Re: CEPH_RBD_API: options on image create

> > I am concerned about passing a void* + length to specify the option > > value since you really can't protect against the user providing data > > in the incorrect format. For example, if the backend treated > > RBD_OPTION_STRIPE_UNIT as a 4byte int, what happens if someone > > passes a 2- or

Re: CEPH_RBD_API: options on image create

> > But we don't need them to match between different platforms, no? Is > linking 64bit code with 32bit possible (supported)? > > Also, for this particular (char*) case, length would actually be the > length of the string, not the pointer length. From my example: > > const char*

[PATCH v2] libceph: add support for CEPH_OSD_OP_GETXATTR

Allows for xattr retrieval. Response data buffer allocation is the responsibility of the osd_req_op_xattr_init() caller. Signed-off-by: David Disseldorp --- include/linux/ceph/osd_client.h | 8 ++- net/ceph/osd_client.c | 47

Re: CEPH_RBD_API: options on image create

On Thu, Oct 15, 2015 at 08:05:07AM -0400, Jason Dillaman wrote: > > > I am concerned about passing a void* + length to specify the option > > > value since you really can't protect against the user providing data > > > in the incorrect format. For example, if the backend treated > > >

Re: does newstore skip some data ??

On Thu, 15 Oct 2015, Tomy Cheru wrote: > Tried to write 160k objects of 64k size, totaling 10G to newstore via rgw > > PUTS were successful > > # ./rados df > .log 0 1270 254 > 0 762 635 5080 >

Re: CEPH_RBD_API: options on image create

On Wed, Oct 14, 2015 at 03:34:52PM -0400, Jason Dillaman wrote: > In general, I like the approach. > > I am concerned about passing a void* + length to specify the option > value since you really can't protect against the user providing data > in the incorrect format. For example, if the

Re: make check bot delays

On 15/10/2015 11:45, Dałek, Piotr wrote: >> -Original Message- >> From: ceph-devel-ow...@vger.kernel.org [mailto:ceph-devel- >> ow...@vger.kernel.org] On Behalf Of Loic Dachary >> Sent: Thursday, October 15, 2015 11:09 AM >> >> Hi, >> >> TL;DR: the make check bot is fixed (no more

Re: make check bot delays

Hi, TL;DR: the make check bot is fixed (no more delays) but only keeps the last 30 runs It looks like the cause of the jenkins delays was the number of run kept in the archives (set to 300). Lowering it down to 30 have jenkins back to running with no delay between jobs (as opposed to using

does newstore skip some data ??

Tried to write 160k objects of 64k size, totaling 10G to newstore via rgw PUTS were successful # ./rados df .log 0 1270 254 0 762 635 5080 .rgw 8 400

Re: [PATCH] ceph/osd_client: add support for CEPH_OSD_OP_GETXATTR

On Thu, Oct 15, 2015 at 12:51 AM, David Disseldorp wrote: > On Wed, 14 Oct 2015 19:57:46 +0200, Ilya Dryomov wrote: > >> On Wed, Oct 14, 2015 at 7:37 PM, David Disseldorp wrote: > ... >> > Ping, any feedback on the patch? >> >> The patch itself looks OK, except for

Re: CEPH_RBD_API: options on image create

On Wed, Oct 14, 2015 at 08:12:37PM -0700, Josh Durgin wrote: > On 10/14/2015 12:34 PM, Jason Dillaman wrote: > >In general, I like the approach. > > > >I am concerned about passing a void* + length to specify the option value > >since you really can't protect against the user providing data in

RE: make check bot delays

> -Original Message- > From: ceph-devel-ow...@vger.kernel.org [mailto:ceph-devel- > ow...@vger.kernel.org] On Behalf Of Loic Dachary > Sent: Thursday, October 15, 2015 11:09 AM > > Hi, > > TL;DR: the make check bot is fixed (no more delays) but only keeps the last > 30 runs >[..] > What

Re: XFS xattr limit and Ceph

xfs has three store types for xattrs like inline, btree and extents. We only want to let xattr stored inline, so it won't need to hit disk. So we need to limit the number of xattrs On Thu, Oct 15, 2015 at 10:54 PM, Somnath Roy wrote: > Sage, > Why we are using XFS max

RE: does newstore skip some data ??

Thanks Sage, make sense. -Original Message- From: Sage Weil [mailto:s...@newdream.net] Sent: Thursday, October 15, 2015 5:56 PM To: Tomy Cheru Cc: ceph-devel@vger.kernel.org Subject: Re: does newstore skip some data ?? On Thu, 15 Oct 2015, Tomy Cheru wrote: > Tried to write 160k objects

Re: XFS xattr limit and Ceph

Got it, thanks Haomai Sent from my iPhone > On Oct 15, 2015, at 8:08 AM, Haomai Wang wrote: > > xfs has three store types for xattrs like inline, btree and extents. > We only want to let xattr stored inline, so it won't need to hit disk. > So we need to limit the number

Re: CEPH_RBD_API: options on image create

On Thu, Oct 15, 2015 at 08:47:58AM -0400, Jason Dillaman wrote: > > > > But we don't need them to match between different platforms, no? Is > > linking 64bit code with 32bit possible (supported)? > > > > Also, for this particular (char*) case, length would actually be the > > length of the

Re: CEPH_RBD_API: options on image create

On Thu, 15 Oct 2015, Mykola Golub wrote: > On Thu, Oct 15, 2015 at 08:47:58AM -0400, Jason Dillaman wrote: > > > > > > But we don't need them to match between different platforms, no? Is > > > linking 64bit code with 32bit possible (supported)? > > > > > > Also, for this particular (char*) case,

XFS xattr limit and Ceph

Sage, Why we are using XFS max inline xattr value as 10 only ? OPTION(filestore_max_inline_xattrs_xfs, OPT_U32, 10) XFS is supporting 1k limit I guess. Is there any performance reason behind that ? Thanks & Regards Somnath PLEASE NOTE: The information

Re: [PATCH] rbd: don't leak parent_spec in rbd_dev_probe_parent()

On 10/15/2015 01:31 PM, Ilya Dryomov wrote: > On Thu, Oct 15, 2015 at 7:10 PM, Alex Elder wrote: >> On 10/11/2015 01:03 PM, Ilya Dryomov wrote: >>> Currently we leak parent_spec and trigger a "parent reference >>> underflow" warning if rbd_dev_create() in rbd_dev_probe_parent()

Unable to mount CephFS

Dear peoples in mailing-list, I've got a problem which is cannot mount CephFS. This is my ceph status form 'ceph -s' cluster 7e549325-246f-4265-8715-11591a26df79 health HEALTH_OK monmap e3: 3 mons at {rc010=192.168.0.10:6789/0,rc012=192.168.0.12:6789/0,rc013=192.168.0.13:6789/0}

Re: Unable to mount CephFS

Could you explain more about it ? What is it supposed to be ? This is my output of 'ceph auth list' mds.rc010 key: AQCrcxZWPVj9GBAANaJskP61dEYHFwDTCojdDg== caps: [mds] allow caps: [mon] allow profile mds caps: [osd] allow rwx mds.rc012 key:

Re: Unable to mount CephFS

you mount with name=user, but there is no "client.user" in auth list, what's about mount with "name=admin"? 2015-10-16 10:56 GMT+08:00 boukom.p.aa : > Could you explain more about it ? What is it supposed to be ? > > This is my output of 'ceph auth list' > > mds.rc010

Re: Unable to mount CephFS

do you verified that 'user' have sufficient permission? use 'ceph auth list ' to check it. 2015-10-16 9:29 GMT+08:00 boukom.p.aa : > Dear peoples in mailing-list, > > I've got a problem which is cannot mount CephFS. > This is my ceph status form 'ceph -s' > >

Re: [PATCH] rbd: don't leak parent_spec in rbd_dev_probe_parent()

On 10/11/2015 01:03 PM, Ilya Dryomov wrote: > Currently we leak parent_spec and trigger a "parent reference > underflow" warning if rbd_dev_create() in rbd_dev_probe_parent() fails. > The problem is we take the !parent out_err branch and that only drops > refcounts; parent_spec that would've been

RE: chooseleaf may cause some unnecessary pg migrations

I did some test by using crushtool --test (together with this PR and #PR 6004) It doesn't help in quantities way. In a 40 OSDs, 10 OSDs per Nodes demo crush map, test with rep = 2. 4096 PGs, in each run I will randomly kick off an OSD (reweight to 0) and compared the PG mapping. If any OSD in

RE: does newstore skip some data ??

How many osds do you have? I wonder if the overlay layer is that large to keep 160K obj(which is 64K * 32 on default, per OSD)? > -Original Message- > From: ceph-devel-ow...@vger.kernel.org [mailto:ceph-devel- > ow...@vger.kernel.org] On Behalf Of Tomy Cheru > Sent: Thursday, October

Re: CEPH_RBD_API: options on image create

Thank you all for your comments! I will come back with PR and pull request. -- Mykola Golub On Thu, Oct 15, 2015 at 11:29:56AM -0700, Josh Durgin wrote: > On 10/15/2015 06:45 AM, Sage Weil wrote: > >On Thu, 15 Oct 2015, Mykola Golub wrote: > >>On Thu, Oct 15, 2015 at 08:47:58AM -0400, Jason

Re: [PATCH] rbd: don't leak parent_spec in rbd_dev_probe_parent()

On 10/15/2015 01:31 PM, Ilya Dryomov wrote: > On Thu, Oct 15, 2015 at 7:10 PM, Alex Elder wrote: >> On 10/11/2015 01:03 PM, Ilya Dryomov wrote: >>> Currently we leak parent_spec and trigger a "parent reference >>> underflow" warning if rbd_dev_create() in rbd_dev_probe_parent()

Re: [PATCH] rbd: don't leak parent_spec in rbd_dev_probe_parent()

On Thu, Oct 15, 2015 at 7:10 PM, Alex Elder wrote: > On 10/11/2015 01:03 PM, Ilya Dryomov wrote: >> Currently we leak parent_spec and trigger a "parent reference >> underflow" warning if rbd_dev_create() in rbd_dev_probe_parent() fails. >> The problem is we take the !parent

RE: chooseleaf may cause some unnecessary pg migrations

On Thu, 15 Oct 2015, Xusangdi wrote: > > -Original Message- > > From: Sage Weil [mailto:sw...@redhat.com] > > Sent: Wednesday, October 14, 2015 10:18 PM > > To: xusangdi 11976 (RD) > > Cc: ceph-devel@vger.kernel.org > > Subject: RE: chooseleaf may cause some unnecessary pg migrations > > >

[PATCH] mark rbd requiring stable pages

rbd requires stable pages, as it performs a crc of the page data before they are send to the OSDs. But since kernel 3.9 (patch 1d1d1a767206fbe5d4c69493b7e6d2a8d08cc0a0 "mm: only enforce stable page writes if the backing device requires it") it is not assumed anymore that block devices require

Keystone auth v3 changes

The keystone v2 api is slated for removal sometime in the near future making radosgw unusable with new openstack deployments. Hoping to jumpstart the process of adding v3 support I propose the following changes as a minimum to make radosgw compatible with the v3 auth api. There is an open

Re: CEPH_RBD_API: options on image create

On 10/15/2015 06:45 AM, Sage Weil wrote: On Thu, 15 Oct 2015, Mykola Golub wrote: On Thu, Oct 15, 2015 at 08:47:58AM -0400, Jason Dillaman wrote: But we don't need them to match between different platforms, no? Is linking 64bit code with 32bit possible (supported)? Also, for this particular