On Dec 4 14:53, bartels wrote: > On 12/04/2013 02:11 PM, marco atzeri wrote: > >Il 12/4/2013 1:35 PM, bartels ha scritto: > >>On 12/04/2013 01:23 PM, marco atzeri wrote: > >> > >>>> > >>>>I am looking for the cause of the problem, so that I know how to prevent > >>>>it. > >>>>Or is the only answer to simply run rebaseall after installation? > >>> > >>>usually yes > >> > >>If that is the case, then why is it not part of the installation? > > > >currently, it is part of the installation > > > > /etc/postinstall/autorebase.bat.done > > > >$ cygcheck -f /etc/postinstall/autorebase.bat > >_autorebase-000444-1 Your attached cygcheck output claims something else. The rebase version should be updated to 4.4.1 as well. > Okay, thanks. I missed that. > It did run, I suppose: Yes, but... > $ cat /etc/postinstall/autorebase.bat.done > @echo off > rem Postinstall scripts are always started from the Cygwin root dir > rem so we can just call dash from here > path .\bin;%path% > dash /bin/rebaseall -p > > > > > >so or last rebase went wrong or something else > >is using cyggcc_s.dll space > > It is intermittent. Does that help? > > > >>I do not understand why it happens on the one Windows 7 machine and not > >>the other. > > > >MS mistery ;-) > > Yikes. Not really funny, but still funny. > > > > >what is the outcome of "rebase -si |grep cyggcc_s" ? > > Ah, that is an interesting one: > > $ rebase -si > rebase: failed to open rebase database "/etc/rebase.db.i386": > No such file or directory ...this here means that rebase never created the database, which in turn could point to rebase crashing or not having sufficient privileges on /etc. Something like that. What happens if you stop all Cygwin processes, including any service you installed, then start dash, make sure you're in /bin, and then call `./rebaseall -p'. Any helpful output? Corinna -- Corinna Vinschen Please, send mails regarding Cygwin to Cygwin Maintainer cygwin AT cygwin DOT com Red Hat