Autofs 5 failed unmounting Solaris 10 NFS share
Hi,
I am seeing autofs 5 on RHEL/CentOS 5 failed auto umounting NFS shares served by Solaris 10 11/06 server. However, other NFS shares got unmounted just fine after expired. All automount maps are pulled from LDAP servers. All NFS shares are auto mounted under /nfs.
Here are some information:
1. autofs version: autofs-5.0.1-0.rc2.43.0.2
2. kernel version: 2.6.18-8.1.8.el5xen
3. auto.master:
===============================================================================
/misc /etc/auto.misc
/net -hosts
# Include central master map if it can be found using
# nsswitch sources. +auto.ldap ===============================================================================
4. auto.ldap (automount map in LDAP server):
===============================================================================
dn: ou=auto.master,ou=ldap,ou=autofs,ou=services,dc=example,dc=com
objectClass: top
objectClass: automountMap
ou: auto.master
dn: cn=/nfs,ou=auto.master,ou=ldap,ou=autofs,ou=services,dc=example,dc=com automountInformation:
ldap:ou=auto.nfs,ou=ldap,ou=autofs,ou=services,dc=example ,dc=com
objectClass: top
objectClass: automount
cn: /nfs
# share1 dn: cn=share1,ou=auto.nfs,ou=ldap,ou=autofs,ou=services,dc=example,dc=com objectClass: top objectClass: automount automountInformation: -fstype=nfs,rw,hard,intr,nosuid,nfsvers=3,udp server1:/export/zfs/nshare1 cn: share1
===============================================================================
5. /etc/sysconfig/autofs
===============================================================================
DEFAULT_MASTER_MAP_NAME="auto.master"
DEFAULT_TIMEOUT=300
DEFAULT_BROWSE_MODE="no"
DEFAULT_LOGGING="debug"
# Define the default LDAP schema to use for lookups DEFAULT_MAP_OBJECT_CLASS="automountMap" DEFAULT_ENTRY_OBJECT_CLASS="automount" DEFAULT_MAP_ATTRIBUTE="ou" DEFAULT_ENTRY_ATTRIBUTE="cn" DEFAULT_VALUE_ATTRIBUTE="automountInformation"
DEFAULT_AUTH_CONF_FILE="/etc/autofs_ldap_auth.conf"
# General global options
OPTIONS=""
===============================================================================
6. /var/log/messages:
===============================================================================
....
Jul 26 10:26:26 client1 automount[21073]: expiring path /home/joe Jul 26 10:26:26 client1 automount[21073]: unmounting dir = /home/joe Jul 26 10:26:26 client1 automount[21073]: expired /home/joe Jul 26 10:26:39 client1 automount[21073]: expiring path /nfs/share1 Jul 26 10:26:39 client1 automount[21073]: umount_multi: could not stat fs of /nfs/share1 Jul 26 10:26:39 client1 automount[21073]: do_expire: couldn't complete
expire of /nfs/share1
....
===============================================================================
7: /var/log/debug:
===============================================================================
.....
Jul 26 10:26:26 client1 automount[21073]: st_expire: state 1 path /home Jul 26 10:26:26 client1 automount[21073]: expire_proc: exp_proc = 3076074384 path /home Jul 26 10:26:26 client1 automount[21073]: expire_proc_indirect: expire /home/joe Jul 26 10:26:26 client1 automount[21073]: handle_packet: type = 4 Jul 26 10:26:26 client1 automount[21073]: handle_packet_expire_indirect: token 1900, name joe Jul 26 10:26:26 client1 automount[21073]: expiring path /home/joe Jul 26 10:26:26 client1 automount[21073]: umount_multi: path /home/joe incl 1 Jul 26 10:26:26 client1 automount[21073]: unmounting dir = /home/joe Jul 26 10:26:26 client1 automount[21073]: rm_unwanted_fn: removing directory /home/joe Jul 26 10:26:26 client1 automount[21073]: expired /home/joe Jul 26 10:26:26 client1 automount[21073]: send_ready: token = 1900 Jul 26 10:26:26 client1 automount[21073]: expire_cleanup: got thid 3076074384 path /home stat 0 Jul 26 10:26:26 client1 automount[21073]: expire_cleanup: sigchld: exp 3076074384 finished, switching from 2 to 1 Jul 26 10:26:26 client1 automount[21073]: st_ready: st_ready(): state = 2 path /home Jul 26 10:26:39 client1 automount[21073]: st_expire: state 1 path /nfs Jul 26 10:26:39 client1 automount[21073]: expire_proc: exp_proc = 3076074384 path /nfs Jul 26 10:26:39 client1 automount[21073]: expire_proc_indirect: expire /nfs/share1 Jul 26 10:26:39 client1 automount[21073]: handle_packet: type = 4 Jul 26 10:26:39 client1 automount[21073]: handle_packet_expire_indirect: token 1901, name share1 Jul 26 10:26:39 client1 automount[21073]: expiring path /nfs/share1 Jul 26 10:26:39 client1 automount[21073]: umount_multi: path /nfs/share1 incl 1 Jul 26 10:26:39 client1 automount[21073]: umount_multi: could not stat fs of /nfs/share1 Jul 26 10:26:39 client1 automount[21073]: do_expire: couldn't complete expire of /nfs/share1 Jul 26 10:26:39 client1 automount[21073]: send_fail: token = 1901 Jul 26 10:26:39 client1 automount[21073]: expire_proc_indirect: 1 remaining in /nfs Jul 26 10:26:39 client1 automount[21073]: mount still busy /nfs Jul 26 10:26:39 client1 automount[21073]: expire_cleanup: got thid 3076074384 path /nfs stat 2 Jul 26 10:26:39 client1 automount[21073]: expire_cleanup: sigchld: exp 3076074384 finished, switching from 2 to 1 Jul 26 10:26:39 client1 automount[21073]: st_ready: st_ready(): state =
2 path /nfs
===============================================================================
Anyone sees similar problem?
Simon