I am experiencing the same problem. Looks like root_squash
has serious issues.
Vlad
On 2/4/22 00:40, Nicola Mori wrote:
Dear Ceph users,
I'm facing a problem when setting the root_squash cap for my
fs client:
# ceph auth get client.wizardfs
[client.wizardfs]
key = . . .
caps mds = "allow rw fsname=wizardfs root_squash"
caps mon = "allow r fsname=wizardfs"
caps osd = "allow rw tag cephfs data=wizardfs"
When I create a test file with one client machine mounting
the fs everything works:
[07:23 mori@farm mori]$ ls -l test.txt
-rw-r--r-- 1 mori lsfuser 21 Feb 4 07:23 test.txt
[07:23 mori@farm mori]$ cat test.txt
Content of test file
But accessing it from another machine using the same client
the file is empty:
[07:34 mori@farm-34 mori]$ ls -l test.txt
-rw-r--r-- 1 mori lsfuser 0 Feb 4 07:23 test.txt
[07:34 mori@farm-34 mori]$ cat test.txt
Removing root_squash from the mds client capabilities makes
everything work as expected.
Both the machines used for the test are just fs clients
running CentOS 7 and Ceph Octopus 15.2.15, and not part of
the cluster (which in turn consists of RockyLinux 8 machines
managed by cephadm and running Ceph Pacific 16.2.7). Both
clients mount the fs via autofs with the kernel driver
using this autofs configuration:
ceph-test -fstype=ceph,name=wizardfs,noatime
172.16.253.2,172.16.253.1:/
I guess this could be a configuration problem but I can't
figure out what I might be doing wrong here. So I'd greatly
appreciate any help or suggestion.
Nicola Mori
_______________________________________________
ceph-users mailing list -- ceph-users@xxxxxxx
To unsubscribe send an email to ceph-users-leave@xxxxxxx
_______________________________________________
ceph-users mailing list -- ceph-users@xxxxxxx
To unsubscribe send an email to ceph-users-leave@xxxxxxx
