Backup/Restore for KeeperMap tables

[antonio2368]

Changelog category (leave one):

• Improvement

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

Add support for backing up and restoring tables using KeeperMap engine.

cc @vitlibar there are tests missing and some polishing is left but created a draft PR so you can take a look while I'm still working on it

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/

[robot-ch-test-poll]

This is an automated comment for commit 9bcedf3 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
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
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 image for servers	The check to build and optionally push the mentioned image to docker hub	✅ 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
Integration tests	The integration tests report. In parenthesis the package type is given, and in square brackets are the optional part/total tests	✅ 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
Push to Dockerhub	The check for building and pushing the CI related docker images to docker hub	✅ 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
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
Stateless tests	Runs stateless functional tests for ClickHouse binaries built in various configurations -- release, debug, with sanitizers, etc	❌ failure

[vitlibar]

This line is not necessary anymore.

[vitlibar]

I like this approach. And it can be faster too because instead of copying the same file again to another destination you just copy it inside a backup.

[vitlibar]

encrypted_by_disk should be copied too:

reference.encrypted_by_disk = target_info->encrypted_by_disk;

[vitlibar]

We can use this copy-inside-backup approach for other copies too if we insert here something like:

SizeAndChecksum size_and_checksum{info.size, info.checksum};
auto [it, inserted] = data_file_index_by_checksum.emplace(size_and_checksum, i);
if (inserted)
{
    info.data_file_name = info.file_name;
    info.data_file_index = i;
}
else
{
    /// specify that info must not be written in the usual way ...
    info.data_file_name.clear();
    info.data_file_index = -1;

    /// ... because target_info will be copied to its place instead
    auto & target_info = it->second;
    target_info->reference_sources.push_back(info.file_name);
}

[antonio2368]

so we are okay with using size and checksum in plain backups as unique id?

[vitlibar]

A plain backup means every file is stored in its place: there can files of the same contents; empty files are stored too; and references to the base backup are not used. So a plain backup is about how it must look on Disk or S3 when it's completed. It's not about how we should write it - we can write it any way we like if the result is the same. I mean there is no difference in the result whether we

copy SRC TO DEST[0], DEST[1], ... DEST[N]

or

copy SRC TO DEST[0], then
copy DEST[0] to DEST[1], ... DEST[N]

[antonio2368]

but there is a difference if we use size and checksum
if files from different backup entries match in size and checksum this will produce a different result

[vitlibar]

if files from different backup entries match in size and checksum this will produce a different result

I didn't quite understand. Do you mean that two files of the same size but different by contents can have the same checksum?

[antonio2368]

yes

[antonio2368]

I mean chances are almost nonexistent for a collision to happen, especially with 128bits we use for checksum but just asking if we are okay with that.

[vitlibar]

The formula is (https://en.wikipedia.org/wiki/Birthday_attack)

p(n, H) = 1-exp(-n^2 / 2 / H)

So if we used 128-bit hashes without size, for a backup with 1000000 files we would have the probability like 10^-27 that two hashes match. With sizes the probability must be even smaller.

[vitlibar]

Though that optimization is not very crucial anyway.

[antonio2368]

let me do this in a different PR so we are not blocking backups of KeeperMap and then we can discuss if we are okay with the solution.

[rschu1ze]

Broke my local build:

/home/ubuntu/repo/ClickHouse/src/Backups/BackupIO_S3.cpp:255:5: error: no matching function for call to 'copyS3File'
  255 |     copyS3File(
      |     ^~~~~~~~~~
/home/ubuntu/repo/ClickHouse/src/IO/S3/copyS3File.h:31:6: note: candidate function not viable: no known conversion from 'std::shared_ptr<S3::Client>' to 'const String' (aka 'const basic_string<char, char_traits<char>, allocator<char>>') for 2nd argument
   33 | void copyS3File(
      |      ^
   34 |     const std::shared_ptr<const S3::Client> & s3_client,
   35 |     const String & src_bucket,
      |     ~~~~~~~~~~~~~~~~~~~~~~~~~
1 error generated.
[10301/12493] Building CXX object src/CMakeFiles/dbms.dir/Functions/CastOverloadResolver.cpp.o
ninja: build stopped: subcommand failed.

Fix: #56974

[antonio2368]

Actually it's this PR #56314 which modifies copyS3File
It didn't have the latest master so it didn't catch the new code that calls copyS3File