v8: fix deadlock when releasing caps after a cancelled operation
set epoch_barrier before cancelling operations on full conditions
v7: don't request a continuous sub when the epoch is not yet at the
barrier. Don't set epoch barrier unless we aborted a call that
may have already been sent. Use READ_ONCE in lockless ERROR_WRITE
flag handling.
v6: reset barrier to current epoch when receiving map with full pool or
at quota condition. Show epoch barrier in debugfs. Don't take osd->lock
unnecessarily. Remove req->r_replay_version. Other cleanups and
fixes suggested by Ilya.
v5: rebase onto ACK vs. commit changes
v4: eliminate map_cb and just call ceph_osdc_abort_on_full directly
Revert earlier patch flagging individual pages with error on writeback
failure. Add mechanism to force synchronous writes when writes start
failing, and reallowing async writes when they succeed.
v3: track "abort_on_full" behavior with a new bool in osd request
instead of a protocol flag. Remove some extraneous arguments from
various functions. Don't export have_pool_full, call it from the
abort_on_full callback instead.
v2: teach libcephfs how to hold on to requests until the right map
epoch appears, instead of delaying cap handling in the cephfs layer.
This patchset is an updated version of the patch series originally
done by John Spray and posted here:
http://www.spinics.net/lists/ceph-devel/msg21257.html
This is ai small update to the set I posted a couple of weeks ago.
We hit a deadlock in testing that was due to taking the osdc->lock
recursively. While diagnosing that, I realized that we can end up
releasing caps as a result of a cancelled operation.
In that case, we probably want to ensure that the cap release message
has an updated epoch barrier in it, so there's also a change to
ceph_osdc_abort_on_full to set the epoch barrier prior to cancelling the
operations.
So far this seems to do well with the re-enabled ENOSPC testing in
teuthology, though we do have to make a small change to the testsuite to
compensate for a difference between kcephfs and ceph-fuse. ceph-fuse
returns errors on a close if there were writeback issues prior to
closing. The kernel does not do that currently.
For now, we're leaving the kcephfs behavior as-is. I think this is a
place where we need to have consistent behavior across Linux
filesystems, and that means a larger discussion about what userspace
consumers really want here.
Jeff Layton (7):
libceph: remove req->r_replay_version
libceph: allow requests to return immediately on full conditions if
caller wishes
libceph: abort already submitted but abortable requests when map or
pool goes full
libceph: add an epoch_barrier field to struct ceph_osd_client
ceph: handle epoch barriers in cap messages
Revert "ceph: SetPageError() for writeback pages if writepages fails"
ceph: when seeing write errors on an inode, switch to sync writes
fs/ceph/addr.c | 10 ++--
fs/ceph/caps.c | 21 +++++--
fs/ceph/file.c | 32 ++++++-----
fs/ceph/mds_client.c | 20 +++++++
fs/ceph/mds_client.h | 7 ++-
fs/ceph/super.h | 26 +++++++++
include/linux/ceph/osd_client.h | 4 +-
net/ceph/debugfs.c | 7 +--
net/ceph/osd_client.c | 118 +++++++++++++++++++++++++++++++++++++---
9 files changed, 208 insertions(+), 37 deletions(-)
--
2.9.3
--
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
