We continued working on eclipse-evasion. We also analysed and fixed a bug
when using DNS names in local root peers. We continued working on engineering
blog post about P2P. We released a new version of packages for
cardano-node-8.1.0 release.
We improved our CI, removed obsolete scripts add extra validation which checks
if CHANGELOG.md files were updated.
We also improved release scripts.
We started working on a new way to switch between root & ledger peers (see
below). We continued to work on eclipse-evasion. We merged changes to
Handshake contributed by Galois Inc. We made improvements to our tests (fixed
a flaky test, added cddl specs for NodeToNodeVersionData and
NodeToClientVersionData). We improved our CI and automated the process of
releasing new package version to CHaP.
We continued to work on testing eclipse-evasion.
We came up with an idea to limit how full node wallets relay on root peers (currently operated by IOG, in future also CF and Emurgo). We designed a switch to use ledger peers if the node tip is close enough to the current time. For more details see #4530.
We merged changes to the handshake mini-protocol which allow one to query
server's node-to-node / node-to-client parameters. We are grateful to
Galois Inc. for implementing it, #4256 and #4538. We published new version
of packages to CHaP chap-#253.
We added DiffusionError wrapper. Thanks to it, ouroboros-consensus will
not duplicate diffusion errors messages in the log, #4537.
We fixed an issue which caused one of our tests to be flaky, #4515.
We added cddl tests for NodeToNodeVersionData and NodeToClientVersionData:
#4540, #4544 (in review).
We wrote scripts which will help us release packages as well as verify that we released all the package necessary to build the newest set of packages, #4542.
We renamed the consensus startup tracer and make sure it doesn't log
ExitSuccess exceptions, consensus-#71.
We reviewed PR which adds RawBearer API, #4395.
We made series of improvements to our CI:
consensus moved
to a new repo;The networking team has released the following packages to Hackage:
We published an announcement blog post. We made progress also on another blog post which will describes the design of the P2P networking layer.
After moving ouroboros-consensus to its own repository, we made
ouroboros-network compatible with GHC 9.4 and 9.6.
We made cardano-client library independent of
ouroboros-consensus-diffusion.
We prepared release of network packages for cardano-node-8.0.
io-classes-1.1.0.0: ouroboros-network#4281.GHC 9.4 and 9.6: ouroboros-network#4527.cardano-client independent of ouroboros-cosnensus-diffusion: ouroboros-network#4526.cardano-node-8.0: ouroboros-network#4532.io-classes-1.0.0.0, and following up releases: 1.0.0.1 (ghc-9.6
support), 1.1.0.0 (documentation fix).io-sim in a series of PRs: io-sim#81,
io-sim#82, io-sim#86, io-sim#87.registerDelayCancellable test: [io-sim#80].io-sim repository.stylish-haskell: io-sim#83.typed-protocols-0.1.0.4 which support io-classes-1.1.0.0:
typed-protocols#227.typed-protocols: typed-protocols#30.stylish-haskell: typed-protocols#29.In the last spring we released cardano-node-1.35.6 with dynamic P2P
functionality.
We received reports from some SPOs who encountered problems with their non P2P block producing nodes not being able to connect to their P2P relay. Karl Knutsson (from Cardano Foundation) reproduced this issue between two nodes (a non P2P and a P2P one) on mainnet. Karl and the IOG Networking Team analysed it and found a bug in the legacy non p2p code. The bug is only possible to trigger with a P2P node which is binding its outbound connection port to a fixed IP address and port (default in p2p). A possible solution was found. For more information see #4465.
We released cardano-ping-0.1.0.0 package to CHaP. cardano-ping is no
longer available as a standalone binary, but instead it will become part of
cardano-cli (see #4664)
We are testing cardano-node with peer sharing functionality (#4019).
We are working on eclipse evasion. We added new class of peers: big ledger peers to the outbound governor, implemented tests and fixed found issues (#4462). We also made the information if a given peer plays the role of a big ledger peer to the mini-protocols. This will allow to modify mini-protocol applications for such peers. As part of this functionality we refactored some core types in the network code which simplifies exposed API.
Together with Moritz Angerman we started to update io-sim to ghc-9.6.1 (see
#73).
We merged a fix of configuration of accepted connections limit in
cardano-node (see #4902).
In the current sprint the networking team focused on fixing bugs and pushing forward implementation of eclipse evasion. We also found a bug in our simulation testing setup (in integration of test node). We also overviewed the work on extending handshake protocol which is delivered by Galois Inc.
We published ouroboros-network-0.4.0.1 and
ouroboros-network-protocols-0.3.0.0 to CHaP.
We also fixed a bug in cardano-node which results in not being able to
configure inbound connection limits, see PR #4902.
Together with Karl Knutsson (CF) we realised an issue in cardano-cli: it's
validation of DNS names, IP address & ports when registering a stake pool
should be more strict to protect against common mistakes which we identified on
the chain. See issue #4929.
In PR #4385 we fixed two bugs in peer state actions. First one results in
a busy loop if demotion from hot to warm times outs. This busy loop is
eventually exited when mux exits (we reported this in our previous report).
This fix made it to 1.35.6 release as well.
In addition the PR #4385 also fixes another bug which results in hot -> warm
-> hot demotion / promotion busy loop.
The PR #4385 also fixed a bug in a node only used in simulation which
resulted in not using chain-sync or block-fetch mini-protocols. In the
review process, we realised that the header-body split in the simulated node
requires further work (see PR #4419, which is under review).
The PR #4385 also extend our generators, which together with the above fix,
cover the hot -> warm -> hot demotion / promotion busy loop.
In PR #4419 we introduce a ChainDB for our simulation node, which plays
similar role to ChainDB in the ouroboros-consensus: a persistent (across
simulated restarts) store of blocks which does chain selection. This ensures
that the simulated node is using block-fetch to download blocks announced by
chain-sync mini-protocol.
We also made progress with reviewing PR #4019 - peer sharing.
We also fixed issue #4370 - a connection manager test failure, see PR #4384.