-
Notifications
You must be signed in to change notification settings - Fork 354
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
Fix tests that fail on macOS 14 #5441
Conversation
b1b5c2f
to
c2fc533
Compare
b912a87
to
33bba9e
Compare
DES-443 Fix failing tests on macOS 14
Most tests currently fail on macOS 14. David mentioned it might be because the test framework is picking the wrong network interface for something. |
ab224fb
to
f79a0aa
Compare
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Great job, looking forward to getting this merged! 🎉
Reviewed 21 of 21 files at r1, all commit messages.
Reviewable status: all files reviewed, 3 unresolved discussions (waiting on @dlon)
test/test-manager/src/tests/helpers.rs
line 99 at r1 (raw file):
pub async fn send_guest_probes( rpc: ServiceClient, interface: Option<String>,
This argument does not have to be an Option<String>
, it could safely be just a String
.
I did not find any instance where None
was passed as the interface
argument.
Code quote:
interface: Option<String>
test/test-runner/src/net.rs
line 56 at r1 (raw file):
log::debug!("Connecting from {bind_addr} to {destination}/TCP"); tokio::task::spawn_blocking(move || {
What is the motivation for this change?
I suggest reverting this change and use the tokio::net::TcpStream
, instead of manually putting synchronous I/O on a blocking-friendly thread.
Code quote:
tokio::task::spawn_blocking(move || {
test/test-runner/src/net.rs
line 121 at r1 (raw file):
} let _ = tokio::task::spawn_blocking(move || {
dito
Code quote:
let _ = tokio::task::spawn_blocking(move || {
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Reviewable status: all files reviewed, 3 unresolved discussions (waiting on @MarkusPettersson98)
test/test-runner/src/net.rs
line 56 at r1 (raw file):
Previously, MarkusPettersson98 (Markus Pettersson) wrote…
What is the motivation for this change?
I suggest reverting this change and use thetokio::net::TcpStream
, instead of manually putting synchronous I/O on a blocking-friendly thread.
connect()
sometimes fails due to EINPROGRESS
. I'm not sure whether it's fine to convert the socket to a tokio::net::TcpStream
in this state. I suppose we could use blocking I/O only for connect()
.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Reviewable status: all files reviewed, 3 unresolved discussions (waiting on @MarkusPettersson98)
test/test-manager/src/tests/helpers.rs
line 99 at r1 (raw file):
Previously, MarkusPettersson98 (Markus Pettersson) wrote…
This argument does not have to be an
Option<String>
, it could safely be just aString
.
I did not find any instance whereNone
was passed as theinterface
argument.
We're calling send_guest_probes_without_monitor
in install.rs
. Should this function not have the same interface as that one?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Reviewable status: all files reviewed, 3 unresolved discussions (waiting on @dlon)
test/test-manager/src/tests/helpers.rs
line 99 at r1 (raw file):
Previously, dlon (David Lönnhager) wrote…
We're calling
send_guest_probes_without_monitor
ininstall.rs
. Should this function not have the same interface as that one?
I think a function's interface should reflect what it actually needs - in this case I find it a bit leaky to have send_guest_probes
also take interface
as an Optional<String>
just because it calls send_guest_probes_without_monitor
😊
test/test-runner/src/net.rs
line 56 at r1 (raw file):
Previously, dlon (David Lönnhager) wrote…
connect()
sometimes fails due toEINPROGRESS
. I'm not sure whether it's fine to convert the socket to atokio::net::TcpStream
in this state. I suppose we could use blocking I/O only forconnect()
.
Alright, nevermind then!
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Reviewable status: 18 of 21 files reviewed, 3 unresolved discussions (waiting on @MarkusPettersson98)
test/test-manager/src/tests/helpers.rs
line 99 at r1 (raw file):
Previously, MarkusPettersson98 (Markus Pettersson) wrote…
I think a function's interface should reflect what it actually needs - in this case I find it a bit leaky to have
send_guest_probes
also takeinterface
as anOptional<String>
just because it callssend_guest_probes_without_monitor
😊
Done.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Reviewed 2 of 3 files at r2, 1 of 1 files at r3, all commit messages.
Reviewable status: complete! all files reviewed, all discussions resolved
… 'checkout' action
…machine has actually handled the request
6ad878e
to
56970ba
Compare
Main changes:
set_allow_lan
,set_dns_options
,set_block_when_disconnected
.fetch-tags
in the GHA e2e workflow. Usegit fetch --tags
directly.test_multihop
andtest_udp2tcp
(analogously to Remedy connection check timeouts intest_bridge
#5442).Fix DES-443.
This change is