Environment
Novell Open Enterprise Server 1 (OES 1) Linux
Novell Open Enterprise Server 2 (OES 2) Linux
Novell Open Enterprise Server 2 SP1 (OES 2 SP1) Linux
Novell Open Enterprise Server 2 (OES 2) Linux
Novell Open Enterprise Server 2 SP1 (OES 2 SP1) Linux
Situation
Error -676 When Trying To Create A Pool Or Volume via NSSMU or iManager
Error -676 When Trying To Create/Modify the NSSADMIN User In YAST Or At The Command Line
ERROR: NSS Admin Failed Creating Edirectory Objects
Server has multiple NICs configured
eDirectory is only configured to communicate on one NIC
eDirectory is synchronizing with other eDirectory server without any errors.
eDirectory LDAP (NLDAP) is communicating without errors.
Creation of objects in eDirectory other than with NSS utilities is working without errors.
Error -676 When Trying To Create/Modify the NSSADMIN User In YAST Or At The Command Line
ERROR: NSS Admin Failed Creating Edirectory Objects
Server has multiple NICs configured
eDirectory is only configured to communicate on one NIC
eDirectory is synchronizing with other eDirectory server without any errors.
eDirectory LDAP (NLDAP) is communicating without errors.
Creation of objects in eDirectory other than with NSS utilities is working without errors.
Resolution
Option 1:
Disable the second Network Interface Card (NIC) in the server and make sure the enabled NIC has the proper host name assigned to it. Restart network services or the server.
Option 2
Make sure the /etc/hosts file is correct and that you don't have multiple references to one NIC with multiple ip addresses.
Option 3
Delete the NSS admin user from the context in which the server resides and then add NSS admin user manually using command line
"nssAdminInstall -a <admin name> -p <admin password> -o <nss admin object name>"
The -a is for specifying a user with admin rights. (e.g. admin.novell)
The -p is for specifying the password.
The -o is for specifying the nss admin object name. (e.g. servernameadmin.novell)
Once this is created we should be able to create pools and volumes with either iManager or NSSMU.
Option 4
Create the volume and pool objects manually using Console One 1.36h.
Option 5
It has been reported that this issue can be seen on multi-homed servers (servers with multiple ip addresses configured). As reported by the customer, they had restored a server with only the primary ip address configured--the one that it previously had (as the other addresses were for iSCSI connections no longer needed). Even though the primary ip was up and active and eDirectory showed no issues, NSSMU gave this same error. After checking the nds.conf they discovered that it contained the two iSCSI addresses. After removing them from the config and restarting ndsd, they reran through the OES install to make sure NSS and any other possible components configured themselves correctly based on the configured address. After doing this the update nds option worked without issue.
Option 6
The fix for this issue is expected to ship in OES 2 SP2.
Disable the second Network Interface Card (NIC) in the server and make sure the enabled NIC has the proper host name assigned to it. Restart network services or the server.
Option 2
Make sure the /etc/hosts file is correct and that you don't have multiple references to one NIC with multiple ip addresses.
Option 3
Delete the NSS admin user from the context in which the server resides and then add NSS admin user manually using command line
"nssAdminInstall -a <admin name> -p <admin password> -o <nss admin object name>"
The -a is for specifying a user with admin rights. (e.g. admin.novell)
The -p is for specifying the password.
The -o is for specifying the nss admin object name. (e.g. servernameadmin.novell)
Once this is created we should be able to create pools and volumes with either iManager or NSSMU.
Option 4
Create the volume and pool objects manually using Console One 1.36h.
Option 5
It has been reported that this issue can be seen on multi-homed servers (servers with multiple ip addresses configured). As reported by the customer, they had restored a server with only the primary ip address configured--the one that it previously had (as the other addresses were for iSCSI connections no longer needed). Even though the primary ip was up and active and eDirectory showed no issues, NSSMU gave this same error. After checking the nds.conf they discovered that it contained the two iSCSI addresses. After removing them from the config and restarting ndsd, they reran through the OES install to make sure NSS and any other possible components configured themselves correctly based on the configured address. After doing this the update nds option worked without issue.
Option 6
The fix for this issue is expected to ship in OES 2 SP2.
Additional Information
RESOLVED IN OES 2 SP 2
Formerly known as TID# 10100343
Formerly known as TID# 10100343