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

changefeedccl: Add timeout to testfeed library. #84007

Merged
merged 1 commit into from
Jul 8, 2022

Conversation

miretskiy
Copy link
Contributor

@miretskiy miretskiy commented Jul 7, 2022

Add timeout to testfeed sink implementations to timeout
if no messages are received for too long.

Informs #83946

Release Notes: None

@miretskiy miretskiy requested a review from a team as a code owner July 7, 2022 15:54
@cockroach-teamcity
Copy link
Member

This change is Reviewable

Copy link
Collaborator

@stevendanna stevendanna left a comment

Choose a reason for hiding this comment

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

Looks good. No idea what the appropriate timeout is for Race builds, but it will definitely be nice for these to time out.

latestResolved hlc.Timestamp
}

var _ cdctest.TestFeed = (*sinklessFeed)(nil)

func deadline() time.Duration {
if util.RaceEnabled {
return time.Minute
Copy link
Collaborator

Choose a reason for hiding this comment

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

Race can be really slow in CI sometimes. I might even consider something like 5 minutes.

Copy link
Contributor Author

Choose a reason for hiding this comment

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

Ack.

@miretskiy miretskiy force-pushed the flake branch 2 times, most recently from 31cd6c3 to 2e17cff Compare July 7, 2022 18:37
@stevendanna
Copy link
Collaborator

stevendanna commented Jul 7, 2022

I'm fine with this merging as forward progress, but I think there is probably a long tail of places that need to be able to respect the contexts. Based on the code, I kinda expected this test to complete in roughly 30 seconds per sink, but it doesn't seem to (depending on the sink, kafka appears to work, sinkless and cloudsink don't)

func TestChangefeedTestTimesOut(t *testing.T) {
	defer leaktest.AfterTest(t)()
	defer log.Scope(t).Close(t)

	testFn := func(t *testing.T, s TestServer, f cdctest.TestFeedFactory) {
		sqlDB := sqlutils.MakeSQLRunner(s.DB)
		sqlDB.Exec(t, `CREATE TABLE foo (a INT PRIMARY KEY)`)
		nada := feed(t, f, "CREATE CHANGEFEED FOR foo")
		defer closeFeed(t, nada)
		require.Error(t, withTimeout(func(ctx context.Context) error {
			return assertPayloadsBaseErr(ctx, nada, []string{`not gonna happen: [2]->{"after": {"a": 2}}`}, false, false)
		}))
	}

	cdcTest(t, testFn)
}

Comment on lines 171 to 184
ctx := context.Background()
c.ctx, c.cancel = context.WithCancel(context.Background())
Copy link
Collaborator

Choose a reason for hiding this comment

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

For sinkless feeds, I think we need a timeout here. I don't think Query will return until at least one row returns, so if our changefeed never returns any rows, we don't time out.

Copy link
Contributor Author

@miretskiy miretskiy Jul 7, 2022

Choose a reason for hiding this comment

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

Not sure actually; I arrange to call cancel in Next():

defer time.AfterFunc(deadline(), c.cancel).Stop()

Copy link
Collaborator

Choose a reason for hiding this comment

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

But we never make it to next. Because Query never returns:

goroutine 38 [IO wait, 1 minutes]:
internal/poll.runtime_pollWait(0x370ba790, 0x72)
        GOROOT/src/runtime/netpoll.go:303 +0x85
internal/poll.(*pollDesc).wait(0xc0006dda00, 0xc000196400, 0x0)
        GOROOT/src/internal/poll/fd_poll_runtime.go:84 +0x32
internal/poll.(*pollDesc).waitRead(...)
        GOROOT/src/internal/poll/fd_poll_runtime.go:89
internal/poll.(*FD).Read(0xc0006dda00, {0xc000196400, 0x8e7, 0x8e7})
        GOROOT/src/internal/poll/fd_unix.go:167 +0x25a
net.(*netFD).Read(0xc0006dda00, {0xc000196400, 0x8e7, 0x0})
        GOROOT/src/net/fd_posix.go:56 +0x29
net.(*conn).Read(0xc001fed320, {0xc000196400, 0xc00601c9f8, 0xc00601ca08})
        GOROOT/src/net/net.go:183 +0x45
crypto/tls.(*atLeastReader).Read(0xc001fcab10, {0xc000196400, 0x0, 0x400ab8d})
        GOROOT/src/crypto/tls/conn.go:778 +0x3d
bytes.(*Buffer).ReadFrom(0xc005fa7e78, {0x9ea9140, 0xc001fcab10})
        GOROOT/src/bytes/buffer.go:204 +0x98
crypto/tls.(*Conn).readFromUntil(0xc005fa7c00, {0x9eb77c0, 0xc001fed320}, 0x4042391)
        GOROOT/src/crypto/tls/conn.go:800 +0xe5
crypto/tls.(*Conn).readRecordOrCCS(0xc005fa7c00, 0x0)
        GOROOT/src/crypto/tls/conn.go:607 +0x112
crypto/tls.(*Conn).readRecord(...)
        GOROOT/src/crypto/tls/conn.go:575
crypto/tls.(*Conn).Read(0xc005fa7c00, {0xc002b4622d, 0x1dd3, 0xc0028d8000})
        GOROOT/src/crypto/tls/conn.go:1278 +0x16f
io.ReadAtLeast({0x13e10e20, 0xc005fa7c00}, {0xc002b4622d, 0x1dd3, 0x1dd3}, 0x5)
        GOROOT/src/io/io.go:328 +0x9a
github.com/jackc/chunkreader/v2.(*ChunkReader).appendAtLeast(...)
        github.com/jackc/chunkreader/v2/external/com_github_jackc_chunkreader_v2/chunkreader.go:88
github.com/jackc/chunkreader/v2.(*ChunkReader).Next(0xc00505a340, 0x5)
        github.com/jackc/chunkreader/v2/external/com_github_jackc_chunkreader_v2/chunkreader.go:78 +0x2ac
github.com/jackc/pgproto3/v2.(*Frontend).Receive(0xc005189000)
        github.com/jackc/pgproto3/v2/external/com_github_jackc_pgproto3_v2/frontend.go:74 +0x50
github.com/jackc/pgconn.(*PgConn).peekMessage(0xc0056cec60)
        github.com/jackc/pgconn/external/com_github_jackc_pgconn/pgconn.go:499 +0x136
github.com/jackc/pgconn.(*ResultReader).readUntilRowDescription(0xc0056cecf8)
        github.com/jackc/pgconn/external/com_github_jackc_pgconn/pgconn.go:1529 +0x31
github.com/jackc/pgconn.(*PgConn).execExtendedSuffix(0xc0056cec60, {0xc0028d8000, 0xc004fbce00, 0x0}, 0xc0056cecf8)
        github.com/jackc/pgconn/external/com_github_jackc_pgconn/pgconn.go:1131 +0x24b
github.com/jackc/pgconn.(*PgConn).ExecPrepared(0xc0056cec60, {0x9f549c0, 0xc004fbce00}, {0xc000567760, 0xa}, {0x0, 0x0, 0x0}, {0x0, 0x0, ...}, ...)
        github.com/jackc/pgconn/external/com_github_jackc_pgconn/pgconn.go:1076 +0x196
github.com/jackc/pgx/v4.(*Conn).Query(0xc003934870, {0x9f549c0, 0xc004fbce00}, {0x88e5005, 0x19}, {0x0, 0x68, 0x78})
        github.com/jackc/pgx/v4/external/com_github_jackc_pgx_v4/conn.go:651 +0xd8d
github.com/cockroachdb/cockroach/pkg/ccl/changefeedccl.(*sinklessFeed).start(0xc0006dd900)
        github.com/cockroachdb/cockroach/pkg/ccl/changefeedccl/testfeed_test.go:201 +0x37e
github.com/cockroachdb/cockroach/pkg/ccl/changefeedccl.(*sinklessFeedFactory).Feed(0xc00235eb60, {0x88e5005, 0x19}, {0x0, 0x0, 0x0})
        github.com/cockroachdb/cockroach/pkg/ccl/changefeedccl/testfeed_test.go:88 +0x19d
github.com/cockroachdb/cockroach/pkg/ccl/changefeedccl.feed({0xa023850, 0xc00235eb60}, {0x9eaa780, 0xc0040aa780}, {0x88e5005, 0x19}, {0x0, 0x0, 0x0})
        github.com/cockroachdb/cockroach/pkg/ccl/changefeedccl/helpers_test.go:596 +0x98
github.com/cockroachdb/cockroach/pkg/ccl/changefeedccl.TestChangefeedTestTimesOut.func1(_, {0xc005d5fc70, {0xa055510, 0xc001435dc0}, {{0xc003e15890}, {0xc003e15890}, 0x0}, {{0x0, 0x0}, {0x0, ...}, ...}}, ...)
        github.com/cockroachdb/cockroach/pkg/ccl/changefeedccl/changefeed_test.go:701 +0xe6
github.com/cockroachdb/cockroach/pkg/ccl/changefeedccl.cdcTestNamed.func1(_, {{0xc005d5fc70, {0xa055510, 0xc001435dc0}, {{0xc003e15890}, {0xc003e15890}, 0x0}, {{0x0, 0x0}, {0x0, ...}, ...}}, ...}, ...)
        github.com/cockroachdb/cockroach/pkg/ccl/changefeedccl/helpers_test.go:820 +0x5d
github.com/cockroachdb/cockroach/pkg/ccl/changefeedccl.cdcTestNamedWithSystem.func1(0xc00235eb60)
        github.com/cockroachdb/cockroach/pkg/ccl/changefeedccl/helpers_test.go:848 +0x24d
testing.tRunner(0xc00235eb60, 0xc00089fdd0)
        GOROOT/src/testing/testing.go:1259 +0x102
created by testing.(*T).Run
        GOROOT/src/testing/testing.go:1306 +0x35a

Copy link
Contributor Author

Choose a reason for hiding this comment

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

But we never make it to next. Because Query never returns:

Ahh.. i see.

@miretskiy
Copy link
Contributor Author

Made some additional changes; added the test you wrote -- almost verbatim; seems to correctly
timeout in 100 runs... So, 🤞 ....

@miretskiy miretskiy force-pushed the flake branch 3 times, most recently from b69d2b8 to 0c8a129 Compare July 8, 2022 00:50
@miretskiy
Copy link
Contributor Author

cancelling sinkless feed proving to be more problematic; removed WithCancel logic and replaced it with direct call to connection cancellation instead. Let's see if that makes CI happy.

Add timeout to testfeed sink implementations to timeout
if no messages are received for too long.

Release Notes: None
@miretskiy
Copy link
Contributor Author

bors r+

@craig
Copy link
Contributor

craig bot commented Jul 8, 2022

Build succeeded:

@craig craig bot merged commit 4d8d0b5 into cockroachdb:master Jul 8, 2022
miretskiy pushed a commit to miretskiy/cockroach that referenced this pull request Jul 19, 2022
Fixes cockroachdb#84121

cockroachdb#84007 introduced a change to add a timeout
to test feed library to prevent flaky tests from hanging for a long
time.  This timeout lead to `TestChangefeedBackfillCheckpoint` test
to become flaky.  The main contributor of the slowness of that test
was the fact that the test processes 1000 messages (twice), and
the fact that a `webhook` sink and it's mock sink implementation
are very slow (50+ms per message).

The webhook sink, and mock webhook sink performance will be
addressed separately (cockroachdb#84676)

For now, marginally improve mock webhook sink performance
by detecting when messages become available directly, instead
of relying on resolved timestamps.  Also, significantly increase
the internal test timeout when reading many messages in a unit test.

While troubleshooting this issue, observed large number of
error messages `http: TLS handshake error from 127.0.0.1:34276: EOF`.
The problem is that the webhook sink specified an arbitrary, and
very small default timeout of 3 seconds.  The default in Go
library is 0 -- no timeout; and we should have this default
as well.  Fixes 75745

Release Notes: None
craig bot pushed a commit that referenced this pull request Jul 20, 2022
84572: obsservice: update README r=andreimatei a=dhartunian

update README to reference the `./dev` command
which includes the UI build.

Release note: None

84678: changefeedccl: De-flake TestChangefeedBackfillCheckpoint test. r=miretskiy a=miretskiy

Fixes #84121

#84007 introduced a change to add a timeout
to test feed library to prevent flaky tests from hanging for a long
time.  This timeout lead to `TestChangefeedBackfillCheckpoint` test
to become flaky.  The main contributor of the slowness of that test
was the fact that the test processes 1000 messages (twice), and
the fact that a `webhook` sink and it's mock sink implementation
are very slow (50+ms per message).

The webhook sink, and mock webhook sink performance will be
addressed separately (#84676)

For now, marginally improve mock webhook sink performance
by detecting when messages become available directly, instead
of relying on resolved timestamps.  Also, significantly increase
the internal test timeout when reading many messages in a unit test.

While troubleshooting this issue, observed large number of
error messages `http: TLS handshake error from 127.0.0.1:34276: EOF`.
The problem is that the webhook sink specified an arbitrary, and
very small default timeout of 3 seconds.  The default in Go
library is 0 -- no timeout; and we should have this default
as well.  Fixes #75745

Release Notes: None

84682: opt: fix crdb_internal.decode_plan_gist to work with unknown index r=rytaft a=rytaft

Release note (bug fix): `crdb_internal.decode_plan_gist` will no longer
produce an internal error when it is used to decode a plan gist for which
no schema information is available.

Co-authored-by: David Hartunian <davidh@cockroachlabs.com>
Co-authored-by: Yevgeniy Miretskiy <yevgeniy@cockroachlabs.com>
Co-authored-by: Rebecca Taft <becca@cockroachlabs.com>
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