Known Problems

In addition to known problems documented in Sun Cluster 3.0 12/01 Release Notes, the following known problems affect the operation of the Sun Cluster 3.0 12/01 release.

Bug ID 4362925

Problem Summary:

nodeA# scshutdown -g0 -y
scshutdown: Unmount of /dev/md/sc/dsk/d30 failed: Device busy.
scshutdown: Could not unmount all PxFS filesystems.

The Networker packages were bundled and installed during the Oracle installation. Therefore, the nsrmmd daemon is running and mounting the /global/oracle directory, which prevents the unmount of all cluster file systems.

nodeA# umount /global/oracle
umount: global/oracle busy
nodeA# fuser -c /global/oracle
/global/oracle: 335co 317co 302co 273co 272co
nodeA# ps -ef|grep 335
 root 335 273 0 17:17:41 ?       0:00 /usr/sbin/nsrmmd -n 1
 root 448 397 0 17:19:37 console 0:00 grep 335

This problem occurs during Sun Cluster shutdown when the shutdown tries to unmount a cluster file system that the process nsrmmd is still referencing.

Workaround: Run the fuser(1M) command on each node to establish a list of all processes still using the cluster file systems that cannot be unmounted. Check that no Resource Group Manager resources have been restarted since the failed scshutdown(1M) command was first run. Kill all these processes with the kill -9 command. This kill list should not include any processes under the control of the Resource Group Manager. After all such processes have terminated, rerun the scshutdown command, and the shutdown should run to successful completion.

Bug ID 4525403

Problem Summary: VxFS installation adds the following line to the /etc/system file:

* vxfs_START -- do not remove the following lines:
*
* VxFS requires a stack size greater than the default 8K.
* The following values allow the kernel stack size
* for nfs threads to be increased to 16K.
*
set rpcmod:svc_default_stksize=0x4000
* vxfs_END

If you install VxFS after Sun Cluster software has been installed, the VxFS installation overrides the following required Sun Cluster 3.0 12/01 setting in the /etc/system file:

* Start of lines added by SUNWscr
set rpcmod:svc_default_stksize=0x6000

Workaround: Change the setting value from 0x4000 to 0x6000.

set rpcmod:svc_default_stksize=0x6000

Bug ID 4511699

Problem Summary: Sun Cluster HA for NFS requires files [SUCCESS=return] for the hosts lookup entry in the /etc/nsswitch.conf file, and requires that all cluster private IP addresses be present in the /etc/inet/hosts file on all cluster nodes.

Otherwise, Sun Cluster HA for NFS will not be able to fail over correctly in the presence of public network failures.

Workaround: Perform the following steps on each node of the cluster.

1. Modify the hosts entry in the /etc/nsswitch.conf file so that, upon success in resolving a name locally, it returns success immediately and does not contact NIS or DNS.

   hosts: cluster files [SUCCESS=return] nis dns

2. Add entries for all cluster private IP addresses to the /etc/inet/hosts file.

You only need to list the IP addresses plumbed on the physical private interfaces in the /etc/nsswitch.conf and /etc/inet/hosts files. The logical IP addresses are already resolvable through the cluster nsswitch library.

To list the physical private IP addresses, run the following command on any cluster node.

% grep ip_address /etc/cluster/ccr/infrastructure

Each IP address in this list must be assigned a unique hostname that does not conflict with any other hostname in the domain.

Sun Cluster software already requires that any HA IP addresses (LogicalHostname/SharedAddresses) be present in /etc/inet/hosts on all cluster nodes and that files is listed before nis or dns. The additional requirements mandated by this bug is to also list [SUCCESS=return] after files and to list all cluster private IP addresses in the /etc/inet/hosts file.

Bug ID 4626010

Problem Summary: In a Sun StorEdge 3900 Series system, the preconfigured, default hard zones that are used on the Sun StorEdge Network FC Switch-8 and Switch-16 switches are incompatible for use with host-based mirroring in a cluster. The hard zones prevent the cluster nodes from seeing all attached storage because there are only two ports per zone.

Workaround: Remove the default hard zones from the Sun StorEdge Network FC Switch-8 and Switch-16 switches (this is documented as a necessary step in the procedures in Appendix C, Installing and Maintaining a Sun StorEdge 3900 or 6900 Series System. See the SANbox-8/16 Switch Management User's Manual for instructions on removing the preconfigured hard zones on all Sun StorEdge Network FC Switch-8 and Switch-16 switches.