I noticed that OptiPlex 990, 790, and Precision T1600 have signed BIOS updates available. There's an option that gets added to the BIOS before installing a signed version. For example, A05 on the 990 adds an Enable Signed Firmware Update option to the Security section of the BIOS menu. There's no way to enable that option through CCTK (or OMCI or DCCU or DTK). Are those tools going to get updated so that this option can get set through automation? I assume the HAPI dcdbas32/64 driver needs to be updated to support the option since all the tools use that driver (well CCTK and OMCI do).
OK thanks Warren. I wish it had been enabled by default. In the meantime, if you could find out the token value for this setting, then I could just set it that way.
Some of this thread was cut off due to being posted after the migration date to the new forum. I saved the thread just in case. Anyway, I've noticed a number of issues.
On OptiPlex 960s, it looks like the feature was added starting with the A13 BIOS according to the release notes. I tried --istokenactive=0x0325 to see if the feature was reported correctly. Unfortunately this froze the OS almost immediately. I also tried --token=0x0325 and that resulted in the same freeze/hang/lock up behavior. After rebooting and checking the BIOS, I can see the feature was not enabled. I've tested this with CCTK 2.0.1 on Vista SP2 x86 and Windows 7 SP1 x86 and x64 so I don't think the OS matters. I tested on A11, A13, and A14 BIOSes. Since the feature was not present in A11, the command run as expected reporting the state byte is not available. A13 and A14 BIOSes result in an OS freeze when using CCTK though. I'm not sure if this is a BIOS issue or CCTK issue but since this mostly works for other models I've tested, I think it is probably a BIOS issue.
The second issue is with the Latitude E6420. It looks like the feature was added in the A05 version of the BIOS according to the release notes. Unfortunately --istokenactive and --token don't detect 0x0325 as being a valid token. I manually checked the BIOS dialog and the feature is there. Once I installed A06, then CCTK reported the token as being valid. Seems like a weird issue. My guess is that the Latitude E5420 and Latitude E5320 also have this same problem because their A05 and A06 BIOSes were released at the exact same time as the E6420. Unfortunately I don't have the hardware to test those two models but I'm trying to find someone who can confirm if the problem exists on those specific models too.
I just did some testing with a Latitude D630. CCTK can enable the setting successfully. I checked the BIOS setup dialog to confirm it. Unfortunately the istokenactive command does not detect the token as being active. This seems like an issue with CCTK. Perhaps it is the same bug that is affecting the Latitude E6420. I think the 960 issue is probably with the BIOS though.
Sorry, mis-read your post. I checked with the CCTK team and they stated that some of the desktop BIOS's don't have support for the token for CCTK but can be set manually.
The D630 has the token in A18. CCTK can enable the feature with the token. The problem is that istokenactive does not work correctly with that token. If the token was missing then --token would also not work. I'm still convinced there's a bug with the D630. I can double check the SMBIOS tables for the one that appears to have token number in it.
The Latitude E6420 is missing the token until a later version. --token and --istokenactive both don't work for A05 because the token is missing. They both work in A06 because the token exists. There's not much to do for this other than having it documented so people know this behavior.
New Whitepaper on "Dell Client BIOS: Signed Firmware Update" that probably explains many of the questions on this thread. en.community.dell.com/.../20287278.aspx
DELL-Warren B
1 Rookie
•
1124 Posts
2049
0
Posted November 1st, 2011 22:00