Respect TTL for EC put #1288
No reviewers
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
2 participants
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference: TrueCloudLab/frostfs-node#1288
Loading…
Reference in a new issue
No description provided.
Delete branch "dstepanov-yadro/frostfs-node:fix/ec_put_respect_ttl"
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?
Scenario:
EC 2.1 CBF 1 SELECT 4 FROM *
container, netmap contains 4 nodes, but one node (node 2
) is offnode 1
,node 1
splits object to 3 chunksNode 1
tries to savechunk 0
tonode 2
, gets error, and tries to savechunk 0
tonode 4
.Node 4
tries to savechunk 0
tonode 2
again, gets error, and saveschunk 0
locally or tries to savechunk 0
tonode 3
.Now node takes into account if node gets prepared EC part. It should work so:
8f782503b0
toe4179dac8f
@ -238,6 +251,12 @@ func (e *ecWriter) writePart(ctx context.Context, obj *objectSDK.Object, partIdx
partVisited := make([]bool, len(nodes))
partVisited[partIdx%len(nodes)] = true
select {
Same here
Fixed
@ -263,6 +282,12 @@ func (e *ecWriter) writePart(ctx context.Context, obj *objectSDK.Object, partIdx
partVisited[idx] = true
}
select {
You do it on the first iteration of the following loop, why this check?
Oops, debugee. fixed.
e4179dac8f
toa1347542c8