Re: Best way to repair cygwin?

[LMH <lmh_users-groups at molconn dot com>]

You can select "re-install" in the package manager. Just toggel the icon 
next to "All" (at the top) until it reads "re-install".

I like to keep a back up of anything I have added or configured, such as 
the path folders and such. That way, if I have to wipe out the cygwin 
folder and start again, it is not so much trouble. I would be nice if 
the package manager could dump it's current state to an xml file or 
something like that so I don't have to remember everything I have 
installed. Maybe there is a feature like that, but I'm not aware of it.

I don't know if re-install keeps your configurations or not, so that is 
something to find out about or prepare for.

LMH



Ronald Fischer wrote:
> Cygwin worked well so far. However, starting with today, I get the
> following error - for example when invoking bash:
>
> MUCNL3E6880:~ 2 23 $ bash --norc
>      163 [main] bash 5788 exception::handle: Exception:
>      STATUS_ACCESS_VIOLATION
>     1958 [main] bash 5788 open_stackdumpfile: Dumping stack trace to
>     bash.exe.stackdup
>
> When doing a
>
>     zsh -f
>
> this works at first, but as soon as I do a
>
>    man man
>
> in the shell, I get
>
> 9420794 [main] sh 8004 exception::handle: Exception:
> STATUS_ACCESS_VIOLATION
> Exception: STATUS_ACCESS_VIOLATION at eip=6102048B
> eax=009DD298 ebx=61246414 ecx=758E7B6E edx=003F51F8 esi=00000000
> edi=0022FA10
> ebp=61020C00 esp=0022C7E0 program=C:\cygwin\bin\sh.exe, pid 8004, thread
> main
> cs=001B ds=0023 es=0023 fs=003B gs=0000 ss=0023
> Stack trace:
> Frame     Function  Args
> End of stack trace
> 9455844 [main] sh 5776 exception::handle: Exception:
> STATUS_ACCESS_VIOLATION
> 9458931 [main] sh 5776 open_stackdumpfile: Dumping stack trace to
> sh.exe.stackdump
> 9487143 [main] sh 348 exception::handle: Exception:
> STATUS_ACCESS_VIOLATION
> 9490092 [main] sh 348 open_stackdumpfile: Dumping stack trace to
> sh.exe.stackdump
> 9494562 [main] sh 1860 fork: child -1 - died waiting for longjmp before
> initializatio
> n, retry 0, exit code 0x8B00, errno 11
> sh: fork: retry: Resource temporarily unavailable
> ...
> (and so on, in a endless loop).
>
> I wonder why zsh is invoking "sh.exe" implicitly, when I just call
> "man", but in any case the error message suggests that bash.exe / sh.exe
> is broken.
>
> I had run setup again yesterday evening, in order to install wcd (found
> in the util section). The setup run to the end without error message,
> and I switched off the PC. Today I found bash broken. Since wcd utility
> is an extension to the chdir command, it is conceivable that it fiddled
> around with bash somehow, but I consider it unlikely that this would
> cause the effect I am observing: Such an obvious bug would not have been
> unnoticed.
>
> I wonder whether there is an easy way to "repair" Cygwin (maybe a
> procedure to automatically re-install everything). I searched on the Net
> but didn't find anything in this direction. Any other idea what I could
> do?
>
> Ronald

--
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