Cephfs does have repair tools but I wouldn't jump the gun, your metadata pool is probably fine. Unless you're getting health errors or seeing errors in your MDS log?
Are you exporting a fuse or kernel mount with Ganesha (i.e using the vfs FSAL) or using the Ceph FSAL? Have you tried any tests directly on a CephFS mount (taking Ganesha out of the equation)? On Sat, Sep 30, 2017 at 11:09 PM, Marc Roos <m.r...@f1-outsourcing.eu> wrote: > > > I have on luminous 12.2.1 on a osd node nfs-ganesha 2.5.2 (from ceph > download) running. And when I rsync on a vm that has the nfs mounted, I > get stalls. > > I thought it was related to the amount of files of rsyncing the centos7 > distro. But when I tried to rsync just one file it also stalled. It > looks like it could not create the update of the 'CentOS_BuildTag' file. > > Could this be a problem in the meta data pool of cephfs? Does this sound > familiar? Is there something like an fsck for cephfs? > > drwxr-xr-x 1 500 500 7 Jan 24 2016 .. > -rw-r--r-- 1 500 500 14 Dec 5 2016 CentOS_BuildTag > -rw-r--r-- 1 500 500 29 Dec 5 2016 .discinfo > -rw-r--r-- 1 500 500 946 Jan 12 2017 .treeinfo > drwxr-xr-x 1 500 500 1 Sep 5 15:36 LiveOS > drwxr-xr-x 1 500 500 1 Sep 5 15:36 EFI > drwxr-xr-x 1 500 500 3 Sep 5 15:36 images > drwxrwxr-x 1 500 500 10 Sep 5 23:57 repodata > drwxrwxr-x 1 500 500 9591 Sep 19 20:33 Packages > drwxr-xr-x 1 500 500 9 Sep 19 20:33 isolinux > -rw------- 1 500 500 0 Sep 30 23:49 .CentOS_BuildTag.PKZC1W > -rw------- 1 500 500 0 Sep 30 23:52 .CentOS_BuildTag.gM1C1W > drwxr-xr-x 1 500 500 15 Sep 30 23:52 . > > > > > _______________________________________________ > ceph-users mailing list > ceph-users@lists.ceph.com > http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com >
_______________________________________________ ceph-users mailing list ceph-users@lists.ceph.com http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
