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
doc/cephfs/posix: update #6922
doc/cephfs/posix: update #6922
Conversation
However, there are a few places where CephFS diverges from strict | ||
POSIX semantics for various reasons: | ||
|
||
- In shared simultaneous writer situations, a write that crosses |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Do we actually enable this case these days? I thought that we only had two clients writing simultaneously with the old O_LAZY flag, otherwise we can only have one writing at a time.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
If you have read+write or write+write we flip to LOCK_MIX mode, which gives out RD and WR caps but makes all IO synchronous (on RDCACHE or WRBUFFER). All reads and writes are effectively serialized at the OSD.
CEPH_IOC_LAZY asks the mds to give the client RDCACHE and WRBUFFER despite the sharing, which means the app is responsible for invalidating its cache or flushing changes at the appropriate points...
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
OK, got it.
seems good |
4bc666e
to
d10f61a
Compare
Signed-off-by: Sage Weil <sage@redhat.com>
d10f61a
to
d5cee59
Compare
Signed-off-by: Sage Weil sage@redhat.com