Skip to content

Comments

Add mariadb 11.8#288

Merged
phracek merged 10 commits intosclorg:masterfrom
ndavidova:master
Sep 12, 2025
Merged

Add mariadb 11.8#288
phracek merged 10 commits intosclorg:masterfrom
ndavidova:master

Conversation

@ndavidova
Copy link
Contributor

@ndavidova ndavidova commented Aug 21, 2025

No description provided.

@github-actions
Copy link

github-actions bot commented Aug 21, 2025

Pull Request validation

Failed

🔴 Failed or pending statuses:

  • Testing Farm - RHEL9 - 10.11[pending]
  • Testing Farm - RHEL8 - 10.11[pending]
  • Testing Farm - RHEL8 - 10.3[pending]
  • Testing Farm - RHEL9 - 10.5[pending]

Success

🟢 Review - Reviewed by a member
🟢 Approval - Changes were approved

@phracek
Copy link
Member

phracek commented Aug 21, 2025

[test]

@github-actions
Copy link

github-actions bot commented Aug 21, 2025

Testing Farm results

namecomposearchstatusstarted (UTC)timelogs
CentOS Stream 10 - 11.8CentOS-Stream-10x86_64❌ error21.08.2025 13:37:326min 24stest pipeline
CentOS Stream 9 - 11.8CentOS-Stream-9x86_64❌ error21.08.2025 13:37:407min 3stest pipeline
CentOS Stream 9 - 10.5CentOS-Stream-9x86_64✅ passed11.09.2025 13:37:4410min 18stest pipeline
CentOS Stream 9 - 10.11CentOS-Stream-9x86_64✅ passed12.09.2025 09:37:2411min 24stest pipeline
CentOS Stream 10 - 10.11CentOS-Stream-10x86_64✅ passed12.09.2025 09:37:2310min 1stest pipeline
Fedora - 11.8Fedora-latestx86_64✅ passed12.09.2025 09:37:299min 28stest pipeline
RHEL10 - 11.8RHEL-10-Nightlyx86_64❌ error21.08.2025 13:37:3411min 58stest pipeline
RHEL8 - 11.8RHEL-8.10.0-Nightlyx86_64❌ error21.08.2025 13:37:3412min 34stest pipeline
RHEL9 - 11.8RHEL-9.6.0-Nightlyx86_64❌ error21.08.2025 13:37:4214min 36stest pipeline
RHEL8 - 10.5RHEL-8.10.0-Nightlyx86_64✅ passed12.09.2025 09:37:2417min 4stest pipeline
RHEL10 - 10.11RHEL-10-Nightlyx86_64✅ passed12.09.2025 09:37:2814min 12stest pipeline
RHEL9 - 10.11RHEL-9.6.0-Nightlyx86_64✅ passed12.09.2025 09:37:2120min 58stest pipeline
RHEL8 - 10.3RHEL-8.10.0-Nightlyx86_64✅ passed12.09.2025 09:37:2419min 3stest pipeline
RHEL9 - 10.5RHEL-9.6.0-Nightlyx86_64✅ passed11.09.2025 13:37:4121min 5stest pipeline
Fedora - 10.11Fedora-latestx86_64✅ passed12.09.2025 09:37:219min 3stest pipeline
RHEL8 - 10.11RHEL-8.10.0-Nightlyx86_64✅ passed12.09.2025 09:37:2118min 44stest pipeline

Copy link
Member

@phracek phracek left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks for this pull request.

It looks good. Please add to folder 11.8 exclude files that are not build and not tested.
Like .exclude-rhel9 . Only Fedora is going to be build and tested for now.

@phracek
Copy link
Member

phracek commented Aug 25, 2025

Fedora 11.8 has failed for this reason:

-----------------------------------------------
Running test run_change_password_test (starting at 2025-08-21 13:44:29+00:00) ... 
-----------------------------------------------
    Created container 6ef44a0eea6e40bf3bcdbd3d54fd0aaf36a336ff06ac7ad9fe1264215db6bcc7
  Testing MySQL connection to 10.88.0.90...
    Trying to connect.... FAIL
  Giving up: Failed to connect.
Error: no such object: "testpass1"
  Container finised with exit code .
Logs:
=> sourcing 20-validate-variables.sh ...
=> sourcing 25-validate-replication-variables.sh ...
=> sourcing 30-base-config.sh ...
---> 13:44:29     Processing basic MySQL configuration files ...
=> sourcing 60-replication-config.sh ...
=> sourcing 70-s2i-config.sh ...
---> 13:44:29     Processing additional arbitrary  MySQL configuration provided by s2i ...
=> sourcing 40-paas.cnf ...
=> sourcing 50-my-tuning.cnf ...
---> 13:44:29     Initializing database ...
---> 13:44:29     Running mysql_install_db ...
/usr/bin/mysql_install_db: Deprecated program name. It will be removed in a future release, use 'mariadb-install-db' instead


PLEASE REMEMBER TO SET A PASSWORD FOR THE MariaDB root USER !
To do so, start the server, then issue the following command:

'/usr/bin/mariadb-secure-installation'

which will also give you the option of removing the test
databases and anonymous user created by default.  This is
strongly recommended for production servers.

See the MariaDB Knowledgebase at https://mariadb.com/kb

Please report any problems at https://mariadb.org/jira

The latest information about MariaDB is available at https://mariadb.org/.

Consider joining MariaDB's strong and vibrant community:
https://mariadb.org/get-involved/

---> 13:44:30     Starting MySQL server with disabled networking ...
---> 13:44:30     Waiting for MySQL to start ...
/usr/libexec/mysqld: Deprecated program name. It will be removed in a future release, use '/usr/libexec/mariadbd' instead
2025-08-21 13:44:30 0 [Note] Starting MariaDB 11.8.3-MariaDB source revision b565b3e7e041c480fd8013e3beec6b99544d6bf8 server_uid zw0kt7UZqSetyyfmKEbz/i3k+LY= as process 58
2025-08-21 13:44:30 0 [ERROR] InnoDB: innodb_page_size=16384 requires innodb_buffer_pool_size >= 6 MiB current 5 MiB
2025-08-21 13:44:30 0 [ERROR] Plugin 'InnoDB' registration as a STORAGE ENGINE failed.
2025-08-21 13:44:30 0 [Note] Plugin 'FEEDBACK' is disabled.
2025-08-21 13:44:30 0 [Note] Plugin 'wsrep-provider' is disabled.
2025-08-21 13:44:30 0 [ERROR] Unknown/unsupported storage engine: InnoDB
2025-08-21 13:44:30 0 [ERROR] Aborting
Test for image 'quay.io/fedora/mariadb-118:11.8' FAILED (exit code: 1)
    Created container 98dfb064f82c5dd9498388cc266072c050e86daedefef0707b706a38f7e60c1d
  Testing MySQL connection to 10.88.0.92...
    Trying to connect.... FAIL
  Giving up: Failed to connect.
Error: no such object: "testpass2"
  Container finised with exit code .
Logs:
=> sourcing 20-validate-variables.sh ...
=> sourcing 25-validate-replication-variables.sh ...
=> sourcing 30-base-config.sh ...
---> 13:44:32     Processing basic MySQL configuration files ...
=> sourcing 60-replication-config.sh ...
=> sourcing 70-s2i-config.sh ...
---> 13:44:32     Processing additional arbitrary  MySQL configuration provided by s2i ...
=> sourcing 40-paas.cnf ...
=> sourcing 50-my-tuning.cnf ...
---> 13:44:32     Starting MySQL server with disabled networking ...
---> 13:44:32     Waiting for MySQL to start ...
/usr/libexec/mysqld: Deprecated program name. It will be removed in a future release, use '/usr/libexec/mariadbd' instead
2025-08-21 13:44:32 0 [Note] Starting MariaDB 11.8.3-MariaDB source revision b565b3e7e041c480fd8013e3beec6b99544d6bf8 server_uid PCi+RJyCwaDPfXL6I+4uNcJ2nvY= as process 20
2025-08-21 13:44:32 0 [ERROR] InnoDB: innodb_page_size=16384 requires innodb_buffer_pool_size >= 6 MiB current 5 MiB
2025-08-21 13:44:32 0 [ERROR] Plugin 'InnoDB' registration as a STORAGE ENGINE failed.
2025-08-21 13:44:32 0 [Note] Plugin 'FEEDBACK' is disabled.
2025-08-21 13:44:32 0 [Note] Plugin 'wsrep-provider' is disabled.
2025-08-21 13:44:32 0 [ERROR] Unknown/unsupported storage engine: InnoDB
2025-08-21 13:44:32 0 [ERROR] Aborting
Test for image 'quay.io/fedora/mariadb-118:11.8' FAILED (exit code: 1)
mysql: Deprecated program name. It will be removed in a future release, use '/usr/bin/mariadb' instead
ERROR 2002 (HY000): Can't connect to local server through socket '/var/lib/mysql/mysql.sock' (2)
-----------------------------------------------

@FaramosCZ PTAL and help @ndavidova

@ndavidova
Copy link
Contributor Author

[test]

1 similar comment
@ndavidova
Copy link
Contributor Author

[test]

Copy link
Member

@phracek phracek left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks for bringing new Fedora mariadb-container image to the Fedora land.
One note. As we do not build C9S, or C10S. Please remove it from build-and-push.yaml action.

I will look on INNODB_BUFFER_POOL_SIZE in detail.

Copy link
Member

@phracek phracek left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@ndavidova
Copy link
Contributor Author

[test]

@phracek
Copy link
Member

phracek commented Aug 28, 2025

Let's re-run test and innodb was increased.

[test]

@ndavidova
Copy link
Contributor Author

[test]

Copy link
Member

@FaramosCZ FaramosCZ left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

c8s, rhel8 and their related exclude files are neither expected to be copied over, nor edited or created for MariaDB 11.8.

@phracek
Copy link
Member

phracek commented Sep 1, 2025

[test]

@phracek
Copy link
Member

phracek commented Sep 1, 2025

It is a really BIG progress.
Only one error here:

-----------------------------------------------
Running test run_ssl_test (starting at 2025-09-01 14:24:12+00:00) ... 
-----------------------------------------------
  Testing ssl usage
.......+++++++++++++++++++++++++++++++++++++++*..........+..+.+.....+......+....+......+.....+.+..+++++++++++++++++++++++++++++++++++++++*...+....+...+.....+......+.+......+..+.......+.........+...........+...+.+..+....+...............+...+......+...........+....+......+...+....................+.......+......+..+...............+.+..+..........+.....+....+...........+..........+..+.+..+....+.........+.....+.++++++
..+.....+.+..+.......+...+..+............+...+..................+...+++++++++++++++++++++++++++++++++++++++*.....+...+...+..+.......+...+...+........+.+.....+.........+.......+..+...+...+...............+......+.+..+.+......+++++++++++++++++++++++++++++++++++++++*.+....+..+.........+....+..+.............+......+.........+..........................+.......+..+.+.........+...++++++
-----
    Created container 6208f6881101e1ad6ea0167c19cd431a2daf97504ffcc6fee75a197f8f10e9ca
  Testing MySQL connection to 10.88.0.144...
    Trying to connect....... OK
  Success!
mysql: Deprecated program name. It will be removed in a future release, use '/usr/bin/mariadb' instead
ERROR 2026 (HY000): TLS/SSL error: Hostname verification failed
  FAIL!
mysql: Deprecated program name. It will be removed in a future release, use '/usr/bin/mariadb' instead
ERROR 2026 (HY000): TLS/SSL error: Hostname verification failed
Test for image 'quay.io/fedora/mariadb-118:11.8' FAILED (exit code: 1)

@ndavidova
Copy link
Contributor Author

[test]

@FaramosCZ
Copy link
Member

I am now doing a review regarding the GIT quality:

1 to 4 fixed, thanks !

I'm still not satisfied with the justification of the Increase innodb buffer pool size commit.
I assume there is some equation how to compute the minimal amount required.
Please make the computation in the commit message so it is auditable, and also link to the upstream documentation where the computation is described.

New issue appeared - the two last commits are already present in the origin.
Bump common from 048d05c to 199f500 and Bump common from 199f500 to dc69f42 please rebase on top of them instead.

@phracek
Copy link
Member

phracek commented Sep 11, 2025

[test]

Copy link
Member

@phracek phracek left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM . Thanks for bringing this new container to our upstream repository. Great.

@phracek
Copy link
Member

phracek commented Sep 11, 2025

/auto-merge

Let's get merge it automatically.

ndavidova and others added 10 commits September 12, 2025 09:52
@phracek
Copy link
Member

phracek commented Sep 12, 2025

[test]

@phracek phracek merged commit 3e08819 into sclorg:master Sep 12, 2025
12 checks passed
@FaramosCZ
Copy link
Member

⚠️ @phracek you merged an unfinished PR into production.

The changes haven't passed my review.

Commits 6bc39db and f3fc8b1 should not be merged.

Neither they are complete nor auditable correct.

Can you please force remove them from production and make a standalone PR out of them, so we can resolve their remaining issues there ?

@phracek
Copy link
Member

phracek commented Sep 12, 2025

We can fix it by other PR.

@FaramosCZ
Copy link
Member

I strongly suggest we employ more careful approach @phracek

  1. There was no need to rush the PR
    We have neither external nor internal deadline that won't allow it t be worked for another week. Or a month.

  2. We now have unfinished commits in production, that I disputed not only for the formal deficiencies, but primarily because of the functionality and correctness of the code itself.

  3. You merged the code even though there was an actual open review requesting changes.
    Do you believe this is the best way we can cooperate on delivering high quality work?
    As the MariaDB / MySQL subject matter expert, I hoped to have some say in what I believe is an acceptable solution and what is not, regarding at least the DB behavior itself.

  4. Merging through unresolved reviews without a good justification send an uncomfortable message to the reviewers about how much you (don´t) value their effort.

  5. We are fortunate to have new contributors, like Nikola, around.
    It is especially important to care about teaching them correct habits, and make them familiar with doing the work the right way. Hasty Friday merges are quite the opposite of the values we wish to display, especially in PRs that are no way in a hurry.

I am happy to do hundred rounds of a review, if it gives value to the assignee.

  1. While it seems that we can fix the potential issues later, it's pretty hard to fix low quality commit messages. When someone comes tomorrow, asking why you selected that exact value for buffer pool size, or where the computation comes from what will you tell them?

The code one writes is often other people responsibility later.
That's why it's so important to learn to create high quality code and commit messages, and why reviews from different people brings value that single pair of eyes can possibly not.

--

Would you like to reconsider removing the unfinished commits, if we haven't yet have a production build from the content?

What is your opinion on implementing a guideline for this organization, to require all open questions and requests for changes to be at least answered with justification, before moving forward leaving them unresolved?

@phracek phracek mentioned this pull request Sep 17, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants