metabase: Fix EC search #1371
No reviewers
TrueCloudLab/storage-core-committers
Labels
No labels
P0
P1
P2
P3
badger
frostfs-adm
frostfs-cli
frostfs-ir
frostfs-lens
frostfs-node
good first issue
triage
Infrastructure
blocked
bug
config
discussion
documentation
duplicate
enhancement
go
help wanted
internal
invalid
kludge
observability
perfomance
question
refactoring
wontfix
No milestone
No project
No assignees
3 participants
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference: TrueCloudLab/frostfs-node#1371
Loading…
Add table
Reference in a new issue
No description provided.
Delete branch "dstepanov-yadro/frostfs-node:fix/ec_search"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Closes #1347
Before fix:
After fix:
77128f1e6b
to97b5a1e8d2
metabase: Fix EC searchto WIP: metabase: Fix EC searchSupport/v0.42:
Current branch:
97b5a1e8d2
to119412cdc8
@ -418,3 +421,3 @@
data = []byte(v)
} else {
return f[i].Operation() == objectSDK.MatchNotPresent
return result, f[i].Operation() == objectSDK.MatchNotPresent
Not sure that returning EC chunk in case of
NOT PRESENT
filter is ok, but this is behavior ofsupport/v0.42
branch. See PR comments for example.WIP: metabase: Fix EC searchto metabase: Fix EC search@ -387,3 +388,2 @@
return true
return result, true
}
unrelated
fixed
@ -154,3 +154,3 @@
if !match {
continue // ignore objects with unmatched slow filters
}
unrelated
fixed
@ -415,23 +415,26 @@ func (db *DB) matchSlowFilters(tx *bbolt.Tx, addr oid.Address, f objectSDK.Searc
default: // user attribute
v, ok := attributeValue(obj, f[i].Header())
if ok {
if ech := obj.ECHeader(); ech != nil {
Could you describe in the commit message what is the problem here?
It seems strange that if we have found an attirbute we change the address to the parent.
done
119412cdc8
to7ebd86b751