configurable pubsub signing #5647
Conversation
ghost
assigned 
| @@ -475,14 +475,23 @@ func (n *IpfsNode) startOnlineServicesWithHost(ctx context.Context, host p2phost | |||
|
|
|||
| var service *pubsub.PubSub | |||
|
|
|||
| var pubsubOptions []pubsub.Option | |||
There was a problem hiding this comment.
What happens if they are both set?
There was a problem hiding this comment.
The user won't be able to send any messages. We should error in that case, shouldn't we...
There was a problem hiding this comment.
There was a problem hiding this comment.
Re-reading this now with some more context it seems like a valid case (although I'm not sure how useful it is), that is, we are requiring signed messages but we don't sign them ourselves.
There was a problem hiding this comment.
It is a valid case, technically. But it's so useless that we should probably get rid of it.
There was a problem hiding this comment.
(actually, I may just do that in a followup patch once someone takes a look at that PR)
There was a problem hiding this comment.
if strict signing is set but signing is disabled, then it will be ignored.
There was a problem hiding this comment.
Got it. Now that the option validation patch has been merged, it should return an error on construction. That should be the least surprising result.
| # ipfs pubsub sub is long-running so we need to start it in the background and | ||
| # wait put its output somewhere where we can access it | ||
| ( | ||
| ipfsi 0 pubsub sub --enc=ndpayload testTopic | if read line; then |
There was a problem hiding this comment.
What is ipfsi? (I mean, where can I read some information about it?)
There was a problem hiding this comment.
It's a function defined in test/sharness/lib/iptb-lib.sh.
| @@ -475,14 +475,23 @@ func (n *IpfsNode) startOnlineServicesWithHost(ctx context.Context, host p2phost | |||
|
|
|||
| var service *pubsub.PubSub | |||
|
|
|||
| var pubsubOptions []pubsub.Option | |||
There was a problem hiding this comment.
Re-reading this now with some more context it seems like a valid case (although I'm not sure how useful it is), that is, we are requiring signed messages but we don't sign them ourselves.
It's an arbitrary (binary) string. At one point, it was the hash of a topic descriptor but that never really got finished. |
|
@schomatis mind ticking ipfs/go-ipfs-config#18 as well? |
I'd like to sneak this into the release so we can turn on strict verification ASAP. License: MIT Signed-off-by: Steven Allen <steven@stebalien.com>
Stebalien
force-pushed
the
feat/pubsub-signing
branch
from
95935fb
to
2b0bc7e
Compare
I'd like to sneak this into the release so we can turn on strict verification ASAP.
Depends on ipfs/go-ipfs-config#18 and a gx update.