Re: How to correctly rebase?

[Dr Rainer Woitok <rainer.woitok@gmail.com>]

Warren and All,

On Wednesday, 2015-10-14 11:27:46 -0600, you wrote:

> ...
> The core problem I ran into is that /var/cache/rebase and
> /var/lib/rebase/*.d didn’t exist on my system, and this caused various
> complaints and failures.

On my system all these directories existed, as well as file "/var/cache/
rebase/fullrebase.done".   And I'm pretty  sure I didn't  install any of
these manually,  so I assume  they are  (or at least were) automatically
installed by the "_autorebase" package.

> ...
>     /etc/postinstall/0p_*

After looking at script "0p_000_autorebase.dash" I decided against using
it directly because it does not provide  a means to pass the "--verbose"
option to script "rebaselst".   So I ended up directly running what "0p_
000_autorebase.dash" would have called, saving standard output and stan-
dard error to two files:

   rebaselst --verbose --rebuild update --cleardb rebase

QUESTION: Script "0p_000_autorebase.dash" also passes the options "--no-
aslr" and "peflags" to "rebaselst",  provided the flag file "/var/cache/
rebase/peflags" is existing.  Should I use these options here?

Excerpt from the standard output log file:

removing /var/cache/rebase/rebase_pkg
creating empty /var/cache/rebase/rebase_pkg
Updating package information in /var/cache/rebase/rebase_pkg.
	from /etc/setup/adwaita-icon-theme.lst.gz...
	from /etc/setup/adwaita-themes.lst.gz...
...
	from /etc/setup/_autorebase.lst.gz...
	from /etc/setup/_update-info-dir.lst.gz...
removing /var/cache/rebase/rebase_lst
creating empty /var/cache/rebase/rebase_lst
Updating rebase information for installed dynamic objects in /var/cache/rebase/rebase_lst.
Updating rebase information for installed executables in /var/cache/rebase/rebase_exe.
removing /var/cache/rebase/rebase_dyn
creating empty /var/cache/rebase/rebase_dyn
Looking for dynamic language modules/libraries in:
  /usr/lib/python2.7/site-packages
Updating rebase information for dynamic language modules/libraries /var/cache/rebase/rebase_dyn.
removing /var/cache/rebase/rebase_user
creating empty /var/cache/rebase/rebase_user
Updating rebase information for user-defined dynamic objects /var/cache/rebase/rebase_user.
removing /var/cache/rebase/rebase_user_exe
creating empty /var/cache/rebase/rebase_user_exe
Updating rebase information for user-defined executables /var/cache/rebase/rebase_user_exe.
Rebasing with list /var/cache/rebase/rebase_all, built from /var/cache/rebase/rebase_lst /var/cache/rebase/rebase_dyn /var/cache/rebase/rebase_user.
/usr/share/texmf-dist/tex/plain/gustlib/licz/licz-tst.mex: skipped because not rebaseable
...
/usr/share/texmf-dist/tex/plain/gustlib/plbtx993/test.mex: skipped because not rebaseable
/usr/libexec/coreutils/libstdbuf.so: new base = 32b40000, new size = 10000
/usr/lib/xorg/modules/extensions/cygglx.dll: new base = 32b50000, new size = 60000
...
/usr/lib/python2.7/lib-dynload/select.dll: new base = 37c60000, new size = 10000
...
/usr/lib/python2.7/lib-dynload/operator.dll: new base = 37cd0000, new size = 10000
...
/usr/lib/python2.7/lib-dynload/_socket.dll: new base = 37f10000, new size = 20000
...

End of the  standard ouput log file.   The standard error  log file con-
tains the following:

ReBaseImage (/usr/bin/cygintl-8.dll) failed with last error = 6
ReBaseImage (/usr/bin/cygiconv-2.dll) failed with last error = 6

The following DLLs couldn't be rebased due to errors:
  /usr/bin/cygintl-8.dll
  /usr/bin/cygiconv-2.dll

End of the standard error log file.

QUESTION: Do these errors indicate a problem?

QUESTION: I called "rebaselst" with  the "--cleardb" option which should
cause it to pass "-b 0x70000000" to "rebase" (this is an "i386" system).
Why then are we starting above at a base of 0x32b40000?

And finally, Python still refuses to run, producing at least three diff-
erent error messages when, for instance, calling Mercurial (the "rebase"
output for these three libraries is listed above):

$ hg commit
      1 [main] python2.7 40584 child_info_fork::abort: address space needed by 'select.dll' (0x370000) is already occupied
abort: Resource temporarily unavailable
$ hg commit
      1 [main] python2.7 32648 child_info_fork::abort: unable to remap operator.dll to same address as parent (0x360000) - try running rebaseall
abort: Resource temporarily unavailable
$ hg commit
      1 [main] python2.7 6496 child_info_fork::abort: address space needed by '_ssl.dll' (0x2B0000) is already occupied
abort: Resource temporarily unavailable

QUESTION:What am I supposed to do next?

> ...
> You should not have to reboot.  If you do, it means you’re ignoring
> complaints from setup.exe about files being in use during the
> installation, which could be part of the problem.

There definitely weren't any complaints from "setup_x86.exe".   I didn't
say I HAD to reboot, I just said I did it.

Well, to sum it up: it seems I'm needing more help :-(

Sincerely,
  Rainer

PS: And since I'm still not  on this list because  I can't really handle
the traffic, please also reply to me personally.  Thanks.

 ----------------------------------------------------------------------
| Rainer M Woitok                | Phone : (+49 60 93) 487 95 95       |
| Kolpingstraße 3                | Mobile: (+49 172) 813 6 831         |
| D-63846 Laufach                | Mail  : Rainer.Woitok@Gmail.Com     |
| Germany                        |                                     |
 ----------------------------------------------------------------------

--
Problem reports:       http://cygwin.com/problems.html
FAQ:                   http://cygwin.com/faq/
Documentation:         http://cygwin.com/docs.html
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple