Conversation
5ae4534 to
411fd4d
Compare
Codecov Report❌ Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## master #3591 +/- ##
==========================================
+ Coverage 26.60% 26.64% +0.04%
==========================================
Files 655 655
Lines 49742 49702 -40
==========================================
+ Hits 13232 13244 +12
+ Misses 35466 35421 -45
+ Partials 1044 1037 -7 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
08f094b to
b0f05e9
Compare
b0f05e9 to
72d284c
Compare
| return err | ||
| } | ||
|
|
||
| if t.localNodeInContainer && t.metainfoConsistencyAttr != "" { |
There was a problem hiding this comment.
is t.localOnly branch OK w/ this? seems like empty lists are sent in this case
also, if client puts prepared objects with TTL=1, signatures are not sent. Is that ok?
There was a problem hiding this comment.
client puts prepared objects with TTL=1
I doubt we have any other option in this case. Client is doing something strange.
There was a problem hiding this comment.
we submitted a signature previously. Is it impossible now?
There was a problem hiding this comment.
This never worked correctly, a single signature is just not sufficient for a REP 3 container, transaction won't be accepted.
There was a problem hiding this comment.
seems like empty lists are sent in this case
lets cancel this
also, if client puts prepared objects with TTL=1, signatures are not sent. Is that ok?
i do not know what we can do in this case, this is an explicit wish to store an object and do nothing with meta data, without remote nodes, it will not be accepted by the contract
There was a problem hiding this comment.
@roman-khimov, @cthulhu-rider, check the last commit. maybe we can log this. i do not know what TTL=1 PUT should do with meta containers
There was a problem hiding this comment.
what TTL=1 PUT should do with meta containers
nothing. I thought that previously it worked, and dropping it is unrelated to the contract API fix. As it turns out, it didn't work before either. This could've been made into a separate patch, but I dont mind leaving it as is. Commit message remark as an option
Do not send signatures as a flat slice, use them as a two-dimensional matrix, as it was meant to be according to NeoFS placement policies. Refs https://github.com/nspcc-dev/neofs-contract/blob/50f4c2c6d83ec6661676a4059852f20fde11a78c/contracts/container/contract.go#L249 Refs neo-project/neo#1891. Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
This is likely will not be accepted (only in a strange case when there is REP 1 policy with the only local node). Signed-off-by: Pavel Karpy <carpawell@nspcc.ru>
72d284c to
14f5498
Compare
| } | ||
|
|
||
| if t.localNodeInContainer && t.metainfoConsistencyAttr != "" { | ||
| if !t.localOnly && t.localNodeInContainer && t.metainfoConsistencyAttr != "" { |
There was a problem hiding this comment.
looks correct now. We can now catch t.localOnly case before defer at the func's top, right?
Blocked by nspcc-dev/neofs-contract#519.