Tasks #8535
audit the journaler send-to-OSD ordering
% Done:
0%
Tags:
Reviewed:
Affected Versions:
Component(FS):
MDS
Labels (FS):
Pull request ID:
Description
I've seen several cases recently where users had a journal header indicating that the journal ended one object later than actually existed in RADOS. I suspect we may be updating the header before we've actually received the object commits from the OSDs, which is bad. Check that; fix if so.
History
#1 Updated by Greg Farnum over 9 years ago
This looks fine — we send out the journal header at the same time as the blocks to write, but we update the write_pos based on safe_pos (which is only incremented by safe acks).
#2 Updated by Greg Farnum over 9 years ago
- Status changed from New to Closed
#3 Updated by Greg Farnum over 7 years ago
- Component(FS) MDS added