FYI.
---------- Forwarded message ----------
From: David Casier <david.casier@xxxxxxxx>
Date: 2015-12-21 23:19 GMT+01:00
Subject: FileStore : no wait thread queue_sync
To: Ceph Development <ceph-devel@xxxxxxxxxxxxxxx>, Sage Weil <sage@xxxxxxxxxxxx>
Cc: Benoît LORIOT <benoit.loriot@xxxxxxxx>, Sébastien VALSEMEY
<sebastien.valsemey@xxxxxxxx>
Hi,
What do you think about :
if (!journal && m_filestore_direct) {
apply_manager.commit_finish
}
in FileStore::queue_transactions ?
For direct and no waiting (sync_entry thread) ?
I would also propose putting a parameter "m_omap_is_safe" for bypass
XATTR_SPILL_OUT_NAME and reduce IOPS in hard_drive
if ( !m_omap_is_safe) {
r = chain_fgetxattr(**o, XATTR_SPILL_OUT_NAME, buf, sizeof(buf));
if (r >= 0 && !strncmp(buf, XATTR_NO_SPILL_OUT,
sizeof(XATTR_NO_SPILL_OUT))) {
r = chain_fsetxattr(**n, XATTR_SPILL_OUT_NAME, XATTR_NO_SPILL_OUT,
sizeof(XATTR_NO_SPILL_OUT));
} else {
r = chain_fsetxattr(**n, XATTR_SPILL_OUT_NAME, XATTR_SPILL_OUT,
sizeof(XATTR_SPILL_OUT));
}
}
--
________________________________________________________
Cordialement,
David CASIER
________________________________________________________
--
________________________________________________________
Cordialement,
David CASIER
________________________________________________________
--
To unsubscribe from this list: send the line "unsubscribe ceph-devel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
