[OmniOS-discuss] Internal pkg error during a test r151010 to r151014 upgrade

[Dan McDonald]

We've had one other person complain about update problems, and his was a false-positive hit on an anti-virus middlebox.

You don't have such beasts in your network, do you?

> On Apr 6, 2015, at 3:38 PM, Chris Siebenmann <cks at cs.toronto.edu> wrote:

<SNIP!>

> Update Phase                              3141/16901Action upgrade failed for 'opt/gcc-4.8.1/lib/amd64/libstdc++.so.6.0.18-gdb.py' (pkg://omnios/developer/gcc48):
> TypeError: 'NoneType' object is not callable

It's like you got an empty file for this one.

My 014 box does have it:

-bash-4.3$ pkg list -v gcc48
FMRI                                                                         IFO
pkg://omnios/developer/gcc48@4.8.1-0.151014:20150402T205830Z                 i--
-bash-4.3$ pkg contents gcc48 | grep /libstdc++.so.6.0.18-gdb.py
opt/gcc-4.8.1/lib/amd64/libstdc++.so.6.0.18-gdb.py
opt/gcc-4.8.1/lib/libstdc++.so.6.0.18-gdb.py
-bash-4.3$ 

But that just means I got it the time I tried it.

<SNIP!>

>  File "/usr/lib/python2.6/vendor-packages/pkg/client/pkgplan.py", line 441, in execute_update
>    dest.install(self, src)
>  File "/usr/lib/python2.6/vendor-packages/pkg/actions/file.py", line 193, in install
>    stream = self.data()

That's with the old file.py, BTW.  It's barfing because self.data() is being hinky.

> I have this virtual machine still running at the moment so I can poke
> around inside the mounted clone for people. Although I haven't tested
> yet for obvious reasons, I also have a virtual-disk snapshot from before
> I started my test so I expect this is going to be fully reproducable.
> 
> If people want I can make our customization scripts available so people
> can see exactly what we do to our poor OmniOS r151010 systems. (Some of
> the bits are probably slightly gory and improper.)

I doubt your "improper" behavior changes the version of python the pkg(5) system uses, but if it does, that's one other possibility.

Dan