Patch-ID# 115606-04 Keywords: netra, high, availability, suite, saratoga Synopsis: NHAS 2.1 foundation services: core patch for all cluster nodes Date: Dec/03/2004 Install Requirements: See Special Install Instructions Solaris Release: 8 9 SunOS Release: 5.8 5.9 Unbundled Product: Netra High Availability Suite Unbundled Release: 2.1_6/03 Xref: Topic: Relevant Architectures: sparc BugId's fixed with this patch: 4651306 4813153 4846389 4860151 4888517 4889119 4889120 4904926 4918892 4927941 4931439 4935103 4941723 4941749 4951992 4964402 4976748 5008589 5036634 Changes incorporated in this version: 5036634 Patches accumulated and obsoleted by this patch: Patches which conflict with this patch: Patches required with this patch: Obsoleted by: Files included with this patch: /etc/opt/SUNWcgha/init.d/nhcgtp /etc/opt/SUNWcgha/init.d/nhcgtps /etc/opt/SUNWcgha/nhfs.conf.template /etc/rc1.d/HA.K30cmm /etc/rc2.d/HA.S90cmm /etc/rc2.d/S001nhcgtp /etc/rc2.d/S001nhcgtps /etc/rcS.d/HA.K30cmm /etc/rcS.d/HA.S75nhpmd /etc/rcS.d/S75nhpmd /opt/SUNWcgha/lib/adm/CGTP.pm /opt/SUNWcgha/lib/adm/COMMON.pm /opt/SUNWcgha/lib/adm/CONF.pm /opt/SUNWcgha/lib/adm/COPY.pm /opt/SUNWcgha/lib/adm/OBP.pm /opt/SUNWcgha/lib/adm/OS.pm /opt/SUNWcgha/lib/adm/SYNC.pm /opt/SUNWcgha/lib/adm/VAR.pm /opt/SUNWcgha/lib/adm/nhadm.pl /opt/SUNWcgha/lib/libcmm_dl_serial.so.1 /opt/SUNWcgha/lib/libcmmcgtp.so.1 /opt/SUNWcgha/lib/locale/C/LC_MESSAGES/CMM.cat /opt/SUNWcgha/lib/locale/C/LC_MESSAGES/nhadm.mo /opt/SUNWcgha/sbin/nhcmmadm /opt/SUNWcgha/sbin/nhcmmd /opt/SUNWcgha/sbin/nhcmmstat /etc/opt/SUNWcgha/init.d/nhcmm.HA Problem Description: 5036634 during booting of dual servers, telnet connections to clients are disconnected. (from 115606-03) 4651306 Diskless Node Does Not Boot After Failover 4846389 [RFE] Sequential synchronising of partitions 4860151 Diskless ct410 client boot hangs trying to access master-cgtp node 4941723 Wrong CMM_STALE_CLUSTER event received on diskless node when Master is disqualif 4941749 lost of target.conf file and the node goes in a reboot loop. 4976748 NHAS must support full Sanibel chassis (18 nodes), with RR HW and SW (DVD11) 5008589 nhadm check fails (PERL error) when eeprom is not installed (from 115606-02) 4918892 The "lack of membership" timeout that reboots MINs should be configurable 4951992 During booting nhcrfsd failed to stay up and pbind: must specify processor ... 4964402 New SNDR patch 116710-01: nhinstall and checker must be adapted (from 115606-01) 4927941 README.115606-01: Wrong reference to CGTP patch and Solaris version for N240 4931439 Wrong syntax for RNFS.CheckReplicatedSlices in README.115606-01 4904926 CGTP patch does not support Solaris 8 PSR 3 4935103 The checker must not check 108727-22 on S8 PSR3 4813153 No Master Disk Data Results in Endless Reboot No Cluster Master 4889120 Adaptability Data Management Policy is not supported 4888517 nhpmd is no more accessible if /var is used as a mounting point in /etc/vfstab 4889119 Data Management Policy property is tested case-sensitively Patch Installation Instructions: -------------------------------- Refer to the man pages for instructions on using 'patchadd' and 'patchrm' scripts provided with Solaris. Any other special or non-generic installation instructions are described below as special instructions. The following example installs a patch to a standalone machine: example# patchadd /var/spool/patch/104945-02 The following example removes a patch from a standalone system: example# patchrm 104945-02 For additional examples please see the appropriate man pages. Special Install Instructions: ----------------------------- Note that Solaris release 8 HW 7/03 (also known as Solaris 8 PSR3) is supported with this patch, but it requires a specific patch T116036-02 that was not part of the original product distribution. Adding this patch is not described in the upgrade procedure and thus a full installation should be performed. Upgrade of a cluster running Netra HA Suite 2.1 06/03 ===================================================== To install the patches, perform the following procedure: 1. Stop the diskless nodes (e.g. shutdown) if any are running. 2. On the master node, perform the following steps as super-user: a. Prevent the Foundation Services from starting by creating the /etc/opt/SUNWcgha/not_configured file. # touch /etc/opt/SUNWcgha/not_configured b. Reboot the node: # sync # init 6 The vice-master node is elected as the new master node. 3. Repeat step 2 on the new master node. 4. On both master-eligible nodes, perform the following steps as super-user: a. Install the patches: # patchadd -M 116710-01 115606-04 114175-04 b. Delete the /etc/opt/SUNWcgha/not_configured file: # rm /etc/opt/SUNWcgha/not_configured 5. Reboot both master-eligible nodes simultaneously: # sync # init 6 If your cluster does not contain diskless nodes, the installation is complete. If your cluster does contain diskless nodes, perform the following steps: 1. Log in to the master node as super-user 2. For each diskless node, install the core patch: # patchadd -M \ -R /export/root/ 115606-04 3. Ensure that the following files exist and that they are empty: ./etc/hostname. ./etc/hostname. 4. Create the following files and ensure that they are empty: ./etc/dhcp. ./etc/dhcp. 5. Reboot the diskless nodes. Verify that the cluster is operational by running the nhadm check tool on all nodes: # /opt/SUNWcgha/sbin/nhadm check Installation of a cluster using nhinstall ========================================= 1) Check that a Netra HA Suite 2.1 06/03 distribution is present on the installation server (location later referenced as ). 2) Copy the new Netra HA Suite patches 115644-04 114175-04 and 115606-04 to /NetraHAS2.1/Patches, where: 115644-04 = nhinstall and SMCT patch 114175-04 = patch for MEN with SNDR dependency 115606-04 = patch for all nodes 3) Copy the new SNDR patch 116710-01 on /NetraHAS2.1/Patches. 4) Copy the new CGTP patch T116036-02 on /NetraHAS2.1/Patches if you plan to use Solaris 8 HW 7/03. 5) Install the patch 115644-04 on the installation server (to patch nhinstall). patchadd -M /NetraHAS2.1/Patches 115644-04 6) On the directory containing the nhinstall configuration files, add the following lines to the addon.conf file : PATCH=114175-04 NHAS - F LOCAL Y N PATCH=115606-04 NHAS - F LOCAL Y Y Create the addon.conf file if it does not exist. 7) Check that the default values for the new options meet your requirements. Otherwise update the cluster_definition.conf file with the correct settings. 8) Perform the installation as explained in the "Netra High Availability Suite Foundation Services 2.1 06/03 Custom Installation Guide" document. ---------------------------------------------------------------------- FEATURE DETAILS ---------------------------------------------------------------------- Refer to the product documentation that has been updated with patch 114170. ---------------------------------------------------------------------- PATCH RELEASE NOTES ---------------------------------------------------------------------- NOTE 1: ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 4964345 SNDR sets using sector 0 fail, which is not detected by nhinstall/nhadm ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ The RNFS/SNDR component of the Foundation Services cannot operate with replicated slices that use sector 0 of a disk. This limitation is not detected by the nhinstall or nhadm tools. For further information, see the public summary and workaround of bug 4963345. README -- Last modified date: Friday, December 3, 2004