MDS pinning: ceph.dir.pin: No such attribute
Hi, I seem to have a problem with the extended attributes for MDS pinning. Ceph version is mimic-13.2.10 and the documentation (https://web.archive.org/web/20190716110503/http://docs.ceph.com/docs/mimic/c...) says I should be able to setfattr -n ceph.dir.pin -v 2 path/to/dir When I try this, I get the following: # setfattr -n ceph.dir.pin -v 1 . # getfattr -n ceph.dir.pin . .: ceph.dir.pin: No such attribute Is this expected? How can I check that pinning is in effect as intended? Thanks and best regards, ================= Frank Schilder AIT Risø Campus Bygning 109, rum S14
On Mon, Mar 15, 2021 at 2:58 AM Frank Schilder <frans@dtu.dk> wrote:
Hi, I seem to have a problem with the extended attributes for MDS pinning. Ceph version is mimic-13.2.10 and the documentation (https://web.archive.org/web/20190716110503/http://docs.ceph.com/docs/mimic/c...) says I should be able to
setfattr -n ceph.dir.pin -v 2 path/to/dir
When I try this, I get the following:
# setfattr -n ceph.dir.pin -v 1 . # getfattr -n ceph.dir.pin . .: ceph.dir.pin: No such attribute
Is this expected? How can I check that pinning is in effect as intended?
IIRC, getfattr support was recently added. What client are you using? -- Patrick Donnelly, Ph.D. He / Him / His Principal Software Engineer Red Hat Sunnyvale, CA GPG: 19F28A586F808C2402351B93C3301A3E258DD79D
Hi Patrick, thanks for your reply. I'm using the kernel client on 5.9.9-1.el7.elrepo.x86_64. Is there a way to check what each MDS is assigned to? Best regards, ================= Frank Schilder AIT Risø Campus Bygning 109, rum S14 ________________________________________ From: Patrick Donnelly <pdonnell@redhat.com> Sent: 15 March 2021 14:49:01 To: Frank Schilder Cc: ceph-users@ceph.io Subject: Re: [ceph-users] MDS pinning: ceph.dir.pin: No such attribute On Mon, Mar 15, 2021 at 2:58 AM Frank Schilder <frans@dtu.dk> wrote:
Hi, I seem to have a problem with the extended attributes for MDS pinning. Ceph version is mimic-13.2.10 and the documentation (https://web.archive.org/web/20190716110503/http://docs.ceph.com/docs/mimic/c...) says I should be able to
setfattr -n ceph.dir.pin -v 2 path/to/dir
When I try this, I get the following:
# setfattr -n ceph.dir.pin -v 1 . # getfattr -n ceph.dir.pin . .: ceph.dir.pin: No such attribute
Is this expected? How can I check that pinning is in effect as intended?
IIRC, getfattr support was recently added. What client are you using? -- Patrick Donnelly, Ph.D. He / Him / His Principal Software Engineer Red Hat Sunnyvale, CA GPG: 19F28A586F808C2402351B93C3301A3E258DD79D
Looking at the client, most of the dir_pin support went in ~v5.1: -------------8<--------------- static bool ceph_vxattrcb_dir_pin_exists(struct ceph_inode_info *ci) { return ci->i_dir_pin != -ENODATA; } static ssize_t ceph_vxattrcb_dir_pin(struct ceph_inode_info *ci, char *val, size_t size) { return ceph_fmt_xattr(val, size, "%d", (int)ci->i_dir_pin); } -------------8<--------------- ...and the parse_reply_info_in() is what sets i_dir_pin: -------------8<--------------- /* dir pin */ if (struct_v >= 2) { ceph_decode_32_safe(p, end, info->dir_pin, bad); } else { info->dir_pin = -ENODATA; } -------------8<--------------- The question is, does the MDS you're using return an inode structure version >=2 ? On Mon, 2021-03-15 at 14:00 +0000, Frank Schilder wrote: Hi Patrick, thanks for your reply. I'm using the kernel client on 5.9.9-1.el7.elrepo.x86_64. Is there a way to check what each MDS is assigned to? Best regards, ================= Frank Schilder AIT Risø Campus Bygning 109, rum S14 ________________________________________ From: Patrick Donnelly <pdonnell@redhat.com> Sent: 15 March 2021 14:49:01 To: Frank Schilder Cc: ceph-users@ceph.io Subject: Re: [ceph-users] MDS pinning: ceph.dir.pin: No such attribute On Mon, Mar 15, 2021 at 2:58 AM Frank Schilder <frans@dtu.dk> wrote:
Hi, I seem to have a problem with the extended attributes for MDS pinning. Ceph version is mimic-13.2.10 and the documentation (https://web.archive.org/web/20190716110503/http://docs.ceph.com/docs/mimic/c...) says I should be able to
setfattr -n ceph.dir.pin -v 2 path/to/dir
When I try this, I get the following:
# setfattr -n ceph.dir.pin -v 1 . # getfattr -n ceph.dir.pin . .: ceph.dir.pin: No such attribute
Is this expected? How can I check that pinning is in effect as intended?
IIRC, getfattr support was recently added. What client are you using? -- Patrick Donnelly, Ph.D. He / Him / His Principal Software Engineer Red Hat Sunnyvale, CA GPG: 19F28A586F808C2402351B93C3301A3E258DD79D _______________________________________________ ceph-users mailing list -- ceph-users@ceph.io To unsubscribe send an email to ceph-users-leave@ceph.io -- Jeff Layton <jlayton@redhat.com>
On Mon, Mar 15, 2021 at 10:42 AM Jeff Layton <jlayton@redhat.com> wrote:
The question is, does the MDS you're using return an inode structure version >=2 ?
Yes, he needs to upgrade to at least nautilus. Mimic is missing commit 8469a81625180668a9dec840293013be019236b8. -- Patrick Donnelly, Ph.D. He / Him / His Principal Software Engineer Red Hat Sunnyvale, CA GPG: 19F28A586F808C2402351B93C3301A3E258DD79D
Thanks for your support.
The question is, does the MDS you're using return an inode structure version >=2 ?
How do I check that? I'm somewhat certain that the pinning actually works, the load distribution between the two active MDSes is consistent with expectations of pinning. Is it possible that setting the pin xattrib succeeded even though I can't query its value? Is there a way to check which MDS is working on what?
Yes, he needs to upgrade to at least nautilus. Mimic is missing commit 8469a81625180668a9dec840293013be019236b8.
Or could I patch and compile from source? Best regards, ================= Frank Schilder AIT Risø Campus Bygning 109, rum S14 ________________________________________ From: Patrick Donnelly <pdonnell@redhat.com> Sent: 15 March 2021 18:43:36 To: Jeff Layton Cc: Frank Schilder; ceph-users@ceph.io Subject: Re: [ceph-users] Re: MDS pinning: ceph.dir.pin: No such attribute On Mon, Mar 15, 2021 at 10:42 AM Jeff Layton <jlayton@redhat.com> wrote:
The question is, does the MDS you're using return an inode structure version >=2 ?
Yes, he needs to upgrade to at least nautilus. Mimic is missing commit 8469a81625180668a9dec840293013be019236b8. -- Patrick Donnelly, Ph.D. He / Him / His Principal Software Engineer Red Hat Sunnyvale, CA GPG: 19F28A586F808C2402351B93C3301A3E258DD79D
participants (3)
-
Frank Schilder
-
Jeff Layton
-
Patrick Donnelly