Comment 4 for bug 332677

I can confirm that this problem still exists in Ubuntu 9.10

Dec 9 01:26:07 human automount[11065]: using kernel protocol version 5.01
Dec 9 01:26:07 human automount[11065]: lookup_nss_read_master: reading master file /etc/auto.master
Dec 9 01:26:07 human automount[11065]: parse_init: parse(sun): init gathered global options: (null)
Dec 9 01:26:07 human automount[11065]: lookup_read_master: lookup(file): read entry +auto.master
Dec 9 01:26:07 human automount[11065]: lookup_nss_read_master: reading master files auto.master
Dec 9 01:26:07 human automount[11065]: parse_init: parse(sun): init gathered global options: (null)
Dec 9 01:26:07 human automount[11065]: lookup_nss_read_master: reading master ldap auto.master
Dec 9 01:26:07 human automount[11065]: open_lookup:90: cannot open lookup module ldap (/usr/lib/autofs/lookup_ldap.so: undefined symbol: do_verbose)
Dec 9 01:26:07 human automount[11065]: lookup(file): failed to read included master map auto.master
Dec 9 01:26:07 human automount[11065]: no mounts in table

My /etc/auto.master has nothing in it except the default uncommented line (+auto.master at the end of the file).

I have made the following changes to /etc/default/autofs

LOGGING="debug"

LDAP_URI="ldap://localhost"

SEARCH_BASE="ou=automount,dc=mydomain,dc=com"

MAP_OBJECT_CLASS="automountMap"
ENTRY_OBJECT_CLASS="automount"
MAP_ATTRIBUTE="ou"
ENTRY_ATTRIBUTE="cn"
VALUE_ATTRIBUTE="automountInformation"

I have several machines running CentOS 5.3, and everything worked perfectly as soon as I renamed auto.master to auto_bak.master (literally -- install, set up ldap authentication and rename the file. Nothing more.). I assume this means we can rule out LDAP-related problems.

If it makes a difference, this bug is occurring on both of my Ubuntu servers, and they are the OpenLDAP servers. All configuration is done with a non-LDAP user. Slapd confirmed to be up and running from the point of view of a CentOS machine.

LDAP automounts are in the form:

dn: cn=myNfsShare,ou=auto.u,ou=automount,dc=mydomain,dc=com
objectClass: automount
objectClass: top
automountInformation: -fstype=nfs,rw,soft,intr,rsize=8192,wsize=8192 shareHostName:/myNfsShare
cn: myNfsShare
creatorsName: cn=root,dc=com
hasSubordinates: FALSE
structuralObjectClass: automount
subschemaSubentry: cn=Subschema

and maps are in the form:

dn: ou=auto.machines,ou=automount,dc=mydomain,dc=com
objectClass: automountMap
objectClass: top
ou: auto.machines
creatorsName: cn=root,dc=com
hasSubordinates: TRUE
structuralObjectClass: automountMap
subschemaSubentry: cn=Subschema

I have to manually keep my auto.* files in sync on the ubuntu machines until this bug gets fixed, so please let me know if there's any other information you need in order to investigate.