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.

Sign up for GitHub

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

Jump to bottom

roachtest: kv95limited-spans/enc=false/nodes=1/splt=0/seq/no-load-splitting failed #71262

Closed
cockroach-teamcity opened this issue Oct 7, 2021 · 1 comment · Fixed by #71280
Closed

roachtest: kv95limited-spans/enc=false/nodes=1/splt=0/seq/no-load-splitting failed #71262

cockroach-teamcity opened this issue Oct 7, 2021 · 1 comment · Fixed by #71280
Labels
branch-master Failures and bugs on the master branch. C-test-failure Broken test (automatically or manually discovered). O-roachtest O-robot Originated from a bot. release-blocker Indicates a release-blocker. Use with branch-release-2x.x label to denote which branch is blocked.

Comments

@cockroach-teamcity
Copy link
Member

roachtest.kv95limited-spans/enc=false/nodes=1/splt=0/seq/no-load-splitting failed with artifacts on master @ cc6296c24ddb048215dabe5cc41339f306db4f41:

		  |   587.0s        0          889.1          898.9     37.7     52.4     58.7     65.0 write
		  |   588.0s        0        16616.7        17033.8      1.2      6.8     13.6     27.3 span
		  |   588.0s        0          846.8          898.8     35.7     52.4     60.8     71.3 write
		  |   589.0s        0        16763.5        17033.3      1.2      6.6     13.1     25.2 span
		  |   589.0s        0          864.2          898.7     35.7     52.4     58.7     62.9 write
		  |   590.0s        0        16302.6        17032.1      1.2      7.1     14.7     30.4 span
		  |   590.0s        0          830.9          898.6     35.7     52.4     56.6     62.9 write
		  | _elapsed___errors__ops/sec(inst)___ops/sec(cum)__p50(ms)__p95(ms)__p99(ms)_pMax(ms)
		  |   591.0s        0        16409.9        17031.0      1.2      6.6     12.6     25.2 span
		  |   591.0s        0          872.0          898.6     35.7     50.3     58.7     67.1 write
		  |   592.0s        0        16392.4        17030.0      1.2      6.6     11.5     25.2 span
		  |   592.0s        0          893.0          898.6     35.7     50.3     58.7     67.1 write
		  |   593.0s        0        16438.3        17029.0      1.2      7.1     14.2     28.3 span
		  |   593.0s        0          838.1          898.5     35.7     50.3     60.8     83.9 write
		  |   594.0s        0        15788.1        17026.9      1.2      7.3     14.7     33.6 span
		  |   594.0s        0          828.0          898.3     37.7     52.4     60.8     71.3 write
		Wraps: (4) exit status 20
		Error types: (1) *withstack.withStack (2) *errutil.withPrefix (3) *cluster.WithCommandDetails (4) *exec.ExitError

	monitor.go:128,kv.go:137,kv.go:246,test_runner.go:777: monitor failure: monitor task failed: t.Fatal() was called
		(1) attached stack trace
		  -- stack trace:
		  | main.(*monitorImpl).WaitE
		  | 	/home/agent/work/.go/src/github.com/cockroachdb/cockroach/pkg/cmd/roachtest/monitor.go:116
		  | main.(*monitorImpl).Wait
		  | 	/home/agent/work/.go/src/github.com/cockroachdb/cockroach/pkg/cmd/roachtest/monitor.go:124
		  | github.com/cockroachdb/cockroach/pkg/cmd/roachtest/tests.registerKV.func2
		  | 	/home/agent/work/.go/src/github.com/cockroachdb/cockroach/pkg/cmd/roachtest/tests/kv.go:137
		  | github.com/cockroachdb/cockroach/pkg/cmd/roachtest/tests.registerKV.func3
		  | 	/home/agent/work/.go/src/github.com/cockroachdb/cockroach/pkg/cmd/roachtest/tests/kv.go:246
		  | main.(*testRunner).runTest.func2
		  | 	/home/agent/work/.go/src/github.com/cockroachdb/cockroach/pkg/cmd/roachtest/test_runner.go:777
		Wraps: (2) monitor failure
		Wraps: (3) attached stack trace
		  -- stack trace:
		  | main.(*monitorImpl).wait.func2
		  | 	/home/agent/work/.go/src/github.com/cockroachdb/cockroach/pkg/cmd/roachtest/monitor.go:172
		Wraps: (4) monitor task failed
		Wraps: (5) attached stack trace
		  -- stack trace:
		  | main.init
		  | 	/home/agent/work/.go/src/github.com/cockroachdb/cockroach/pkg/cmd/roachtest/monitor.go:81
		  | runtime.doInit
		  | 	/usr/local/go/src/runtime/proc.go:6309
		  | runtime.main
		  | 	/usr/local/go/src/runtime/proc.go:208
		  | runtime.goexit
		  | 	/usr/local/go/src/runtime/asm_amd64.s:1371
		Wraps: (6) t.Fatal() was called
		Error types: (1) *withstack.withStack (2) *errutil.withPrefix (3) *withstack.withStack (4) *errutil.withPrefix (5) *withstack.withStack (6) *errutil.leafError
Reproduce

See: roachtest README

/cc @cockroachdb/kv-triage

This test on roachdash | Improve this report!
@cockroach-teamcity cockroach-teamcity added branch-master Failures and bugs on the master branch. C-test-failure Broken test (automatically or manually discovered). O-roachtest O-robot Originated from a bot. release-blocker Indicates a release-blocker. Use with branch-release-2x.x label to denote which branch is blocked. labels Oct 7, 2021
@stevendanna stevendanna mentioned this issue Oct 7, 2021
Merged
craig bot pushed a commit that referenced this issue Oct 7, 2021
@tukeJonny @rickystewart @stevendanna
Merge #71083 #71219 #71220 #71227 #71265 #71280
5b1bd56 
71083: opt, parser: add support for nulls first, nulls last r=nehageorge a=nehageorge

Previously, we only supported the default Postgres ordering for
ORDER BY. That is, NULLS come first for ascending order and NULLS
come last for descending order. Postgres also supports the keywords
"NULLS FIRST" and "NULLS LAST", which explicitly places all the NULLS
first or last, respectively. This PR adds support for this non-default
ordering.

Informs #6224.

Release note (sql change): NULLS FIRST and NULLS LAST specifiers now
supported for ORDER BY. 

71219: sql: UserHasAdminRole should return true for the admin role itself r=rafiss a=tukeJonny

This PR makes `planner.UserHasAdminRole` returns true on the `admin` role for improving introspection.

This PR had checked by running `make testbaselogic` .

Fixes #70779 

Release note: None

71220: dev: update how we test for presence of `--dev` config in `doctor` r=rail a=rickystewart

Some people are running Linux and can therefore use the `crosslinux`
config on a day-to-day basis. This is OK and `dev doctor` should allow
this. Instead of failing the `doctor` check if `--config dev` isn't set,
we instead test whether the user has stamping enabled on a `bazel
build`. The built `cockroach` binary won't start if stamping is disabled
and both the `dev` and `crosslinux` configs enable stamping.

Release note: None

71227: bazel: avoid `read-only file system` error when calling into `autom4te` r=rail a=rickystewart

See 3f50217 for more context. Many
people are encountering the error described in that commit and it's not
feasible to ask everyone to install an `autom4te` shim. Instead we
include the shim in the build definition. We also remove it from the
builder image to demonstrate that the fix works.

Closes #70986.

Release note: None

71265: changefeedccl: don't warn when Timestamp() is called on flush event r=miretskiy a=stevendanna

When investigating a recent test failure, I noticed the following log
messages:

    ccl/changefeedccl/kvevent/event.go:153 ⋮ [-] 1830  setting empty timestamp for unknown event type
    ccl/changefeedccl/kvevent/event.go:153 ⋮ [-] 1831  setting empty timestamp for unknown event type
    ccl/changefeedccl/kvevent/event.go:153 ⋮ [-] 1832  setting empty timestamp for unknown event type
    ccl/changefeedccl/kvevent/event.go:153 ⋮ [-] 1833  setting empty timestamp for unknown event type
    ccl/changefeedccl/kvevent/event.go:153 ⋮ [-] 1834  setting empty timestamp for unknown event type

One possibility is that these were generated by our new Flush event,
which is an in-band signal between the kvFeed and the changeAggregator
and has no associated timestamp.

This PR adds handling for TypeFlush to cut down on the log spam.

Not however, this raises the question of why we were hitting the
condition that forces a flush so often.

One possibility is that we were unable to allocate resources to add
items to the queue because of some other user of our memory monitor
and thus while we kept trying to flush, these flushes had no
effect. We may consider tracking whether there has been a successful
resource acquisition since the last flush and only resending a flush
if there has been.

Release note: None

71280: sql: clear (rather than emptying) bound account r=[Azhng,miretskiy] a=stevendanna

Empty() leaves some number of bytes in the bound account as
reserved. When we abandon a Container, those "reserved" bytes were
never released back to the parent. I believe we currently abandon one
Container at the end of each explicit transaction.

Fixes #71262
Fixes #71263
Fixes #71264
Fixes #65279

We may want to provide two different methods. One that calls Empty() and one that
calls Clear() in case there are users of this who really need to keep their reserved bytes.

Release note: None

Co-authored-by: Neha George <neha.george@cockroachlabs.com>
Co-authored-by: tukeJonny <ne250143@yahoo.co.jp>
Co-authored-by: Ricky Stewart <ricky@cockroachlabs.com>
Co-authored-by: Steven Danna <danna@cockroachlabs.com>
@tbg
Copy link
Member

tbg commented Oct 7, 2021

#71280

@tbg tbg closed this as completed Oct 7, 2021
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
branch-master Failures and bugs on the master branch. C-test-failure Broken test (automatically or manually discovered). O-roachtest O-robot Originated from a bot. release-blocker Indicates a release-blocker. Use with branch-release-2x.x label to denote which branch is blocked.
Projects
None yet
Milestone
No milestone
Development

Successfully merging a pull request may close this issue.

sql: clear (rather than emptying) bound account stevendanna/cockroach
2 participants
@tbg @cockroach-teamcity