poc impersonate #39
Labels
No labels
P0
P1
P2
P3
good first issue
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-s3-gw#39
Loading…
Reference in a new issue
No description provided.
Delete branch "KirillovDenis/poc/impersonate"
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?
close #81
Should be merged after TrueCloudLab/frostfs-node#68
poc impersonateto WIP: poc impersonateWIP: poc impersonateto poc impersonate36edd0d1d0
to5909633649
5909633649
tob387f5e6ff
b387f5e6ff
tofb1817f672
fb1817f672
to72ed3b1f45
@ -388,3 +387,1 @@
if bktInfo.Owner.Equals(bearer.ResolveIssuer(*bd.Gate.BearerToken)) {
return bd.Gate.BearerToken.Marshal()
}
return bd.Gate.BearerToken.Marshal()
To be fully backward compatible, it seems that bearer token without impersonation flag should be treated as before: return token if
bktInfo.Owner.Equals(bearer.ResolveIssuer(*bd.Gate.BearerToken))
. Need to check that by accessing public-read-write bucket with foreign credentials without impersonation flag.You are right, we have to handle not impersonate tokens as before
72ed3b1f45
tob366e75366
LGTM.
We can also use lighter extended ACLs with less records by using impersonated tokens (private buckets may remove a set of allow rules for gate public keys).
Probably it should be discussed with new ACL scheme. /cc @realloc