Actually, scratch that. I went ahead and opened this:

https://tracker.ceph.com/issues/43649

Feel free to watch that one for updates.

On Fri, 2020-01-17 at 07:43 -0500, Jeff Layton wrote:
> No problem. Can you let me know the tracker bug number once you've
> opened it?
> 
> Thanks,
> Jeff
> 
> On Thu, 2020-01-16 at 20:24 -0500, Aaron wrote:
> > This debugging started because the ceph-provisioner from k8s was making 
> > those users...but what we found was doing something similar by hand caused 
> > the same issue. Just surprised no one else using k8s and ceph backed 
> > PVC/PVs  ran into this issue. 
> > 
> > Thanks again for all your help!
> > 
> > Cheers
> > Aaron
> > 
> > On Thu, Jan 16, 2020 at 8:21 PM Aaron <aarongm...@gmail.com> wrote:
> > > No worries, can definitely do that. 
> > > 
> > > Cheers
> > > Aaron
> > > 
> > > On Thu, Jan 16, 2020 at 8:08 PM Jeff Layton <jlay...@redhat.com> wrote:
> > > > On Thu, 2020-01-16 at 18:42 -0500, Jeff Layton wrote:
> > > > > On Wed, 2020-01-15 at 08:05 -0500, Aaron wrote:
> > > > > > Seeing a weird mount issue.  Some info:
> > > > > > 
> > > > > > No LSB modules are available.
> > > > > > Distributor ID: Ubuntu
> > > > > > Description: Ubuntu 18.04.3 LTS
> > > > > > Release: 18.04
> > > > > > Codename: bionic
> > > > > > 
> > > > > > Ubuntu 18.04.3 with kerne 4.15.0-74-generic
> > > > > > Ceph 14.2.5 & 14.2.6
> > > > > > 
> > > > > > With ceph-common, ceph-base, etc installed:
> > > > > > 
> > > > > > ceph/stable,now 14.2.6-1bionic amd64 [installed]
> > > > > > ceph-base/stable,now 14.2.6-1bionic amd64 [installed]
> > > > > > ceph-common/stable,now 14.2.6-1bionic amd64 [installed,automatic]
> > > > > > ceph-mds/stable,now 14.2.6-1bionic amd64 [installed]
> > > > > > ceph-mgr/stable,now 14.2.6-1bionic amd64 [installed,automatic]
> > > > > > ceph-mgr-dashboard/stable,stable,now 14.2.6-1bionic all [installed]
> > > > > > ceph-mon/stable,now 14.2.6-1bionic amd64 [installed]
> > > > > > ceph-osd/stable,now 14.2.6-1bionic amd64 [installed]
> > > > > > libcephfs2/stable,now 14.2.6-1bionic amd64 [installed,automatic]
> > > > > > python-ceph-argparse/stable,stable,now 14.2.6-1bionic all 
> > > > > > [installed,automatic]
> > > > > > python-cephfs/stable,now 14.2.6-1bionic amd64 [installed,automatic]
> > > > > > 
> > > > > > I create a user via get-or-create cmd, and I have a users/secret 
> > > > > > now.
> > > > > > When I try to mount on these Ubuntu nodes,
> > > > > > 
> > > > > > The mount cmd I run for testing is:
> > > > > > sudo mount -t ceph -o
> > > > > > name=user-20c5338c-34db-11ea-b27a-de7033e905f6,secret=AQC6dhpeyczkDxAAhRcr7oERUY4BcD2NCUkuNg==
> > > > > > 10.10.10.10:6789:/work/20c5332d-34db-11ea-b27a-de7033e905f6 
> > > > > > /tmp/test
> > > > > > 
> > > > > > I get the error:
> > > > > > couldn't finalize options: -34
> > > > > > 
> > > > > > From some tracking down, it's part of the get_secret_option() in
> > > > > > common/secrets.c and the Linux System Error:
> > > > > > 
> > > > > > #define ERANGE          34      /* Math result not representable */
> > > > > > 
> > > > > > Now the weird part...when I remove all the above libs above, the 
> > > > > > mount
> > > > > > command works. I know that there are ceph.ko modules in the Ubuntu
> > > > > > filesystems DIR, and that Ubuntu comes with some understanding of 
> > > > > > how
> > > > > > to mount a cephfs system.  So, that explains how it can mount
> > > > > > cephfs...but, what I don't understand is why I'm getting that -34
> > > > > > error with the 14.2.5 and 14.2.6 libs installed. I didn't have this
> > > > > > issue with 14.2.3 or 14.2.4.
> > > > > 
> > > > > This sounds like a regression in mount.ceph, probably due to something
> > > > > that went in for v14.2.5. I can reproduce the problem on Fedora, and I
> > > > > think it has something to do with the very long username you're using.
> > > > > 
> > > > > I'll take a closer look and let you know. Stay tuned.
> > > > > 
> > > > 
> > > > I think I see the issue. The SECRET_OPTION_BUFSIZE is just too small for
> > > > your use case. We need to make that a little larger than the largest
> > > > name= parameter can be. Prior to v14.2.5, it was ~1000 bytes, but I made
> > > > it smaller in that set thinking that was too large. Mea culpa.
> > > > 
> > > > The problem is determining how big that size can be. AFAICT EntityName
> > > > is basically a std::string in the ceph code, which can be an arbitrary
> > > > size (up to 4g or so).
> > > > 
> > > > Aaron, would you mind opening a bug for this at tracker.ceph.com? We
> > > > should be able to get it fixed up, once I do a bit more research to
> > > > figure out how big to make this buffer.

-- 
Jeff Layton <jlay...@redhat.com>

_______________________________________________
ceph-users mailing list
ceph-users@lists.ceph.com
http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com

Reply via email to