Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Default validator support #525

Merged
merged 7 commits into from
Mar 1, 2023
Merged

Default validator support #525

merged 7 commits into from
Mar 1, 2023

Conversation

vyzo
Copy link
Collaborator

@vyzo vyzo commented Feb 28, 2023

Adds support for default validators and provides an implementation that validates the message seqno as a per peer nocne to prevent infinite propagation when the network diameter exceeds the seen cache span.
See #524
Implements the logic described in ipfs/kubo#9665

TBD:

  • unit test

Copy link
Contributor

@MarcoPolo MarcoPolo left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

One suggestion to the doc comment. Would be nice to have some tests here as well, but understood that this is urgent so we can defer to a future PR.

validation_builtin.go Show resolved Hide resolved
@vyzo vyzo merged commit 829f902 into master Mar 1, 2023
@vyzo vyzo deleted the feat/default-validator branch March 1, 2023 14:16
@stbrody
Copy link

stbrody commented Mar 1, 2023

Is there a risk here if messages are delivered out of order? Like if a node receives message 5 before message 4 from the same peer, will that node then never forward message 4 on to its other peers? Does that risk some messages not being well propagated to the whole network?

@vyzo
Copy link
Collaborator Author

vyzo commented Mar 2, 2023

This can only happen if one message from the peer is dropped (because of a full queue) and the next message propagates before the dropped message arrives from other paths.

So in theory there is some risk, but this would be quite difficult to trigger.
Maybe with an extremely high sending rate from some peer.

Worth noting, but i am not convinced it is a problem.

@vyzo
Copy link
Collaborator Author

vyzo commented Mar 4, 2023

There is a simple way to alleviate concerns here: keep a bitvector (say 32 or 64 bits, as an integer) that encodes whether the last few messages have been delivered.

How does that sound?

@stbrody
Copy link

stbrody commented Mar 5, 2023

@vyzo I don't think I fully understand how the bitvector would work, but I think keeping a small cache of the last few seqnos and whether they've been delivered once before does make sense and would address the issue I brought up!

@vyzo
Copy link
Collaborator Author

vyzo commented Mar 5, 2023

Ok, i will implement something along these lines.

yhassanzadeh13 added a commit to yhassanzadeh13/go-libp2p-pubsub that referenced this pull request Jul 3, 2023
* feat: expire messages from the cache based on last seen time (libp2p#513)

* feat: expire messages from the cache based on last seen time

* chore: minor renaming

* fix: messages should not be found after expiration

* chore: editorial

* fix: use new time cache strategy consistently

* fix: default to old time cache and add todo for background gc

* chore: update to go-libp2p v0.25 (libp2p#517)

* Update to go-libp2p v0.25

* Use go 1.19

* chore: update go version and dependencies (libp2p#516)

* fix(timecache): remove panic in first seen cache on Add (libp2p#522)

* Refactor timecache implementations (libp2p#523)

* reimplement timecache for sane and performant behaviour

* remove seenMessagesMx, take advantage of new tc api

* fix timecache tests

* fix typo

* store expiry, don't make life difficult

* refactor common background sweep procedure for both impls

* add godocs to TimeCache

* Default validator support (libp2p#525)

* add default validator support

* add an implementation for basic seqno as nonce validation

* missing return

* the nonce belongs to the origin peer

* add note about rust predicament

* add seqno validator tests

* minor test tweak, ensure at least 1ms before replay

* Fix Memory Leak In New Timecache Implementations (libp2p#528)

* fix bug

* add for last seen cache

* chore: Update .github/workflows/stale.yml [skip ci]

* chore: Update .github/workflows/stale.yml [skip ci]

* upgrades libp2p version

* upgrades libp2p version

---------

Co-authored-by: Mohsin Zaidi <[email protected]>
Co-authored-by: Marco Munizaga <[email protected]>
Co-authored-by: RichΛrd <[email protected]>
Co-authored-by: Hlib Kanunnikov <[email protected]>
Co-authored-by: vyzo <[email protected]>
Co-authored-by: Nishant Das <[email protected]>
Co-authored-by: GitHub <[email protected]>
yhassanzadeh13 added a commit to yhassanzadeh13/go-libp2p-pubsub that referenced this pull request Feb 20, 2024
* feat: expire messages from the cache based on last seen time (libp2p#513)

* feat: expire messages from the cache based on last seen time

* chore: minor renaming

* fix: messages should not be found after expiration

* chore: editorial

* fix: use new time cache strategy consistently

* fix: default to old time cache and add todo for background gc

* chore: update to go-libp2p v0.25 (libp2p#517)

* Update to go-libp2p v0.25

* Use go 1.19

* chore: update go version and dependencies (libp2p#516)

* fix(timecache): remove panic in first seen cache on Add (libp2p#522)

* Refactor timecache implementations (libp2p#523)

* reimplement timecache for sane and performant behaviour

* remove seenMessagesMx, take advantage of new tc api

* fix timecache tests

* fix typo

* store expiry, don't make life difficult

* refactor common background sweep procedure for both impls

* add godocs to TimeCache

* Default validator support (libp2p#525)

* add default validator support

* add an implementation for basic seqno as nonce validation

* missing return

* the nonce belongs to the origin peer

* add note about rust predicament

* add seqno validator tests

* minor test tweak, ensure at least 1ms before replay

* Fix Memory Leak In New Timecache Implementations (libp2p#528)

* fix bug

* add for last seen cache

* chore: Update .github/workflows/stale.yml [skip ci]

* chore: Update .github/workflows/stale.yml [skip ci]

* bump golang.org/x/net from 0.4.0 to 0.7.0 (libp2p#520)

Bumps [golang.org/x/net](https://github.com/golang/net) from 0.4.0 to 0.7.0.
- [Release notes](https://github.com/golang/net/releases)
- [Commits](golang/net@v0.4.0...v0.7.0)

---
updated-dependencies:
- dependency-name: golang.org/x/net
  dependency-type: indirect
...

Signed-off-by: dependabot[bot] <[email protected]>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>

* fix: topicscore params can't be set for dynamically subscribed topic (libp2p#540)

* fix: topicscore params can't be set for a topic subscribed after gossipsub is initialized

* chore:address review comments

* Revert "fix: topicscore params can't be set for dynamically subscribed topic (libp2p#540)" (libp2p#541)

This reverts commit aa5fd79.

* remove usage of deprecated peerid.Pretty method (libp2p#542)

* chore: update go-libp2p to v0.32 (libp2p#548)

* chore: Update .github/workflows/stale.yml [skip ci]

* make tidy

---------

Signed-off-by: dependabot[bot] <[email protected]>
Co-authored-by: Mohsin Zaidi <[email protected]>
Co-authored-by: Marco Munizaga <[email protected]>
Co-authored-by: RichΛrd <[email protected]>
Co-authored-by: Hlib Kanunnikov <[email protected]>
Co-authored-by: vyzo <[email protected]>
Co-authored-by: Nishant Das <[email protected]>
Co-authored-by: GitHub <[email protected]>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
Co-authored-by: Prem Chaitanya Prathi <[email protected]>
Co-authored-by: Sukun <[email protected]>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

3 participants