List Info

Thread: A(nother) Windows problem in pass 8




A(nother) Windows problem in pass 8
user name
2006-04-06 19:13:47
Guy Rish wrote:
> Michael,
> 
> 
>>Guy Rish wrote:
>>
>>>Oddly, if I remove the --svnadmin switch it
works.  Sorry 
>>
>>to have not 
>>
>>>dig a bit deeper myself and to have troubled
you.
>>
>>Ahhh.  I see now that you were passing the
*directory* of 
>>svnadmin using the --svnadmin option, not the full
path to 
>>the executable including the filename.  This would
be the 
>>problem.  cvs2svn was trying to execute a directory.
 I guess 
>>the docs should be clearer about this option.
>>
> 
> 
> Good to know.
> 
> 
>>>On a distantly related topic, if I could get
your help on 
>>
>>this one as well:
>>
>>>ERROR: The command '['co', '-q', '-x,v',
'-p1.1', '-kk', 
>>>'c:\\cvsnt\\cvsrepo\\CV
SROOT\\historyinfo,v']' failed with exit 
>>>status: 1 and the following output:
>>>co:
c:\cvsnt\cvsrepo\CVSROOT\historyinfo,v:9: missing ';' 
>>
>>after 'author'
>>
>>>co aborted
>>>
>>>>From what I can gather from digging about
the problem 
>>
>>likely results 
>>
>>>>from
>>>
>>>spaces in the username (an increasingly common
practice on Windows, 
>>>and perhaps something to file into your own
notes as you'll 
>>
>>probably 
>>
>>>run into it again before I will).  The specific
link I found that 
>>>leads me this
>>>direction:
>>>http://archive.netbsd.se/index.php?ml=vi
ewcvs&a=2003-03&t=404060
>>
>>The log shows that it is the "co"
program (part of the RCS 
>>suite) that is producing the error.  The --using-cvs
option 
>>should help.  (This exact problem is mentioned in
the 
>>documentation for --using-cvs.)
>>
> 
> 
> I thought about that too (finally).  And I did find the
cvs2svn bug notes on
> this, because it was a CVS specific error I didn't
look there first,
> foolishly.
> 
> *smirk*
> And now I've got a different pass 8 error...
> 
> C:\temp\cvs2svn-1.3.0>python cvs2svn -v -s
c:\codebase --existing-svnrepos
> --tmp
> dir=c:\temp --use-cvs /cvsnt/cvsrepo
> 
> ----- pass 8 -----
> Starting Subversion Repository.
>
============================================================
> Starting Subversion r1 / 976
>   New Directory trunk
>   New Directory branches
>   New Directory tags
>
============================================================
> Starting Subversion r2 / 976
> Committing 18 CVSRevisions
>   New Directory trunk/CVSROOT
> ERROR: The command '['cvs', '-q', 'co',
'-r1.1', '-p', '-kk',
> 'CVSROOT/historyin
> fo']' failed with exit status: 1
> and the following output:
> cvs [checkout aborted]: no such host C
> 
> 
> C:\temp\cvs2svn-1.3.0>echo %CVSROOT%
> :pserver:Guy Rish12.22.179.124:1401:/cvsnt/cvsrepo
> 
> 
> Is this related to the CVSROOT environment variable? 
When I google for the
> message that's typically where it turns up.  But I get
the same error
> whatever my variable is set to (or not, as I even
deleted it once for test).
> 
> 
> rish

[In the future, please send your emails to the
"users" mailing list so
that the conversation gets archived and other people can
benefit from
any solutions that are found.  I hope you don't mind that I
CC this mail
there.]

I am not a Windows user so I can only speculate.  The
CVSROOT
environment variable is *supposed* to be set by cvs2svn
based on the
path to your repository that you specify as script argument
(i.e., in
this case, /cvsnt/cvsrepo).  That easily explains why the
value you set
has no effect.  I don't see where a supposed host name
"C" could come
from, but presumably your repository is not where cvs2svn
thinks it is.

When using --use-cvs, cvs2svn walks up the directory tree
looking for a
directory that has a subdirectory called 'CVSROOT'
(unrelated to the
environment variable with the same name).  It assumes that
this is the
head of the CVS repository.  Could it be that your CVSROOT
directory is
missing, or that there is one in the wrong place?

CVSROOT is set at line 4850 of the script.  You could add a
debugging
output following that line to see what it is really set to;
for example,

    print repr(os.environ['CVSROOT'])

Do you happen to have a CVS_RSH environment variable set?

Michael

------------------------------------------------------------
---------
To unsubscribe, e-mail: users-unsubscribecvs2svn.tigris.org
For additional commands, e-mail: users-helpcvs2svn.tigris.org

A(nother) Windows problem in pass 8
user name
2006-04-07 15:13:59
Michael,

> 
> Guy Rish wrote:
> > Michael,
> > 
> > 
> >>Guy Rish wrote:
> >>
> >>>Oddly, if I remove the --svnadmin switch it
works.  Sorry
> >>
> >>to have not
> >>
> >>>dig a bit deeper myself and to have
troubled you.
> >>
> >>Ahhh.  I see now that you were passing the
*directory* of svnadmin 
> >>using the --svnadmin option, not the full path
to the executable 
> >>including the filename.  This would be the
problem.  cvs2svn was 
> >>trying to execute a directory.  I guess the
docs should be clearer 
> >>about this option.
> >>
> > 
> > 
> > Good to know.
> > 
> > 
> >>>On a distantly related topic, if I could
get your help on
> >>
> >>this one as well:
> >>
> >>>ERROR: The command '['co', '-q',
'-x,v', '-p1.1', '-kk', 
> >>>'c:\\cvsnt\\cvsrepo\\CV
SROOT\\historyinfo,v']' failed with exit
> >>>status: 1 and the following output:
> >>>co:
c:\cvsnt\cvsrepo\CVSROOT\historyinfo,v:9: missing ';' 
> >>
> >>after 'author'
> >>
> >>>co aborted
> >>>
> >>>>From what I can gather from digging
about the problem
> >>
> >>likely results
> >>
> >>>>from
> >>>
> >>>spaces in the username (an increasingly
common practice on 
> Windows, 
> >>>and perhaps something to file into your own
notes as you'll
> >>
> >>probably
> >>
> >>>run into it again before I will).  The
specific link I found that 
> >>>leads me this
> >>>direction:
> >>>http://archive.netbsd.se/index.php?ml=vi
ewcvs&a=2003-03&t=404060
> >>
> >>The log shows that it is the "co"
program (part of the RCS
> >>suite) that is producing the error.  The
--using-cvs option should 
> >>help.  (This exact problem is mentioned in the
documentation for 
> >>--using-cvs.)
> >>
> > 
> > 
> > I thought about that too (finally).  And I did
find the cvs2svn bug 
> > notes on this, because it was a CVS specific error
I didn't 
> look there 
> > first, foolishly.
> > 
> > *smirk*
> > And now I've got a different pass 8 error...
> > 
> > C:\temp\cvs2svn-1.3.0>python cvs2svn -v -s
c:\codebase 
> > --existing-svnrepos --tmp dir=c:\temp --use-cvs
/cvsnt/cvsrepo
> > 
> > ----- pass 8 -----
> > Starting Subversion Repository.
> >
============================================================
> > Starting Subversion r1 / 976
> >   New Directory trunk
> >   New Directory branches
> >   New Directory tags
> >
============================================================
> > Starting Subversion r2 / 976
> > Committing 18 CVSRevisions
> >   New Directory trunk/CVSROOT
> > ERROR: The command '['cvs', '-q', 'co',
'-r1.1', '-p', '-kk', 
> > 'CVSROOT/historyin fo']' failed with exit
status: 1 and the 
> following 
> > output:
> > cvs [checkout aborted]: no such host C
> > 
> > 
> > C:\temp\cvs2svn-1.3.0>echo %CVSROOT%
> > :pserver:Guy Rish12.22.179.124:1401:/cvsnt/cvsrepo
> > 
> > 
> > Is this related to the CVSROOT environment
variable?  When I google 
> > for the message that's typically where it turns
up.  But I get the 
> > same error whatever my variable is set to (or not,
as I 
> even deleted it once for test).
> > 
> > 
> > rish
> 
> [In the future, please send your emails to the
"users" 
> mailing list so that the conversation gets archived and
other 
> people can benefit from any solutions that are found. 
I hope 
> you don't mind that I CC this mail there.]
> 
> I am not a Windows user so I can only speculate.  The
CVSROOT 
> environment variable is *supposed* to be set by cvs2svn
based 
> on the path to your repository that you specify as
script 
> argument (i.e., in this case, /cvsnt/cvsrepo).  That
easily 
> explains why the value you set has no effect.  I don't
see 
> where a supposed host name "C" could come
from, but 
> presumably your repository is not where cvs2svn thinks
it is.
> 
> When using --use-cvs, cvs2svn walks up the directory
tree 
> looking for a directory that has a subdirectory called 
> 'CVSROOT' (unrelated to the environment variable with
the 
> same name).  It assumes that this is the head of the
CVS 
> repository.  Could it be that your CVSROOT directory is

> missing, or that there is one in the wrong place?
> 
> CVSROOT is set at line 4850 of the script.  You could
add a 
> debugging output following that line to see what it is
really 
> set to; for example,
> 
>     print repr(os.environ['CVSROOT'])
> 
> Do you happen to have a CVS_RSH environment variable
set?
> 
> Michael
> 

No, I don't mind about spamming the list if they don't
mind.  If I must be a
cautionary tale then I'm happy to share to fruits of my
suffering.

Here is the latest:

------------------------------------------------------------
----------------
-----------------------------------

C:\temp\cvs2svn-1.3.0>echo %path%
C:\Program
Files\GnuWin32\bin;C:\Languages\Python\...[clipped]...C
:\Program
Files\Subversion\bin;C:\Program
Files\GNU\WinCvs 1.2;C:\Program Files\CVSNT\

C:\temp\cvs2svn-1.3.0>echo %CVSROOT%
/cvsnt/cvsrepo

C:\temp\cvs2svn-1.3.0>echo %CVS_RSH%
%CVS_RSH%

C:\temp\cvs2svn-1.3.0>python cvs2svn.db.py -q -s
c:\codebase
--existing-svnrepos
 --tmpdir=c:\temp --use-cvs /cvsnt/cvsrepo
'/cvsnt/cvsrepo'
----- pass 1 -----
Examining all CVS ',v' files...
Done
----- pass 2 -----
Checking for blocked exclusions...
Checking for forced tags with commits...
Checking for tag/branch mismatches...
Re-synchronizing CVS revision timestamps...
Done
----- pass 3 -----
Sorting CVS revisions...
Done
----- pass 4 -----
Copying CVS revision data from flat file to database...
Finding last CVS revisions for all symbolic names...
Done
----- pass 5 -----
Mapping CVS revisions to Subversion commits...
WARNING: grouping spans more than 300 seconds
WARNING: grouping spans more than 300 seconds
WARNING: grouping spans more than 300 seconds
WARNING: grouping spans more than 300 seconds
WARNING: grouping spans more than 300 seconds
WARNING: grouping spans more than 300 seconds
WARNING: grouping spans more than 300 seconds
WARNING: grouping spans more than 300 seconds
WARNING: grouping spans more than 300 seconds
WARNING: grouping spans more than 300 seconds
WARNING: grouping spans more than 300 seconds
WARNING: grouping spans more than 300 seconds
Done
----- pass 6 -----
Sorting symbolic name source revisions...
Done
----- pass 7 -----
Determining offsets for all symbolic names...
Done.
----- pass 8 -----
Starting Subversion Repository.
ERROR: The command '['cvs', '-q', 'co', '-r1.1',
'-p', '-kk',
'CVSROOT/historyin
fo']' failed with exit status: 1
and the following output:
cvs [checkout aborted]: no such host C

C:\temp\cvs2svn-1.3.0>cvs -q co -r1.1 -p -kk
CVSROOT/historyinfo
# The "historyinfo" file is used to log the
history file output.
# The filter on the right is invoked with the repository
name.  Its
# standard input contains the history line that has just
been written
# to the history file (if it exists)
#
# The first entry on a line is a regular expression which is
tested
# against the directory that the change is being committed
to, relative
# to the $CVSROOT.  For the first match that is found, then
the remainder
# of the line is the name of the filter to run.
#
# If the repository name does not match any of the regular
expressions in
this
# file, the "DEFAULT" line is used, if it is
specified.
#
# If the name "ALL" appears as a regular
expression it is always used
# in addition to the first matching regex or
"DEFAULT".

------------------------------------------------------------
----------------
-----------------------------------

Notice the last bit.  If, after the cvs2svn script bails out
on me, I
execute the same from the command line I get content. 
That's the most
confounding part to me.

Note the line directly after my script invocation, it is the
debug line you
gave me to insert (which I probably should have added more
text to in order
to make it standout more).  The script should be using the
same CVSROOT that
my command shell is using.

I'm not sure I understand the "no such host C"
problem correctly.  It's a
CVS error that, according to my Google searches, pops only
when the
repository path is fouled.  Now I've some different CVSROOT
settings (just
the path, the fully qualified CVS path with username and
server and such)
but they all produce the same effect, "no such
host".

If you want to touch up the script to dump more information
just let me know
what the changes are and I'll apply them.

Thanks.

rish



------------------------------------------------------------
---------
To unsubscribe, e-mail: users-unsubscribecvs2svn.tigris.org
For additional commands, e-mail: users-helpcvs2svn.tigris.org

[1-2]

about | contact  Other archives ( Real Estate discussion Medical topics )