load_extents() will find bgi_extents == NULL and try to load
extent allocated using M0_ALLOC_ARR where le_is_alloc = false, but with
same start/end values. This ends up in having overlapping entry in extent list
and mismatch in fragment count in group info vs total entries in list.
This PR is to add separate flag to decide about group loaded vs not loaded
status, and will avoid overlapping entries scenario mentioned above.
Signed-off-by: Mukund Kanekar <mukund.kanekar@seagate.com>
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.
mukundkanekar
requested review from
mehjoshi,
madhavemuri,
nikitadanilov,
yeshpal-jain-seagate,
huanghua78,
andriytk,
siningwuseagate,
vidyadhar-pinglikar,
shashank-parulekar,
nkommuri,
sg-shankar,
yatin-mahajan and
rkothiya
as code owners
CORTX-34042 - Fix load_extents logic in balloc module.
Signed-off-by: Mukund Kanekar mukund.kanekar@seagate.com
Problem Statement
grp->bgi_extents pointer value.
On restart if free fragment count = 0 then bgi_extents will remain NULL.
Later when any fragment is freed up then corresponding entry in extent list
and extent btree will be added by creating new extent using lext_create()
api i.e le_is_alloc = true.
The subsequent call to m0_balloc_load_extents() will find bgi_extents == NULL
and try to load extent allocated using M0_ALLOC_ARR where le_is_alloc = false,
but with same start/end values.
This ends up in having overlapping entry in extent list and mismatch in fragment
count in group info vs actual total entries in list.
If one of the overlapping extents goes off w.r. to value stored in btree due subsequent
free/alloc operations, then extent btree update operation fails with -2 (ENOENT) error,
Design
This PR is to add separate flag to decide about group loaded vs not loaded
status, and will avoid overlapping entries scenario mentioned above.
Coding
Checklist for Author
Testing
Checklist for Author
Impact Analysis
Checklist for Author/Reviewer/GateKeeper
Review Checklist
Checklist for Author
Documentation
Checklist for Author