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

quincy: os/bluestore: use direct write in BlueStore::_write_bdev_label #48279

Merged
merged 1 commit into from Aug 11, 2023

Conversation

ifed01
Copy link
Contributor

@ifed01 ifed01 commented Sep 28, 2022

On AArch64 with kernel page size 64K, it occurs occasionally "OSD::init(): unable to read osd superblock" when deploying osd. As bluestore use direct write to write the superblock at 0x20001000 and BlueStore::_write_bdev_label use buffer write to write label at 0x01000, The OS flush the buffer write algined to page size, it will overwrite the superblock(0x2000~1000). Use driect write to avoid overwriting the superblock.

Fixes: https://tracker.ceph.com/issues/57688
Signed-off-by: luo rixin luorixin@huawei.com
(cherry picked from commit afdb5e4)
Signed-off-by: Igor Fedotov ifedotov@croit.io

Contribution Guidelines

Checklist

  • Tracker (select at least one)
    • References tracker ticket
    • Very recent bug; references commit where it was introduced
    • New feature (ticket optional)
    • Doc update (no ticket needed)
    • Code cleanup (no ticket needed)
  • Component impact
    • Affects Dashboard, opened tracker ticket
    • Affects Orchestrator, opened tracker ticket
    • No impact that needs to be tracked
  • Documentation (select at least one)
    • Updates relevant documentation
    • No doc update is appropriate
  • Tests (select at least one)
Show available Jenkins commands
  • jenkins retest this please
  • jenkins test classic perf
  • jenkins test crimson perf
  • jenkins test signed
  • jenkins test make check
  • jenkins test make check arm64
  • jenkins test submodules
  • jenkins test dashboard
  • jenkins test dashboard cephadm
  • jenkins test api
  • jenkins test docs
  • jenkins render docs
  • jenkins test ceph-volume all
  • jenkins test ceph-volume tox
  • jenkins test windows

On AArch64 with kernel page size 64K, it occurs occasionally
"OSD::init(): unable to read osd superblock" when deploying osd.
As bluestore use direct write to write the superblock at 0x2000~1000
and BlueStore::_write_bdev_label use buffer write to write label at
0x0~1000, The OS flush the buffer write algined to page size, it will
overwrite the superblock(0x2000~1000). Use driect write to avoid
overwriting the superblock.

Fixes: https://tracker.ceph.com/issues/57537
Signed-off-by: luo rixin <luorixin@huawei.com>
(cherry picked from commit afdb5e4)
@ifed01 ifed01 added this to the quincy milestone Sep 28, 2022
@ifed01 ifed01 requested a review from aclamk September 28, 2022 09:20
@ifed01 ifed01 requested a review from a team as a code owner September 28, 2022 09:20
@github-actions github-actions bot added the core label Sep 28, 2022
@rosinL
Copy link
Member

rosinL commented Jun 30, 2023

jenkins retest this please

@rosinL
Copy link
Member

rosinL commented Jun 30, 2023

@ifed01 @aclamk , It is suspended for a long time, can we merge this?

@ifed01
Copy link
Contributor Author

ifed01 commented Jun 30, 2023

jenkins test make check

@rosinL
Copy link
Member

rosinL commented Jul 4, 2023

jenkins retest this please

1 similar comment
@rosinL
Copy link
Member

rosinL commented Jul 4, 2023

jenkins retest this please

@ifed01 ifed01 added the needs-qa label Jul 4, 2023
@ifed01
Copy link
Contributor Author

ifed01 commented Jul 4, 2023

jenkins test docs

@ifed01
Copy link
Contributor Author

ifed01 commented Aug 7, 2023

jenkins test docs

@ljflores
Copy link
Contributor

ljflores commented Aug 10, 2023

Hey @ifed01, a new failure reproduced twice in this batch: https://tracker.ceph.com/issues/62401

There were 3 bluestore PRs in the batch, so I will leave a similar message on all of them, as I'm not sure which one could be responsible. Can you take a look and see if it's related? If not, the test results were otherwise clean.

Rados suite review here: https://tracker.ceph.com/projects/rados/wiki/QUINCY#httpstrellocomcw1wxAcJO1814-wip-yuri8-testing-2023-07-24-0819-quincy

The other two PRs were:
#51774
#48171

@ljflores
Copy link
Contributor

Feel free to merge if the above failure is unrelated.

@ifed01 ifed01 merged commit 4345431 into ceph:quincy Aug 11, 2023
6 of 7 checks passed
@ifed01 ifed01 deleted the wip-ifed-fix-57688 branch August 11, 2023 12:58
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
5 participants