Thanks for the information.
We started enabling specific firmware years ago because we started with Sunray2's and knew that we
would never have any of the CoronaP1...P7 models at site. To us, it just seems like bad practice to
load up a bunch of code that will never (legitimately) be required, i.e. we were reducing our threat
surface.
As for the docs, they seem a bit sparse on the entire firmware issue, Perhaps this is because of the
recent split into the separate "Sun Ray Operating System" product? If so, perhaps this will be
improved in the future.
re: But now, new units that are not yet at the desired firmware version might not be able to upgrade
properly.
I guess we're fortunate that this doesn't seem to have been the case in this instance, but we'll
have to alter our install notes to avoid this problem in the future.
Sean Walmsley
On 07/13/12 08:47, Jörg Barfurth wrote:
Am 12.07.12 16:42, schrieb Sean Walmsley:
We deliberately didn't enable the extCoronaP9 initially because we
couldn't find any reference to it being required in the docs. When we
added it later, it loaded just fine.
You should never have to configure use of individual firmware files, except for
very special
circumstances.
Is there any place in the docs advising you to enable or disable specific
firmware files for any
purpose?
My question is: what is the extCoronaP9 firmware file for? The SunRay3+
units seem to work properly with only the updated CoronaP9 firmware file
installed, and adding the "ext" file changes the machine description
(from [STOP]-[V]) from "SunrayP9" to "SunrayP9-E". What features does
the extra ~750Kb of code contain, or put another way, what is the
difference between a P9 unit and a P9-E unit?
The ext* file contains model-specific hardware drivers, so when the ext file is
installed, the unit
makes better use of its hardware (to accelerate some features).
To answer your specific question, after utfwinstall we did:
# clear out all firmware links in /tftpboot to get rid of old
CoronaP1..P7 links
/opt/SUNWut/sbin/utfwadm -R
# add in firmware for Sunray 2, 3+, 3
/opt/SUNWut/sbin/utfwadm -A -a -V -G force -f
/opt/SUNWutdfw/lib/firmware/CoronaP8
/opt/SUNWut/sbin/utfwadm -A -a -V -G force -f
/opt/SUNWutdfw/lib/firmware/CoronaP9
/opt/SUNWut/sbin/utfwadm -A -a -V -G force -f
/opt/SUNWutdfw/lib/firmware/CoronaP10
If you point utfwadm to a specific file, it will make available only that file.
You should never
need to use the -f option at all. If you do, you should use it with a directory
containing the full
set of firmware files for a version of SRS. Using it with individual files
should be reserved to
special cases, typically at the request of support personnel.
If you use
/opt/SUNWut/sbin/utfwadm -A -a -V -G force
or
/opt/SUNWut/sbin/utfwadm -A -a -V -G force \
-f /opt/SUNWutdfw/lib/firmware
that should do the right thing.
After this, the Sunray3+ units downloaded the new CoronaP9 ([STOP]-[V]
showed the correct new version number) and seemed to work just fine with
the exception of complaining about attempting to download extCoronaP9 on
each reboot.
We then did:
/opt/SUNWut/sbin/utfwadm -A -a -V -G force -f
/opt/SUNWutdfw/lib/firmware/extCoronaP9
following which the SunRay3+ units downloaded the "ext" firmware and
changed their description (from [STOP]-[V]) from SunrayP9 to SunRayP9-E.
But now, new units that are not yet at the desired firmware version might not
be able to upgrade
properly.
So far, we haven't been able to discern any difference between our
single upgraded "P9-E" unit and a "regular" P9 unit (although they are
in different physical locations so it's hard to do a good comparison)
See above.
Regards,
- Jörg
_______________________________________________
SunRay-Users mailing list
[email protected]
http://www.filibeto.org/mailman/listinfo/sunray-users
