Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add arm to docker #1

Open
wants to merge 149 commits into
base: master
Choose a base branch
from
Open

Add arm to docker #1

wants to merge 149 commits into from

Conversation

yaronkaikov
Copy link
Owner

No description provided.

@yaronkaikov yaronkaikov force-pushed the add-arm-to-docker branch 3 times, most recently from 38f477e to 7d544f4 Compare May 19, 2023 19:55
dgarcia360 and others added 27 commits June 5, 2023 16:33
Fix typo - later -> latter
Changing default system_auth replication strategy is strongly recommended in multi-dc clusters. Testing environment should follow this rule.
The minimal requirements for cql user used for restore tables are that it should be able to ALTER restored tables. Restore tests should be executed on users with only necessary permissions.
* testing: add ipv6 test environment

* testutils: enable ipv6 environment

* backup-tests: enable ipv6 environment

* cluster-tests: enable ipv6 environment

* repair-tests: enable ipv6 environment

* healthcheck-tests: enable ipv6 environment

* scylla-client-tests: enable ipv6 environment

* ping-tests: enable ipv6 environment

* util-tests: enable ipv6 environment

* cicd: populate ipv6 tests

---------

Co-authored-by: Dmitry Kropachev <[email protected]>
It's required for performing automatic repair after restoring tables.
Storing repair task ID is necessary for pausing and resuming automated post-restore repair task.
As repair is now part of the restore procedure, it's progress should also be included.
karol-kokoszka and others added 27 commits September 12, 2023 20:47
Time to time it can happen that agent and server are failing on the start
To exclude all temporary cases `RestartPreventExitStatus` is removed to make systemd restart them automatically
And `StartLimitBurst` is set to make systemd not to start them if problem is not going away
…b#3555)

Up to this point there were two possible values of ping status metric: 1 for success and -1 for failure. The problem was that -1 could mean that either Scylla or Agent are unavailable. To make it more descriptive, now new value -2 describes unavailable Agent and -1 describes unavailable Scylla.
Previously total progress aggregated on monitoring side looked like:
(token_ranges_success + token_ranges_error) / token_ranges_total.
This approach didn't take table size into consideration, so it often lead to misleading progress display (especially since SM 3.2 repairs small tables first).
Now total progress is an additional metric exposed by SM that weights done token ranges with table size.

Fixes scylladb#3547
This approach is safer than printing 0% error when only a fraction of a task has failed.

Fixes scylladb#3581
In general, it shouldn't be important, but this allows for explicitly disabling small table optimization even when table size equals 0. Issue scylladb#3584 has shown that table size report might be inaccurate and that it would be useful to have this behavior.
Adding arm support to our Scylla-manager docker image

This will only build addtional docker image, the publish part will be
done as part of scylladb/scylla-pkg#3239

Closes: scylladb#3278
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

10 participants