Skip to content

Use CLOCK_MONOTONIC_RAW over CLOCK_MONOTONIC on Linux (fixes non monotonic clock)#49819

Merged
alexey-milovidov merged 1 commit intoClickHouse:masterfrom
azat:fix-clock
May 12, 2023
Merged

Use CLOCK_MONOTONIC_RAW over CLOCK_MONOTONIC on Linux (fixes non monotonic clock)#49819
alexey-milovidov merged 1 commit intoClickHouse:masterfrom
azat:fix-clock

Conversation

@azat
Copy link
Copy Markdown
Member

@azat azat commented May 12, 2023

Changelog category (leave one):

  • Not for changelog (changelog entry is not required)

Changelog entry (a user-readable short description of the changes that goes to CHANGELOG.md):

Use CLOCK_MONOTONIC_RAW over CLOCK_MONOTONIC on Linux (fixes non monotonic clock)

There is no bug in Linux, the issue that CLOCK_MONOTONIC returns values less then previous calls likely happens due to adjtime(3) (NTP), since CLOCK_MONOTONIC is affected by it, and I've seen lots of slight time modifications due to NTP on the servers. And even on my desktop (I also have NTP enabled):

CLOCK_MONOTONIC:     189292.803 (2 days +  4h 34m 52s)
CLOCK_MONOTONIC_RAW: 189290.016 (2 days +  4h 34m 50s)

However on Linux there is CLOCK_MONOTONIC_RAW, it is similar to CLOCK_MONOTONIC, but does not affected by the adjtime(3).

About performance, it is the same:

CLOCK_MONOTONIC 10e6:     real=0m0.191s user=0m0.190s sys=0m0.000s
CLOCK_MONOTONIC_RAW 10e6: real=0m0.191s user=0m0.191s sys=0m0.000s

Ops/s:

  • AMD Threadripper: 52.3e6
  • Xeon Silver 4216 2.10: 46.5e6

Fixes: c5d631c
Fixes: #29811 (cc @tavplubix)
Fixes: #29154

@robot-ch-test-poll robot-ch-test-poll added the pr-not-for-changelog This PR should not be mentioned in the changelog label May 12, 2023
@robot-ch-test-poll
Copy link
Copy Markdown
Contributor

robot-ch-test-poll commented May 12, 2023 •
edited by robot-ch-test-poll1
Loading

This is an automated comment for commit fa2074e with description of existing statuses. It's updated for the latest CI running
The full report is available here
The overall status of the commit is 🔴 failure

Check nameDescriptionStatus
AST fuzzerRuns randomly generated queries to catch program errors. The build type is optionally given in parenthesis. If it fails, ask a maintainer for help🟢 success
CI runningA meta-check that indicates the running CI. Normally, it's in success or pending state. The failed status indicates some problems with the PR🟡 pending
ClickHouse build checkBuilds ClickHouse in various configurations for use in further steps. You have to fix the builds that fail. Build logs often has enough information to fix the error, but you might have to reproduce the failure locally. The cmake options can be found in the build log, grepping for cmake. Use these options and follow the general build process🔴 failure
Compatibility checkChecks that clickhouse binary runs on distributions with old libc versions. If it fails, ask a maintainer for help🟢 success
Docker image for serversThe check to build and optionally push the mentioned image to docker hub🟢 success
Fast testNormally this is the first check that is ran for a PR. It builds ClickHouse and runs most of stateless functional tests, omitting some. If it fails, further checks are not started until it is fixed. Look at the report to see which tests fail, then reproduce the failure locally as described here🟢 success
Flaky testsChecks if new added or modified tests are flaky by running them repeatedly, in parallel, with more randomization. Functional tests are run 100 times with address sanitizer, and additional randomization of thread scheduling. Integrational tests are run up to 10 times. If at least once a new test has failed, or was too long, this check will be red. We don't allow flaky tests, read the doc🟢 success
Install packagesChecks that the built packages are installable in a clear environment🟢 success
Integration testsThe integration tests report. In parenthesis the package type is given, and in square brackets are the optional part/total tests🟢 success
Mergeable CheckChecks if all other necessary checks are successful🔴 failure
Performance ComparisonMeasure changes in query performance. The performance test report is described in detail here. In square brackets are the optional part/total tests🟢 success
Push to DockerhubThe check for building and pushing the CI related docker images to docker hub🟢 success
SQLancerFuzzing tests that detect logical bugs with SQLancer tool🟢 success
SqllogicRun clickhouse on the sqllogic test set against sqlite and checks that all statements are passed🟢 success
Stateful testsRuns stateful functional tests for ClickHouse binaries built in various configurations -- release, debug, with sanitizers, etc🟢 success
Stateless testsRuns stateless functional tests for ClickHouse binaries built in various configurations -- release, debug, with sanitizers, etc🟢 success
Stress testRuns stateless functional tests concurrently from several clients to detect concurrency-related errors🟢 success
Style CheckRuns a set of checks to keep the code style clean. If some of tests failed, see the related log from the report🟢 success
Unit testsRuns the unit tests for different release types🟢 success
Upgrade checkRuns stress tests on server version from last release and then tries to upgrade it to the version from the PR. It checks if the new server can successfully startup without any errors, crashes or sanitizer asserts🟢 success

@azat
Use CLOCK_MONOTONIC_RAW over CLOCK_MONOTONIC on Linux (fixes non mono…
fa2074e 
…tonic clock)

There is no bug in Linux, the issue that CLOCK_MONOTONIC returns values
less then previous calls likely happens due to adjtime(3) (NTP), since
CLOCK_MONOTONIC is affected by it, and I've seen lots of slight time
modifications due to NTP on the servers. And even on my desktop (I also
have NTP enabled):

    CLOCK_MONOTONIC:     189292.803 (2 days +  4h 34m 52s)
    CLOCK_MONOTONIC_RAW: 189290.016 (2 days +  4h 34m 50s)

However on Linux there is CLOCK_MONOTONIC_RAW, it is similar to
CLOCK_MONOTONIC, but does not affected by the adjtime(3).

About performance, it is the same:

    CLOCK_MONOTONIC 10e6:     real=0m0.191s user=0m0.190s sys=0m0.000s
    CLOCK_MONOTONIC_RAW 10e6: real=0m0.191s user=0m0.191s sys=0m0.000s

Ops/s:
- AMD Threadripper: 52.3e6
- Xeon Silver 4216 2.10: 46.5e6

Fixes: c5d631c
Fixes: ClickHouse#29811 (cc @tavplubix)
Signed-off-by: Azat Khuzhin <[email protected]>
@azat
Copy link
Copy Markdown
Member Author

azat commented May 12, 2023 •
edited
Loading

ClickHouse special build check — 7/8 artifact groups are OK

  • clang-15-darwin-aarch64
May 12 15:37:39   The CMAKE_C_COMPILER:
May 12 15:37:39 
May 12 15:37:39     clang-15
May 12 15:37:39 
May 12 15:37:39   is not a full path and was not found in the PATH.
May 12 15:37:39 
May 12 15:37:39   Tell CMake where to find the compiler by setting either the environment
May 12 15:37:39   variable "CC" or the CMake cache entry CMAKE_C_COMPILER to the full path to
May 12 15:37:39   the compiler, or to the compiler name if it is in the PATH.
May 12 15:37:39

Images was switched to clang-16 already.

@alexey-milovidov alexey-milovidov self-assigned this May 12, 2023
@alexey-milovidov alexey-milovidov merged commit 12be14b into ClickHouse:master May 12, 2023
@azat azat deleted the fix-clock branch May 13, 2023 05:52
@tavplubix
Copy link
Copy Markdown
Member

tavplubix commented May 15, 2023

There is no bug in Linux, the issue that CLOCK_MONOTONIC returns values less then previous calls likely happens due to adjtime(3) (NTP), since CLOCK_MONOTONIC is affected by it, and I've seen lots of slight time modifications due to NTP on the servers.

CLOCK_MONOTONIC is for Clock that cannot be set and represents monotonic time since some unspecified starting point., it should never go back. Yes, it's affected by adjtime, but adjtime speeds up or slows down the clock, it should not set the clock to a smaller value. So it's clearly a bug (probably related to that speed up/slow down logic, so there's a good chance that CLOCK_MONOTONIC_RAW does not have a similar bug).

@azat azat mentioned this pull request Jul 10, 2024
Closed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@alexey-milovidov alexey-milovidov alexey-milovidov approved these changes

Assignees

@alexey-milovidov alexey-milovidov

Labels

pr-not-for-changelog This PR should not be mentioned in the changelog

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

Timeout exceeded: elapsed 18446744073.709553 seconds

4 participants

@azat @robot-ch-test-poll @tavplubix @alexey-milovidov