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

feat: implement peering service #7362

Merged
merged 9 commits into from
May 26, 2020
Merged

feat: implement peering service #7362

merged 9 commits into from
May 26, 2020

Conversation

Stebalien
Copy link
Member

MVP for #6097

This feature will repeatedly reconnect (with a randomized exponential backoff) to peers in a set of "peered" peers.

In the future, this should be extended to:

  1. Include a CLI for modifying this list at runtime.
  2. Include additional options for peers we want to protect but not connect to.
  3. Allow configuring timeouts, backoff, etc.
  4. Allow groups? Possibly through textile threads.
  5. Allow for runtime-only peering rules.
  6. Different reconnect policies.

But this MVP should be a significant step forward.

@Stebalien
Copy link
Member Author

Depends on ipfs/go-ipfs-config#96

@@ -703,6 +706,26 @@ intentionally re-using the real message's message ID.

Default: `false`

### `Peering`
Copy link
Contributor

Choose a reason for hiding this comment

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

include expected behavior / semantics for how this behaves when one of the two peers has it, and when both have it mutually.

go.mod Show resolved Hide resolved
peering/peering.go Outdated Show resolved Hide resolved
peering/peering.go Outdated Show resolved Hide resolved
nextDelay time.Duration
}

func (ph *peerHandler) stop() {
Copy link
Contributor

Choose a reason for hiding this comment

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

cancel ph.ctx ?

peering/peering_test.go Show resolved Hide resolved
Copy link
Contributor

@aschmahmann aschmahmann left a comment

Choose a reason for hiding this comment

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

Seems reasonable to me, I left a few comments/change requests but nothing super dramatic in case you want to just merge this and come back to it later.

peering/peering.go Outdated Show resolved Hide resolved
Comment on lines +142 to +143
- [`Peering`](#peering)
- [`Peering.Peers`](#peeringpeers)
Copy link
Contributor

Choose a reason for hiding this comment

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

should any of this behavior going under the experimental label or no?

Copy link
Member Author

Choose a reason for hiding this comment

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

I figured it was a minimal enough feature that that wasn't really necessary. I want people to start using this feature. Thoughts?

Copy link
Member Author

Choose a reason for hiding this comment

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

But we can revisit this decision once the RC has shipped.

Copy link
Contributor

Choose a reason for hiding this comment

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

Seems fine, I wasn't really sure what qualified as experimental vs not which is why I asked 😄

peering/peering.go Outdated Show resolved Hide resolved
test/sharness/t0171-peering.sh Outdated Show resolved Hide resolved
peering/peering.go Show resolved Hide resolved
peering/peering.go Show resolved Hide resolved
peering/peering.go Outdated Show resolved Hide resolved
peering/peering.go Outdated Show resolved Hide resolved
Stebalien and others added 5 commits May 25, 2020 20:24
MVP for #6097

This feature will repeatedly reconnect (with a randomized exponential backoff)
to peers in a set of "peered" peers.

In the future, this should be extended to:

1. Include a CLI for modifying this list at runtime.
2. Include additional options for peers we want to _protect_ but not connect to.
3. Allow configuring timeouts, backoff, etc.
4. Allow groups? Possibly through textile threads.
5. Allow for runtime-only peering rules.
6. Different reconnect policies.

But this MVP should be a significant step forward.
Co-authored-by: Adin Schmahmann <[email protected]>
* better name for timer
* cancel context from within stop
* Explain _why_ it exists.
* Explain how it can be used.
@Stebalien
Copy link
Member Author

@aschmahmann could you give this a once over post merge? I think it's good enough to go in the RC, but I'd like you to look at my changes.

@Stebalien Stebalien mentioned this pull request May 26, 2020
77 tasks
Copy link
Contributor

@aschmahmann aschmahmann left a comment

Choose a reason for hiding this comment

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

@Stebalien left a small comment on the go test, but LGTM

return h1.Network().Connectedness(h2.ID()) == network.Connected
}, 30*time.Second, 100*time.Millisecond)

require.Len(t, h1.Network().Peers(), 3)
Copy link
Contributor

Choose a reason for hiding this comment

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

I don't know if it's worth doing anything about this, but noting that there's an unlikely race with the connmgr here where we prune in between connecting and doing the size check.

Copy link
Member Author

Choose a reason for hiding this comment

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

I set the high water to 100 so that shouldn't happen.

@hacdias hacdias deleted the feat/peering branch May 9, 2023 10:59
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