On Sat, Dec 29, 2012 at 9:06 AM, Mailingliste
<ml@bokomoko.de> wrote:
Hello,
I run a Debian squeeze NFS Server and a Debian squeeze NFS client. I see a
strange issue, which looks like a cache issue:
I had on the server a directory
drwxrwsr-x 2 dorsch users 4096 15. Feb 2010 shared
on the client that was translated into
drwxrwsr-x 2 nobody users 4096 15. Feb 2010 shared
(I might have had a uid mismatch for a short time for that folder on client
and server).
I wanted to test if it has something todo with the SGID bit on that directory
and did a
# chmod g-s shared
on the server. This had the expected effect on the server
drwxrwxr-x 17 dorsch users 4096 27. Dez 19:25 shared
but the client still shows
drwxrwsr-x 2 nobody users 4096 15. Feb 2010 shared
(see also the file modification date).
The big surprise for me was that this mismatch was even after a reboot of the
client and the server was still there.
Any idea why this information does not get updated on the client is welcome, I
am also happy to provide additional information.
Something else is wrong here ie not properly unmount/mount nfs share. I suspect automount is to blame. Disable it, unmount the share and mount it back manually.
From the NFS client documentation:
acregmin=n The minimum time in seconds that attributes of a regular file should be cached before requesting
fresh information from a server. The default is 3 seconds.
acregmax=n The maximum time in seconds that attributes of a regular file can be cached before requesting
fresh information from a server. The default is 60 seconds.
acdirmin=n The minimum time in seconds that attributes of a directory should be cached before requesting
fresh information from a server. The default is 30 seconds.
acdirmax=n The maximum time in seconds that attributes of a directory can be cached before requesting fresh
information from a server. The default is 60 seconds.
actimeo=n Using actimeo sets all of acregmin, acregmax, acdirmin, and acdirmax to the same value. There
is no default value.
as you can see the default maximum directory attribute cache time is 60 seconds which doesn't match with your resoults. For both v3 and v4. But in v4 there is also:
noac Disable attribute caching, and force synchronous writes. This extracts a server perfor-
mance penalty but it allows two different NFS clients to get reasonable good results when
both clients are actively writing to common filesystem on the server.
fsc Enable the use of persistent caching to the local disk using the FS-Cache facility for
the given mount point.
so are you using maybe fsc option during the mount?
What happens if you switch to NFS v3?
Also post the result of:
# cat /proc/mounts
# nfsstat -m
# nfsstat -n -c -v -4
# uname -a
on the client and:
# nfsstat -s
on the server.
For reference, what I did on the server side:
Setup NFS Server:
=================
# apt-get install nfs-kernel-server
dell:~# cat /etc/exports
# /etc/exports: the access control list for filesystems which may be exported
# to NFS clients. See exports(5).
#
# Example for NFSv2 and NFSv3:
# /srv/homes hostname1(rw,sync,no_subtree_check)
hostname2(ro,sync,no_subtree_check)
#
# Example for NFSv4:
# /srv/nfs4 gss/krb5i(rw,sync,fsid=0,crossmnt,no_subtree_check)
# /srv/nfs4/homes gss/krb5i(rw,sync,no_subtree_check)
#
/home 192.168.2.27(rw,no_root_squash,subtree_check)
dell:~# cat /etc/hosts.deny
# /etc/hosts.deny: list of hosts that are _not_ allowed to access the system.
# See the manual pages hosts_access(5) and hosts_options(5).
#
# Example: ALL: some.host.name, .some.domain
# ALL EXCEPT in.fingerd: other.host.name, .other.domain
#
# If you're going to protect the portmapper use the name "portmap" for the
# daemon name. Remember that you can only use the keyword "ALL" and IP
# addresses (NOT host or domain names) for the portmapper, as well as for
# rpc.mountd (the NFS mount daemon). See portmap(8) and rpc.mountd(8)
# for further information.
#
# The PARANOID wildcard matches any host whose name does not match its
# address.
# You may wish to enable this to ensure any programs that don't
# validate looked up hostnames still leave understandable logs. In past
# versions of Debian this has been the default.
# ALL: PARANOID
portmap:ALL
lockd:ALL
mountd:ALL
rquotad:ALL
statd:ALL
dell:~# cat /etc/hosts.allow
# /etc/hosts.allow: list of hosts that are allowed to access the system.
# See the manual pages hosts_access(5) and hosts_options(5).
#
# Example: ALL: LOCAL @some_netgroup
# ALL: .foobar.edu EXCEPT terminalserver.foobar.edu
#
# If you're going to protect the portmapper use the name "portmap" for the
# daemon name. Remember that you can only use the keyword "ALL" and IP
# addresses (NOT host or domain names) for the portmapper, as well as for
# rpc.mountd (the NFS mount daemon). See portmap(8) and rpc.mountd(8)
# for further information.
#
portmap: 192.168.2.27
lockd: 192.168.2.27
rquotad: 192.168.2.27
mountd: 192.168.2.27
statd: 192.168.2.27
dell:~#
Make UID matching work with NFSv4:
----------------------------------
dell:/etc/default# diff nfs-common.orig nfs-common
16c16
< NEED_IDMAPD=
---
> NEED_IDMAPD=yes
19c19
For reference, what I did on the client side (happens also when I use without
autoFS):
Mount NFS volumes with AutoFS:
==============================
support automount nfs volumes: /mnt/net/<hostname>/<path>:
----------------------------------------------------------
root@omjuta:~# cat /etc/auto.master.d/net.autofs
/mnt/net -hosts --timeout=60
root@omjuta:~#
# ls -l /mnt/net/dell/home
Make UID matching work with NFSv4:
----------------------------------
root@omjuta:/etc# diff idmapd.conf.orig idmapd.conf
10a11,13
>
> [Translation]
> Method=nsswitch
root@omjuta:/etc#
root@omjuta:~# diff /etc/default/nfs-common.orig /etc/default/nfs-common
16c16
< NEED_IDMAPD=
---
> NEED_IDMAPD=yes
19c19
< NEED_GSSD=
---
> NEED_GSSD=no
root@omjuta:~#
Many thanks,
Rainer
--
To UNSUBSCRIBE, email to debian-user-REQUEST@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmaster@lists.debian.org
Archive: [🔎] 201212282306.30052.ml@bokomoko.de" target="_blank">http://lists.debian.org/[🔎] 201212282306.30052.ml@bokomoko.de