Draft: Nftables sets #1017
Draft: Nftables sets #1017
Conversation
| @@ -46,6 +47,55 @@ | |||
| NFT_MSG_GETFLOWTABLE = 23 | |||
| NFT_MSG_DELFLOWTABLE = 24 | |||
|
|
|||
| # from nftables/include/datatype.h | |||
There was a problem hiding this comment.
Just a comment, what do you think — would we benefit of using enum in these cases instead of a plain definition of variables?
Not starting with this PR, but in some future maybe.
There was a problem hiding this comment.
Yes I think it would be better
pyroute2/nftables/main.py
Outdated
| else: | ||
| kwarg.pop("elements") | ||
|
|
||
| kwarg["elements"] = [ |
There was a problem hiding this comment.
I already started to work on kwarg filters for IPRoute, see /pyroute2/requests/; eventually we could make some common API that solves common transformations like str 255.255.255.0 ⇒ int 24, or iterable ( int interface_index ) ⇒ int interface_index
There was a problem hiding this comment.
I will see that thanks !
|
LGTM, if you could fix pep8 (run I see no issues so far, but tonigh I'll look more closer and start some test routines as well. |
inemajo
force-pushed
the
nftables_sets
branch
5 times, most recently
from
6b16545
to
6767826
Compare
inemajo
force-pushed
the
nftables_sets
branch
from
6767826
to
6c2a59c
Compare
inemajo
force-pushed
the
nftables_sets
branch
2 times, most recently
from
d2e52d5
to
096da94
Compare
|
@inemajo does the functional test succeeds on your system? |
|
Yes, and I try on a VM with ubuntu-18 and they pass too, so I don't understand why they don't pass on the CI |
|
I put some new features on sets (timeout, counter...) I want to add userdata (for the comment entry) and after passing the MR to ready |
inemajo
force-pushed
the
nftables_sets
branch
from
504a423
to
ca75a64
Compare
|
Is it possible that the test has some prerequisites set up on your machines but on the CI VM:s? I'll take a look tonight, maybe will be lucky to find out. |
|
that's what I thought, the test create the filter table but it had no effect. Thanks for the help |
inemajo
force-pushed
the
nftables_sets
branch
from
ca75a64
to
f4746ad
Compare
…oseconds) in kwarg
…of expressions) and fix NFTA_SET_EXPR (only one expression)
…ESSIONS and fix NFTA_SET_ELEM_EXPR, it's only one nft_expr
…ment timeout and counter on sets
inemajo
force-pushed
the
nftables_sets
branch
from
f4746ad
to
4a107fa
Compare
|
Hello @svinota, I move the tests into examples for unlock the CI, do you think it's ok for you ? |
|
Sure thing, thanks. |
Hello @svinota, this is a draft for the issue #1013 (implementation of sets and set_elements for Nftables)