[Orca-users] Problem/fix building snapshot r380

[Jon Tankersley]

Good point, the tool has to distinguish between CSV commas and number
commas.  I don't know for sure if Perfmon puts quotes around comma-filled
data elements.  Double check that.  If you have an partial extraction you
can send to the group that might help, not everybody has that format.
 
There are LOTS of problems with the perfmon data that we've seen empirically
- very strange data in headings including null bytes when an installation
didn't do 'everything' correctly (of course that was in one system out of 4,
so.... who knows what the problems were).  
 
You also get NIC names that can be 20-30 characters in length.  This causes
havoc with the legends, etc.  These appear to be set by the vendor of the
hardware, not by Microsoft, so things can and do get messy.

-----Original Message-----
From: orca-users-bounces+jon.tankersley=eds.com at orcaware.com
[mailto:orca-users-bounces+jon.tankersley=eds.com at orcaware.com] On Behalf Of
Renaat Dumon
Sent: Tuesday, August 24, 2004 4:19 PM
To: orca-users at orcaware.com
Subject: RE: [Orca-users] Problem/fix building snapshot r380



Well,

 

Now you've got something started, I also have a couple of comments on r380
with regards to winallator. (I'm not a member of the dev-list)

 

1.	First of all, one should be aware that in some countries, comma's
are used instead of digits for decimal symbols. The current code only
accepts "." (period), whereas my systems use "," (comma). I was able to add
the required substitions in Sourcefile.pm but still. 

2.	Somebody left debug print commands in there that come up while
parsing files. yikes :-) 

3.	There are counters (like for Citrix e.g.) that produce integers as
output,  i.e.  54 ; 87; etc
That apparently is a problem for winallator, since it always expects numbers
like 54.00  (floats). Hence the error "conversion to float incomplete"
errors by the thousands :-o 

4.	This is a typical headerline of a datafile for me:

"(PDH-TSV 4.0) (Romance Standard Time)(-120)"
\\OVERDRIVE\Processor(_Total)\%
<file:///\\OVERDRIVE\Processor(_Total)\%25%20Processor%20Time> Processor
Time

There is a problem with the first column which in the code apparently
substitutes to "timestamp" , in analogy with orcallator. The first run,
things seem to run OK, but consequent runs just bork with some error on how
the "-" sign in this column didn't work out as a valid substraction. Of
course it didn't, it isn't one ! 

 

Maybe someone could slip these on to the dev working on the winallator part
? 

 

Thanks a bunch!

 

Renaat

 


  _____  


From: orca-users-bounces+orca=renaat.com at orcaware.com
[mailto:orca-users-bounces+orca=renaat.com at orcaware.com] On Behalf Of David
Michaels
Sent: Tuesday, August 24, 2004 20:39
To: orca-users at orcaware.com
Subject: [Orca-users] Problem/fix building snapshot r380

 


This may not be the right forum for this, but here goes anyway:  Scold me if
you like.

Was building snapshot r380 on a Solaris 9 box with stock Perl installed.
Trying to incorporate "winallator" into our infrastructure.  Came across
this error when building "version":

dmichael at npd-flurry <packages/version-0.41><218> make ; make test
mkdir blib
mkdir blib/lib
mkdir blib/arch
mkdir blib/arch/auto
mkdir blib/arch/auto/version
mkdir blib/lib/auto
mkdir blib/lib/auto/version
mkdir blib/man3
cp lib/version.pm blib/lib/version.pm
cc -c        -DVERSION=\"0.41\"  -DXS_VERSION=\"0.41\"
-I/usr/perl5/5.00503/sun4-solaris/CORE  util.c
/usr/bin/perl -I/usr/perl5/5.00503/sun4-solaris -I/usr/perl5/5.00503
/usr/perl5/5.00503/ExtUtils/xsubpp  -typemap
/usr/perl5/5.00503/ExtUtils/typemap -typemap typemap version.xs >xstmp.c &&
mv xstmp.c version.c
cc -c        -DVERSION=\"0.41\"  -DXS_VERSION=\"0.41\"
-I/usr/perl5/5.00503/sun4-solaris/CORE  version.c
Running Mkbootstrap for version ()
chmod 644 version.bs
LD_RUN_PATH="" cc -o blib/arch/auto/version/version.so  -G util.o  version.o
chmod 755 blib/arch/auto/version/version.so
cp version.bs blib/arch/auto/version/version.bs
chmod 644 blib/arch/auto/version/version.bs
Manifying blib/man3/version.3
PERL_DL_NONLAZY=1 /usr/bin/perl -Iblib/arch -Iblib/lib
-I/usr/perl5/5.00503/sun4-solaris -I/usr/perl5/5.00503 -e 'use Test::Harness
qw(&runtests $verbose); $verbose=0; runtests @ARGV;' t/*.t
t/01base............Can't locate Test/More.pm in @INC (@INC contains:
blib/arch blib/lib /usr/perl5/5.00503/sun4-solaris
/usr/perl5/5.00503/sun4-solaris /usr/perl5/5.00503/sun4-solaris
/usr/perl5/5.00503 /usr/perl5/5.00503/sun4-solaris
/usr/perl5/5.00503/sun4-solaris /usr/perl5/5.00503
/usr/perl5/site_perl/5.005/sun4-solaris
/usr/perl5/site_perl/5.005/sun4-solaris /usr/perl5/site_perl/5.005 .
/usr/perl5/5.00503/sun4-solaris /usr/perl5/5.00503
/usr/perl5/site_perl/5.005/sun4-solaris /usr/perl5/site_perl/5.005 .) at
t/01base.t line 7.
BEGIN failed--compilation aborted at t/01base.t line 7.
dubious
        Test returned status 2 (wstat 512, 0x200)
FAILED--1 test script could be run, alas--no output ever seen
*** Error code 2
make: Fatal error: Command failed for target `test_dynamic'



The "Test/More.pm" file I found in ../Storable-2.13/t/Test, so I did this:

dmichael at npd-flurry <packages/version-0.41><219> ln -s
../Storable-2.13/t/Test .


And then the "make test" went smoothly.

--Dragon

-------------- next part --------------
An HTML attachment was scrubbed...
URL: </pipermail/orca-users/attachments/20040825/6a2fc39e/attachment.html>