| Document Audience: | INTERNAL |
| Document ID: | I0540-1 |
| Title: | Upgrade of HA 1.x clusters to Sun Cluster 2.2 may fail if HA-NFS is absent. |
| Copyright Notice: | Copyright © 2005 Sun Microsystems, Inc. All Rights Reserved |
| Update Date: | 1999-11-19 |
---------------------------------------------------------------------
- Sun Proprietary/Confidential: Internal Use Only -
---------------------------------------------------------------------
FIELD INFORMATION NOTICE
(For Authorized Distribution by SunService)FIN #: I0540-1
Synopsis: Upgrade of HA 1.x clusters to Sun Cluster 2.2 may fail if HA-NFS is absent.Create Date: Nov/16/99
Keywords:
Upgrade of HA 1.x clusters to Sun Cluster 2.2 may fail if HA-NFS is absent.
Top FIN/FCO Report: No
Products Reference: HA 1.X upgrade failures
Product Category: Server / SW Admin; Storage / SW Admin;
Product Affected:
Mkt_ID Platform Model Description Serial Number
------ -------- ----- ----------- -------------
Systems Affected
----------------
- A14 ALL Ultra 2 -
- E3000 ALL Ultra Enterprise 3000 -
- E3500 ALL Ultra Enterprise 3500 -
- E4000 ALL Ultra Enterprise 4000 -
- E4500 ALL Ultra Enterprise 4500 -
- E5000 ALL Ultra Enterprise 5000 -
- E5500 ALL Ultra Enterprise 5500 -
- E6000 ALL Ultra Enterprise 6000 -
- E6500 ALL Ultra Enterprise 6500 -
X-Options Affected
------------------
X6590A - 112 12.6GB SPARCstorage Array -
X6591A - 112 37.8GB SPARCstorage Array -
X6592A - 112 63.0GB SPARCstorage Array -
X6517A - 114 25.2GB SPARCstorage Array -
X6518A - 114 75.6GB SPARCstorage Array -
X6519A - 114 126GB SPARCstorage Array -
X6594A - 214 75.6GB SPARCstorage Array -
X6593A - 214 176.4GB SPARCstorage Array -
X6595A - 219 164GB SPARCstorage Array -
X6596A - 219 382GB SPARCstorage Array -
X5511A - - DISK 4.2GB/7200 (2*2.1) FWSCSI -
X5506A - - DISK 54.6GB/7200 (6*9.1) FWSCSI -
Parts Affected:
Part Number Description Model
----------- ----------- -----
HA6-1.3-S SOLST HA 1.3 E6000, CD DOC LI
HA5-1.3-S SOLST HA 1.3 E5000, CD DOC LI
HA4-1.3-S SOLST HA 1.3 E4000, CD DOC LI
HA3-1.3-S SOLST HA 1.3 E3000, CD DOC LI
HA2-1.3-S SOLST HA 1.3 E2, CD DOC LI
References:
BugId: 4281365
ESC: 522728
MANUAL: 805-4243-10 Sun Cluster 2.2 Release Notes
MANUAL: (No part number) Sun Cluster 2.2 Release Notes Addendum
URL: http://suncluster.eng.sun.com/engineering/SC2.2/fcs_docs/fcs_docs.html
Issue Description:
During an upgrade of HA 1.x clusters to Sun Cluster 2.2, if the
pre-upgrade system does not include the HA-NFS data service, the
upgrade will fail because the /etc/opt/SUNWhadf/nfs/dfstab.
file does not exist on the machine being upgraded.
This occurs for all HA 1.x clusters that do NOT include the HA-NFS data
service. The error messages look similar to the following:
# /opt/SUNWcluster/bin/scadmin startcluster phys-node-1 clustname
Node specified is phys-node-1
Cluster specified is clustname
Error: Invalid cluster name: clustname
The file "/etc/opt/SUNWcluster/conf/clustname.cdb" was not found
The root cause is the missing file /etc/opt/SUNWhadf/nfs/dfstab..
Failure will occur on all HA 1.x systems that do not include the HA-NFS data
service.
Implementation:
---
| | MANDATORY (Fully Pro-Active)
---
---
| | CONTROLLED PRO-ACTIVE (per Sun Geo Plan)
---
---
| x | REACTIVE (As Required)
---Corrective Action:
Authorized Enterprise Field Service Representatives and Enterprise
Customers can avoid the above mentioned problem encountered during an
upgrade from HA 1.x to Sun Cluster 2.2 by following the recommendations
as shown below:
Use the following procedure to work around the problem, at the time
of upgrade from HA 1.x to Sun Cluster 2.2:
If your pre-upgrade HA 1.x configuration does not include the HA-NFS data
service, you must create /etc/opt/SUNWhadf/nfs/dfstab.tmpl file
(if it does not exist already) before beginning the upgrade process.
Run the following commands as root.
# /usr/bin/mkdir -p /etc/opt/SUNWhadf/nfs 2> /dev/null
# /usr/bin/touch -a /etc/opt/SUNWhadf/nfs/dfstab.tmpl
If you attempt an upgrade while the above file does not exist, the resulting
cluster will have an invalid /etc/opt/SUNWcluster/conf/ccd.database file and
non-existing /etc/opt/SUNWcluster/conf/.cdb file, and the cluster
will not start.
If HA-NFS does not exist on the pre-upgrade cluster, you may see error
messages similar to the following, during startup of the cluster:
# /opt/SUNWcluster/bin/scadmin startcluster phys-node-1 clustname
Node specified is phys-node-1
Cluster specified is clustname
Error: Invalid cluster name: clustname
The file "/etc/opt/SUNWcluster/conf/clustname.cdb" was not found
When such error messages appear, use the following steps to recover.
Run all commands as root.
(In this procedure, the sample pre-upgrade configuration is two-node HA 1.x
cluster. The sample node names are phys-node-1 and phys-node-2. Assume
that phys-node-1 was upgraded to SC 2.2 and is getting the above-mentioned
errors messages while the cluster is starting. Assume that phys-node-2 is
still running HA 1.x and is serving all logical hosts.)
1. Remove the server packages on phys-node-1
(using /opt/SUNWcluster/bin/scinstall).
2. Quit scinstall on phys-node-1
(You must quit scinstall because the SUNWscins package is now un-installed).
3. Create the /etc/opt/SUNWhadf/nfs/dfstab.tmpl file on BOTH NODES
(if it does not exist already), as mentioned earlier.
4. Run the following command on phys-node-2 to make a copy of the /.rhosts
file.
# cp /.rhosts /.rhosts.prev
5. In the /.rhosts file on phys-node-2, create a new line that contains
only the symbol "+", to enable remote access to phys-node-1. Otherwise,
a segmentation violation will occur during post-installation of the
SUNWsccf package (the haupgrd binary will dump core). Run the following
command on phys-node-2.
# echo "+" >> /.rhosts
WARNING : ADDING "+" TO THE /.rhosts FILE CREATES A SECURITY HOLE.
REMOVE THE ENTRY DURING STEP 7. As an alternative to Step 5, add both
host names and the corresponding logical host names to the /.rhosts file
on phys-node-2.
6. Run scinstall again on phys-node-1 to load server packages. Ignore any
errors similar to the following:
Assuming a default cluster name of
/opt/SUNWcluster/bin/scinstall[8]: PDBAPPS=: bad number
7. Once the scinstall has completed successfully, undo the /.rhosts file
changes that you made in Step 5:
# mv /.rhosts.prev /.rhostsComments:
--------------------------------------------------------------------------
Implementation Footnote:
i) In case of MANDATORY FINs, Enterprise Services will attempt to
contact all affected customers to recommend implementation of
the FIN.
ii) For CONTROLLED PROACTIVE FINs, Enterprise Services mission critical
support teams will recommend implementation of the FIN (to their
respective accounts), at the convenience of the customer.
iii) For REACTIVE FINs, Enterprise Services will implement the FIN as the
need arises.
----------------------------------------------------------------------------
All released FINs and FCOs can be accessed using your favorite network
browser as follows:
SunWeb Access:
--------------
* Access the top level URL of http://sdpsweb.ebay/FIN_FCO/
* From there, select the appropriate link to query or browse the FIN and
FCO Homepage collections.
SunSolve Online Access:
-----------------------
* Access the SunSolve Online URL at http://sunsolve.Corp/
* From there, select the appropriate link to browse the FIN or FCO index.
Supporting Documents:
---------------------
* Supporting documents for FIN/FCOs can be found on Edist. Edist can be
accessed internally at the following URL: http://edist.corp/.
* From there, follow the hyperlink path of "Enterprise Services Documenta-
tion" and click on "FIN & FCO attachments", then choose the appropriate
folder, FIN or FCO. This will display supporting directories/files for
FINs or FCOs.
Internet Access:
----------------
* Access the top level URL of https://infoserver.Sun.COM
--------------------------------------------------------------------------
General:
--------
* Send questions or comments to finfco-manager@Sun.COM
---------------------------------------------------------------------------