Unexpected SU+J inconsistency AGAIN -- please, don't shift topic to ZFS!
lev at FreeBSD.org
Thu Feb 28 15:00:57 UTC 2013
You wrote 28 февраля 2013 г., 18:56:47:
LS> There was commit r246876, but I don't understand exactly what it
LS> means, as no real FS or driver's code was touched.
And, yes, barriers are much stronger than "sync writes", as they
should flush all previous writes, even that is not related to journal
or metadata and could wait more (simple file data could be fixed on
plates out of order without destroying filesystem structure).
// Black Lion AKA Lev Serebryakov <lev at FreeBSD.org>
More information about the freebsd-current