HP-UX Software Assistant Administration Guide (5900-3003, March 2013)

1. Using a system with Internet access (this system may be running Linux or Windows), download
the catalog from the HPSC.
1. Get /opt/swa/lbin/swaFetch.jar from an HP-UX system running SWA version
C.02.80 or later and transfer the file to the system that will be running the download.
2. On the system to do the download, run the following command:
# java -jar swaFetch.jar -x hp_id=uname \
-x hp_pw=pw -x proxy=http://user:[e-mail]:8088 \
-x file=/export/patches/swa_catalog.xml.gz
3. When swaFetch.jar has completed, the current directory will contain the catalog
swa_catalog.xml.gz.
2. Transfer the catalog to the system to be analyzed using ssh or media. The catalog's default
location is $HOME/.swa/cache. Uncompress the file with
# gunzip swa_catalog.xml.gz
3. Create an inventory, run an analysis, and generate a report on the system with
# swa report -x catalog_max_age=-1
The catalog_max_age=-1 extended option setting instructs SWA to skip the catalog
download step. Note that you can use the extended option catalog to specify the catalog
location if it is other than the default $HOME/.swa/cache/swa_catalog.xml.
4. Evaluate the reports and determine the patches to be downloaded.
5. Contact the HPSC from a system connected to the Internet and select the patches you wish to
install. Once you have a selected patch list, download them in your desired format. HP
recommends using the depot creation script included with the patches since it will make
installation easier.
Note that when using media or other means to relocate the swcache files to a new system
(the swa get and swa step download commands are not used), the MD5 cryptographic
hash validation of the patches is not repeated.
6. Continue with the patch installation procedure as outlined in Chapter 3 (page 11).
For more information, see the Security Considerations section of swa(1M).
Running SWA on a system without access to the Internet 25