Allow to define volume_priority in storage_configuration

[zvonand]

Allow to manually set volume priorities. Closes #31264.

Changelog category (leave one):

• Improvement

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

Allow to define volume_priority in storage_configuration.

Documentation entry for user-facing changes

• Documentation is written (mandatory for new features)

[ilejn]

Consider std::stable_sort instead of std::sort to avoid unpleasant surprises for users.
.. and, may be, to get rid of has_information_about_priority flag.

[ilejn]

The statement in doc "If two or more volumes have the same setting value, order among them is undefined" is not true.

To me it is enough to say that

• default - 0
• items with equal priorities are ordered by their position in config.

I am a bit worried that higher value actually means lower priority, it may be contrintuitive.

[zvonand]

The statement in doc "If two or more volumes have the same setting value, order among them is undefined" is not true.

Already fixed

[ilejn]

Already fixed

Hm ... not pushed yet?

[zvonand]

Hm ... not pushed yet?

A-ha, fixed one line and left another one :) Fixed.

[robot-clickhouse]

This is an automated comment for commit b4e582f with description of existing statuses. It's updated for the latest CI running

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

Successful checks

Check name	Description	Status
A Sync	There's no description for the check yet, please add it to tests/ci/ci_config.py:CHECK_DESCRIPTIONS	✅ success
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
ClickBench	Runs [ClickBench](https://github.com/ClickHouse/ClickBench/) with instant-attach table	✅ success
ClickHouse build check	Builds 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	✅ 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. 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 packages	Checks that the built packages are installable in a clear environment	✅ success
Mergeable Check	Checks if all other necessary checks are successful	✅ 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
SQLTest	There's no description for the check yet, please add it to tests/ci/ci_config.py:CHECK_DESCRIPTIONS	✅ success
SQLancer	Fuzzing tests that detect logical bugs with SQLancer tool	✅ success
Sqllogic	Run clickhouse on the sqllogic test set against sqlite and checks that all statements are passed	✅ 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
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	✅ success

Check name	Description	Status
CI running	A 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
Integration tests	The integration tests report. In parenthesis the package type is given, and in square brackets are the optional part/total tests	❌ failure

[zvonand]

• Stateless tests (release, analyzer):
  • 02901_parallel_replicas_rollup -- fixed in Flaky test 02901_parallel_replicas_rollup #59030
  • 02891_array_shingles fixed in Fix 02891_array_shingles with analyzer #58982
• Upgrade check (msan) :
  • not_processed should be empty -- Logical error: not_processed should be empty #58319

[zvonand]

Could anyone pls take a look at this PR? It's tiny, non-complicated, well-documented and the tests are fine :)

cc @alexey-milovidov (as you are the only one from ClickHouse Inc. in this thread 😃 )

[alexey-milovidov]

It will be much better if:

• we allow arbitrary unsigned numbers for priorities;
• we allow identical priorities;
• if the priority is identical, the order in the configuration matters;
• omitting the priority value is equivalent to priority zero.

[zvonand]

we allow arbitrary unsigned numbers for priorities

This is about a mismatch of internal representation vs what user specifies in config. E.g. if we specify priorities 1,3,5,7, they would be stored as 1,2,3,4 internally. Moreover, we will have 1,2,3,4 in output of storage_policies. Wouldn't that be confusing? However, I understand that changing this config may be tough as user will have to reset all priorities. If this internal vs config mismatch is ok, we can do it like this

omitting the priority value is equivalent to priority zero

I think it is a matter of taste. In both ways, it is easy to add a low-priority and high-priority volume.
However, the way it works now we can also easily add a volume in the middle, e.g. 2nd priority volume. With your idea, it would be harder

[ilejn]

Basically these restrictions were introduced in 47e75bd to avoid confusion
and to guarantee that volume_priority in config and in system.storage_policies are the same.
@alexey-milovidov , if you think that it does not make sense, we can return to initial approach.

[alexey-milovidov]

Ok.

[zvonand]

Investigating fails:

• Stateless tests (aarch64):
  • 02221_system_zookeeper_unrestricted_like -- Flaky test 02221_system_zookeeper_unrestricted_like #60190
• Stateless tests (tsan), Stateless tests (tsan, s3 storage):
  • 00984_parser_stack_overflow -- Flaky test 00984_parser_stack_overflow #60192
• Upgrade check (asan):
  • S3_ERROR No such key thrown looks like bursts of detached parts as broken-on-start #51781
• Stress test (tsan):
  Strange and unrelated, I can't identify what exactly went wrong, but definitely somewhere else, not in my changes:

  [ 10136 ] {f335ae06-8448-46f2-a339-a5072a0f943a} <Fatal> : Logical error: \'Unexpected return type from equals. Expected Nullable. Got UInt8\'.
  [ 3137 ] {} <Fatal> Application: Child process was terminated by signal 6.
  

  GDB log: https://s3.amazonaws.com/clickhouse-test-reports/58533/e06d1fea267c3e47f84f345641ff256beec7730f/stress_test__tsan_/gdb.log
• All Integration tests:
  • test_replicated_database/test.py::test_startup_without_zk -- Flaky test_replicated_database/test.py::test_startup_without_zk #60213
• Integration tests (release):
  • test_replicated_table_attach/test.py::test_startup_with_small_bg_pool_partitioned -- Flaky test test_replicated_table_attach/test.py::test_startup_with_small_bg_pool_partitioned #60217

[zvonand]

@alexey-milovidov updated according to your comment.
Also, looks like there are no related fails (plenty of them, but not related to my changes)

[alexey-milovidov]

Could you please merge with master?

[zvonand]

The only fails now are:

Integration tests (tsan) [3/6]:

• test_reload_clusters_config -- Flaky integration tests test_reload_clusters_config #60422

CC @alexey-milovidov