Add replication lag and recovery time metrics by mstetsyuk · Pull Request #66703 · ClickHouse/ClickHouse

mstetsyuk · 2024-07-18T10:34:04Z

Changelog category (leave one):

Improvement

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

Add replication lag and recovery time metrics

Documentation entry for user-facing changes

Documentation is written (mandatory for new features)

Information about CI checks: https://clickhouse.com/docs/en/development/continuous-integration/

CI Settings (Only check the boxes if you know what you are doing):

Allow: All Required Checks
Allow: Stateless tests
Allow: Stateful tests
Allow: Integration Tests
Allow: Performance tests
Allow: All Builds
Allow: batch 1, 2 for multi-batch jobs
Allow: batch 3, 4, 5, 6 for multi-batch jobs

Exclude: Style check
Exclude: Fast test
Exclude: All with ASAN
Exclude: All with TSAN, MSAN, UBSAN, Coverage
Exclude: All with aarch64, release, debug

robot-ch-test-poll2 · 2024-07-18T15:25:49Z

This is an automated comment for commit 01ca36c with description of existing statuses. It's updated for the latest CI running

❌ Click here to open a full report in a separate page

Check name	Description	Status
Upgrade check	Runs 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	❌ failure

Successful checks

Check name	Description	Status
AST fuzzer	Runs randomly generated queries to catch program errors. The build type is optionally given in parenthesis. If it fails, ask a maintainer for help	✅ success
Builds	There's no description for the check yet, please add it to tests/ci/ci_config.py:CHECK_DESCRIPTIONS	✅ success
ClickBench	Runs [ClickBench](https://github.com/ClickHouse/ClickBench/) with instant-attach table	✅ success
Compatibility check	Checks that clickhouse binary runs on distributions with old libc versions. If it fails, ask a maintainer for help	✅ success
Docker keeper image	The check to build and optionally push the mentioned image to docker hub	✅ success
Docker server image	The check to build and optionally push the mentioned image to docker hub	✅ success
Docs check	Builds and tests the documentation	✅ success
Fast test	Normally 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 tests	Checks 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. Integration 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 packages	Checks that the built packages are installable in a clear environment	✅ success
Integration tests	The integration tests report. In parenthesis the package type is given, and in square brackets are the optional part/total tests	✅ success
Performance Comparison	Measure changes in query performance. The performance test report is described in detail here. In square brackets are the optional part/total tests	✅ success
Stateful tests	Runs stateful functional tests for ClickHouse binaries built in various configurations -- release, debug, with sanitizers, etc	✅ success
Stateless tests	Runs stateless functional tests for ClickHouse binaries built in various configurations -- release, debug, with sanitizers, etc	✅ success
Stress test	Runs stateless functional tests concurrently from several clients to detect concurrency-related errors	✅ success
Style check	Runs a set of checks to keep the code style clean. If some of tests failed, see the related log from the report	✅ success
Unit tests	Runs the unit tests for different release types	✅ success

mstetsyuk · 2024-07-24T09:57:27Z

Failing CI:

02911_backup_restore_keeper_map is unrelated to this PR and is broken (according to furrious bober)

tavplubix · 2024-07-24T23:10:45Z

src/Databases/DatabaseReplicated.cpp

+                UInt32 log_ptr = parse<UInt32>(zookeeper->get(fs::path(zookeeper_path) / "replicas" / full_name / "log_ptr"));
+                bool is_active = zookeeper->exists(fs::path(zookeeper_path) / "replicas" / full_name / "active");


It will work, but it will drastically slow down selects from system.clusters (especially for clusters having a lot of replicas). For each replica, it makes 2 sequential zk requests, so it's 2 * num_of_replicas RTT for each tryGetReplicasInfo call. Let's put all paths in a vector, make one mutli-get request (exists is basically a get request + a check for the error code (ZOK/ZNONODE)), and process the list of responses/error codes

tavplubix · 2024-07-24T23:22:47Z

src/Databases/DatabaseReplicatedWorker.cpp

 bool DatabaseReplicatedDDLWorker::initializeMainThread()
 {
+    initialization_duration_timer.restart();
+    initializing.store(true, std::memory_order_release);


This is not a performance-critical part of code, so it's okay to use the default std::memory_order_seq_cst. Using anything besides std::memory_order_seq_cst and std::memory_order_relaxed usually requires a comment explaining why it will work correctly, and I would prefer not to think about this when it's not required for performance reasons

See also https://github.com/ClickHouse/ClickHouse/pull/67312/files :)

tavplubix · 2024-07-24T23:35:17Z

src/Databases/DatabaseReplicatedWorker.cpp


+UInt64 DatabaseReplicatedDDLWorker::getCurrentInitializationDurationMs() const
+{
+    return initializing.load(std::memory_order_acquire) ? initialization_duration_timer.elapsedMilliseconds() : 0;


There's still a (theoretical) race condition:

thread X checks initializing here, it's true, but kernel decides to pause this thread for a few seconds for some reason

the DDLWorker thread finishes the initialization successfully, but instantly gets an unexpected error and calls initializeMainThread again

thread X wakes up and reads start_ns inside initialization_duration_timer.elapsedMilliseconds() at the same time when the DDLWorker thread resets it in initialization_duration_timer.restart()

tavplubix · 2024-07-24T23:39:13Z

tests/integration/test_recovery_time_metric/test.py

+node = cluster.add_instance(
+    "node",
+    main_configs=["configs/config.xml"],
+    with_zookeeper=True,


with_zookeeper is redundant since there's embedded Keeper enabled in the server config

tavplubix · 2024-07-24T23:42:06Z

tests/integration/test_recovery_time_metric/test.py

+    node.query("CREATE TABLE rdb.t (x UInt32) ENGINE = MergeTree ORDER BY x;")
+    node.exec_in_container(["bash", "-c", "rm /var/lib/clickhouse/metadata/rdb/t.sql"])
+    node.restart_clickhouse()
+    assert node.query("SELECT any(recovery_time) FROM system.clusters;") != "0\n"


"NULL\n" != "0\n" is true, so the assert will not fail if all recovery times are NULL

alexey-milovidov · 2024-08-03T16:47:39Z

Reverted. Please resubmit.

robot-ch-test-poll1 added the pr-improvement Pull request with some product improvements label Jul 18, 2024

mstetsyuk force-pushed the add-replication-lag-and-recovery-time-metrics branch from 6df4b4d to 1ad9765 Compare July 18, 2024 16:36

add replication lag and recovery time metrics

7fc8ee7

mstetsyuk force-pushed the add-replication-lag-and-recovery-time-metrics branch from 1ad9765 to 7fc8ee7 Compare July 18, 2024 16:38

tavplubix self-assigned this Jul 18, 2024

robot-clickhouse and others added 9 commits July 18, 2024 16:45

Automatic style fix

a6d4db3

test replication lag metric

3f66b39

small fix

245626e

fix tests

79ef630

fix tests

c2ac132

Automatic style fix

d040e43

fix 01293_show_clusters stateless test

5fa2db8

fxs

7f5c58f

use atomic to avoid data race

932033f

mstetsyuk marked this pull request as ready for review July 24, 2024 09:57

mstetsyuk requested a review from tavplubix July 24, 2024 11:37

tavplubix reviewed Jul 24, 2024

View reviewed changes

fxs

aa26291

tavplubix approved these changes Jul 30, 2024

View reviewed changes

mstetsyuk added 4 commits July 30, 2024 21:57

fxs

8d2b804

fix style

e664a14

make it possible to rerun test_recovery_time_metric multiple times

27f4f46

empty

01ca36c

mstetsyuk added this pull request to the merge queue Aug 2, 2024

Merged via the queue into master with commit f1df59d Aug 2, 2024

mstetsyuk deleted the add-replication-lag-and-recovery-time-metrics branch August 2, 2024 15:52

robot-ch-test-poll added the pr-synced-to-cloud The PR is synced to the cloud repo label Aug 2, 2024

This was referenced Aug 3, 2024

Revert "Add replication lag and recovery time metrics" #67731

Merged

DatabaseReplicated: segmentation fault in SHOW CLUSTERS #67730

Closed

Test 01293_show_clusters has failed #67732

Closed

mstetsyuk restored the add-replication-lag-and-recovery-time-metrics branch August 6, 2024 10:46

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Add replication lag and recovery time metrics#66703

Add replication lag and recovery time metrics#66703
mstetsyuk merged 15 commits intomasterfrom
add-replication-lag-and-recovery-time-metrics

mstetsyuk commented Jul 18, 2024

Uh oh!

robot-ch-test-poll2 commented Jul 18, 2024 •

edited by robot-ch-test-poll

Loading

Uh oh!

mstetsyuk commented Jul 24, 2024

Uh oh!

tavplubix Jul 24, 2024

Uh oh!

tavplubix Jul 24, 2024

Uh oh!

tavplubix Jul 30, 2024

Uh oh!

tavplubix Jul 24, 2024

Uh oh!

tavplubix Jul 24, 2024

Uh oh!

tavplubix Jul 24, 2024

Uh oh!

alexey-milovidov commented Aug 3, 2024

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

7 participants

		UInt32 log_ptr = parse<UInt32>(zookeeper->get(fs::path(zookeeper_path) / "replicas" / full_name / "log_ptr"));
		bool is_active = zookeeper->exists(fs::path(zookeeper_path) / "replicas" / full_name / "active");

Conversation

mstetsyuk commented Jul 18, 2024

Changelog category (leave one):

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

Documentation entry for user-facing changes

CI Settings (Only check the boxes if you know what you are doing):

Uh oh!

robot-ch-test-poll2 commented Jul 18, 2024 • edited by robot-ch-test-poll Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

mstetsyuk commented Jul 24, 2024

Uh oh!

tavplubix Jul 24, 2024

Choose a reason for hiding this comment

Uh oh!

tavplubix Jul 24, 2024

Choose a reason for hiding this comment

Uh oh!

tavplubix Jul 30, 2024

Choose a reason for hiding this comment

Uh oh!

tavplubix Jul 24, 2024

Choose a reason for hiding this comment

Uh oh!

tavplubix Jul 24, 2024

Choose a reason for hiding this comment

Uh oh!

tavplubix Jul 24, 2024

Choose a reason for hiding this comment

Uh oh!

alexey-milovidov commented Aug 3, 2024

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

7 participants

robot-ch-test-poll2 commented Jul 18, 2024 •

edited by robot-ch-test-poll

Loading