- Adds our S3 bucket for caching docker images as Protocol Labs shut
down their shared one.
- Remove the free disk space workaround that prevented the jobs from
failing for using too much space for the images.
---------
Co-authored-by: diegomrsantos <diego@status.im>
Hello!
This PR aim to refactor rendezvous code so that it is easier to impl.
Waku rdv strategy. The hardcoded min and max TTL were out of range with
what we needed and specifying which peers to interact with is also
needed since Waku deals with peers on multiple separate shards.
I tried to keep the changes to a minimum, specifically I did not change
the name of any public procs which result in less than descriptive names
in some cases. I also wanted to return results instead of raising
exceptions but didn't. Would it be acceptable to do so?
Please advise on best practices, thank you.
---------
Co-authored-by: Ludovic Chenut <ludovic@status.im>
Updating libp2p.nimble to 1.6.0.
This commit is planned to be tagged with 1.6.0.
The only new feature in this version is the highly experimental Quick
transport. Still, it is a feature and justifies a minor version upgrade.
It also contains a few fixes and improvements (see commit history).
Co-authored-by: ksr <kaiserd@users.noreply.github.com>
- fixes https://github.com/vacp2p/nim-libp2p/issues/1194.
- fixes ambiguous `KeyError`
- removes an unnecessary type param for `newSeqWith`
- fixes tests for `safeConvert`
The main fixes relate to Nim 2.2 being more strict and not accepting
calls with a wrong number of type parameters.
Our quic effort is blocked by bearssl not supporting TLS1.3, but since
Mark did most of the work to implement Quic here:
https://github.com/status-im/nim-libp2p/pull/563 and on nim-quic, this
PR is going to bring encryption-less Quic into nim-libp2p
This allows us to test it, and make sure it doesn't bitrot.
Heavily WiP:
- [X] Extract code from #563
- [X] Create custom muxer & upgrader
- [X] Basic E2E switch test working
- [x] Update nim-quic to get address informations in libp2p (for
`observed address` and port 0 resolving)
- [ ] More tests
- [ ] Cleanup
Since encryption is not yet supported, we're not compatible with any
other libp2ps, and have to rely on home made protocols to retrieve the
peer's id
---------
Co-authored-by: markspanbroek <mark@spanbroek.net>
Co-authored-by: Diego <diego@status.im>
This change is necessary before enabling merge queues. The `merge_group`
event is needed to trigger the GitHub Actions workflow when a pull
request is added to a merge queue.
The merge queue provides the same benefits as the Require branches to be
up to date before merging branch protection but does not require a pull
request author to update their pull request branch and wait for status
checks to finish before trying to merge. More info on
https://docs.github.com/en/repositories/configuring-branches-and-merges-in-your-repository/configuring-pull-request-merges/managing-a-merge-queue.
`LPError` is the top level error type of libp2p, it makes more sense to
raise a multi address specific subtype to avoid requiring callers to
catch errors too broadly. As `MaError` inherits from `LPError`, existing
error handlers will still work fine.
---------
Co-authored-by: diegomrsantos <diego@status.im>
- Moving the `mm:refc` configuration to the main `config.nims` file
makes it be used in tests and tasks that build examples.
- It's not clear why `skipParentCfg`was being used and it doesn't seem
to be necessary.
Bring a bit more detail when the "Too many connections for peer" message
is being logged. Particularly, we are adding the offending `peerId` and
the stream direction
Co-authored-by: diegomrsantos <diego@status.im>
For example:
```
/Users/runner/work/nim-libp2p/nim-libp2p/nimbledeps/pkgs2/websock-0.1.0-94f836ae589056b2deb04bdfdcd614fff80adaf5/websock/http/client.nim(173, 5) template/generic instantiation of `async` from here
/Users/runner/work/nim-libp2p/nim-libp2p/nimbledeps/pkgs2/websock-0.1.0-94f836ae589056b2deb04bdfdcd614fff80adaf5/websock/http/client.nim(165, 1) Warning: The raises pragma doesn't work on async procedures - use `async: (raises: [...]) instead. [User]
/Users/runner/work/nim-libp2p/nim-libp2p/nimbledeps/pkgs2/websock-0.1.0-94f836ae589056b2deb04bdfdcd614fff80adaf5/websock/websock.nim(257, 5) template/generic instantiation of `async` from here
/Users/runner/work/nim-libp2p/nim-libp2p/nimbledeps/pkgs2/websock-0.1.0-94f836ae589056b2deb04bdfdcd614fff80adaf5/websock/websock.nim(251, 1) Warning: The raises pragma doesn't work on async procedures - use `async: (raises: [...]) instead. [User]
/Users/runner/work/nim-libp2p/nim-libp2p/libp2p/transports/wstransport.nim(77, 18) template/generic instantiation of `async` from here
/Users/runner/work/nim-libp2p/nim-libp2p/libp2p/transports/wstransport.nim(83, 10) template/generic instantiation of `setResult` from here
/Users/runner/work/nim-libp2p/nim-libp2p/libp2p/transports/wstransport.nim(78, 26) template/generic instantiation of `mapExceptions` from here
/Users/runner/work/nim-libp2p/nim-libp2p/nimbledeps/pkgs2/chronos-4.0.2-c5e9517b9189713210e2abab8b77a68da71ded12/chronos/internal/asyncmacro.nim(542, 60) Error: expression 'value(cast[type(recv(s.session, pbytes, nbytes))](chronosInternalRetFuture.internalChild))' is of type 'int' and has to be used (or discarded); start of expression here: /Users/runner/work/nim-libp2p/nim-libp2p/libp2p/transports/wstransport.nim(78, 26)
stack trace: (most recent call last)
```
from
https://github.com/vacp2p/nim-libp2p/actions/runs/10655841970/job/29533846606?pr=1145
For minimal example of this:
```nim
template g(body: untyped) =
try:
body
except CatchableError:
raise newException(CatchableError, "")
discard g(0)
```
Also, even in 2.0.8, a variation doesn't work:
```
template g(body: untyped) = body
discard g(0)
```
Free disk space before running the steps of interop's transport test.
The original job has enough space but in our repo it crashes midway due
to missing disk space.
---------
Co-authored-by: kaiserd <1684595+kaiserd@users.noreply.github.com>
Co-authored-by: Diego <diego@status.im>
### Description
Adding an `onValidated` observer which will run every time a message is
received and validated. This comes from the necessity of precisely track
message deliveries and network activity.
`onRecv` observers run before any check is performed on the received
message, which means that it runs every time a duplicate or invalid
message arrives, which is inefficient and inaccurate for our purpose of
tracking only received, unique and valid messages. Therefore, adding
this extra option of running an observer for every message after all
validation checks pass.
The failure is due to incompatibility (in caching) after Nimble's
v.0.14.0 update, where they changed the dependencies directory name from
`pkgs` to `pkgs2`.
This PR includes the nim branch in the cache key to avoid the directory
name issue.
In the future, if we deprecate support for Nim 1.6 we may remove this.
fixes https://github.com/vacp2p/nim-libp2p/issues/1157
---------
Co-authored-by: Diego <diego@status.im>
We are trying to reduce the logs load in our fleets, and one of the most
recurrent one is
```
Dialing canceled topics="libp2p dialer" tid=1 file=dialer.nim:67 err="Future operation cancelled!" peerId=16U*XAFJX3
```
which is quite spammy and doesn't give much info.
In addition to that, its corresponding `Dialing address` log is in
trace.
So adjusting the log level of `Dialing canceled` to trace :)
The main motivation was to update the Ubuntu version on the daily job as
it seemed it wasn't supported anymore.
macOS 14 fails immediately with no error msg, so we are using 13 for
now.
This PR enables Nim 2.0.x with `refc` garbage collector on CI.
The following compilation error had to be fixed: Error: undeclared
identifier: 'acceptHandler`gensym435'; if declared in a template, this
identifier may be inconsistently marked inject or gensym
Ludovic Chenut
Álex
Simon-Pierre Vivier
kaiserd
diegomrsantos
Tanguy
Etan Kissling
Ivan FB
Cofson
tersec
gabrielmer