check return value of mkdir
Signed-off-by: Xing Lin <xinglin@cs.utah.edu>
test/filestore/workloadgen: fix return value check for mkdir
allow EEXIST from mkdir()
Signed-off-by: Xing Lin <xinglin@cs.utah.edu>
blkid -s TYPE /dev/fd0 has been verified to hang forever on a
H8DMR-82 supermicro motherboard running
3.8.0-33-generic #48~precise1-Ubuntu SMP Thu Oct 24 16:28:06 UTC 2013
x86_64
It is unlikely that ceph will ever be used on floppy disks, they
can be blacklisted.
http://tracker.ceph.com/issues/6827fixes: #6827
Signed-off-by: Loic Dachary <loic@dachary.org>
These are supported as of kernel version v3.9:
- chooseleaf_descend_once was added in 1604f48
- others were added in 546f04e
And of course these are supported by bobtail and later.
Signed-off-by: Sage Weil <sage@inktank.com>
Set the tunables to the 'default' values on create.
Default to the legacy values when we decode, so that an old map without
values defined gets legacy values.
Signed-off-by: Sage Weil <sage@inktank.com>
4d140a71a1 refactored this and
introduced a bug. peek_meta() was accidentally reading into magic,
then replacing magic with val, which was always the empty string,
resulting in the osd always failing to start due to 'mismatched'
magic values.
Signed-off-by: Josh Durgin <josh.durgin@inktank.com>
One use noted that they have a 'pdu' type in their type hierarchy that
typically spans multiple racks. Others are known to use the 'pod'
terminology; add that to. And I can imagine 'region' above datacenter.
Factor this into a helper to make things a bit less fragile.
Signed-off-by: Sage Weil <sage@inktank.com>
A chassis is usually bigger than a host but smaller than a rack. This will
be useful for a broad class of modern hardware that sticks multiple hosts
in the same chassis (in sleds, or on cards, or blades, or whatever).
Signed-off-by: Sage Weil <sage@inktank.com>
This reverts commit 08327fed82, except
for the hashpspool bit. We switched back to an integer argument in
commit 337195f046.
Signed-off-by: Sage Weil <sage@inktank.com>
This avoids parsing out the wrong value when a long device name makes
df wrap over two lines.
Fixes: #6699
Reported-by: Jan Harkes <jaharkes@cs.cmu.edu>
Reviewed-by: Noah Watkins <noah.watkins@inktank.com>
Signed-off-by: Sage Weil <sage@inktank.com>
Move these from the OSD. Use a generic implementation in ObjectStore that
hopefully all backends can share (so that it can remain in sync with the
start/stop scripts, ceph-disk, and other orchestration machinery).
Signed-off-by: Sage Weil <sage@inktank.com>
This lets ceph_osd.cc handle the config details and use it directly for
all of the random command-line stuff, eliminating a bunch of mostly-
useless static wrappers in OSD.
Signed-off-by: Sage Weil <sage@inktank.com>
Generic way to create an ObjectStore implementation of the required type,
so that users don't need to know anything about it.
Signed-off-by: Sage Weil <sage@inktank.com>
When the replica responds, it responds with a notify
rather than a log, which the primary then ignores since
it is already in the peer_info map. Rather than fix that
we'll simply not send queries to peers we already know to
have no unfound objects.
Fixes: #6910
Signed-off-by: Samuel Just <sam.just@inktank.com>
Reviewed-by: Sage Weil <sage@inktank.com>
Reviewed-by: David Zafman <david.zafman@inktank.com>
If for some reason there are no up OSDs in the history which
happen to have usable copies of the pg, it's possible that
there is a usable copy elsewhere on the cluster which will
become known to the primary if it waits.
Fixes: #6909
Signed-off-by: Samuel Just <sam.just@inktank.com>
Reviewed-by: Greg Farnum <greg@inktank.com>
Reviewed-by: Sage Weil <sage@inktank.com>
Fixes: #6892
Backport: dumpling, emperor
s3cmd specifies empty owner field when trying to set acls on object
/ bucket. We errored out as it didn't match the current owner name, but
with this change we ignore it.
Signed-off-by: Yehuda Sadeh <yehuda@inktank.com>
