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

osd: Add config option osd_read_ec_check_for_errors for testing #5865

Merged
merged 1 commit into from Oct 30, 2015

Conversation

Projects
None yet
4 participants
@dzafman
Copy link
Member

dzafman commented Sep 10, 2015

All rados suites involving erasure codes should set this option

Signed-off-by: David Zafman dzafman@redhat.com

@liewegas

This comment has been minimized.

Copy link
Member

liewegas commented Oct 1, 2015

@athanatos @dzafman what's the status of this pr?

@dzafman

This comment has been minimized.

Copy link
Member Author

dzafman commented Oct 1, 2015

@liewegas @athanatos I didn't want a "Bad hash" to go undetected because we don't return errors if enough EC shards are readable. Now that I think about it assuming that it would be caught by the log scan after a run, we don't need this change.

@dzafman

This comment has been minimized.

Copy link
Member Author

dzafman commented Oct 1, 2015

@liewegas I think we should merge this post-Infernalis even if we don't end up using the option as long as it doesn't break things when turned off (the default).

err = rop.complete[iter->first].errors.begin()->second;
rop.complete[iter->first].r = err;
} else {
dout(10) << __func__ << " Enough copies for " << iter->first << " (ignore errors)" << dendl;

This comment has been minimized.

@liewegas

liewegas Oct 2, 2015

Member

should this log something to the central cluster log?

This comment has been minimized.

@dzafman

dzafman Oct 2, 2015

Author Member

Good idea!

@liewegas

This comment has been minimized.

Copy link
Member

liewegas commented Oct 2, 2015

passed wip-sage-testing. seems like we should also log to the cluster log if we see a disk error (even though we compensate)?

@liewegas liewegas removed this from the infernalis milestone Oct 7, 2015

osd: Add config option osd_read_ec_check_for_errors for testing
All rados suites involving erasure codes should set this option
Fix message when ignoring errors and log to cluster log

Signed-off-by: David Zafman <dzafman@redhat.com>
@dzafman

This comment has been minimized.

Copy link
Member Author

dzafman commented Oct 15, 2015

@liewegas Added cluster logging and improved the message. Will retest with my other pull request.

@dzafman dzafman assigned liewegas and unassigned dzafman Oct 15, 2015

@dzafman dzafman force-pushed the dzafman:wip-13016 branch from c7b5c28 to 07e7496 Oct 20, 2015

@dzafman

This comment has been minimized.

Copy link
Member Author

dzafman commented Oct 20, 2015

@liewegas Passed a subset of rados suite (cluster logging added previously)

@dzafman

This comment has been minimized.

Copy link
Member Author

dzafman commented Oct 27, 2015

@liewegas ping...ready to merge

@dzafman dzafman assigned athanatos and unassigned liewegas Oct 27, 2015

@dzafman

This comment has been minimized.

Copy link
Member Author

dzafman commented Oct 27, 2015

@athanatos Ping, ready to merge after review.

athanatos pushed a commit that referenced this pull request Oct 30, 2015

Samuel Just
Merge pull request #5865 from dzafman/wip-13016
osd: Add config option osd_read_ec_check_for_errors for testing

Reviewed-by: Samuel Just <sjust@redhat.com>

@athanatos athanatos merged commit e74183e into ceph:master Oct 30, 2015

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
You can’t perform that action at this time.