Roland

I am getting similar error messages when doing a fresh checkout. It seems that GetComponents is trying to do things in parallel that should be serialised, e.g. assuming that a repository has been checked out when it hasn't and is still inconsistent. I have posted my error message to trac <https://trac.einsteintoolkit.org/ticket/1395>. I assume the problem with the git repositories is similar.

-erik



On Thu, Jun 27, 2013 at 2:35 PM, Roland Haas <roland.haas@physics.gatech.edu> wrote:
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello all,

> It seemed to work much better.  Here is the summary:
>> -----------------------------------------------------------------
>>
>>
162 components checked out.
>> 0 components updated.
>>
>> Unable to process EinsteinAnalysis/Outflow: symlinks may be
>> broken Unable to process EinsteinExact/doc: symlinks may be
>> broken Unable to process KrancNumericalTools/GenericFD: symlinks
>> may be broken Unable to process McLachlan/doc: symlinks may be
>> broken Unable to process Carpet/doc: symlinks may be broken
>> Unable to process GetComponents: symlinks may be broken
>>
>> Summary of Warnings:
>>
>> Could not checkout EinsteinAnalysis/Outflow, unable to switch to
>> tag . Any existing symlinks to EinsteinAnalysis/Outflow will be
>> broken Could not checkout EinsteinExact/doc, unable to switch to
>> tag . Any existing symlinks to EinsteinExact/doc will be broken
>> Could not checkout KrancNumericalTools/GenericFD, unable to
>> switch to tag . Any existing symlinks to
>> KrancNumericalTools/GenericFD will be broken Could not checkout
>> McLachlan/doc, unable to switch to tag . Any existing symlinks to
>> McLachlan/doc will be broken Could not checkout Carpet/doc,
>> unable to switch to tag . Any existing symlinks to Carpet/doc
>> will be broken Could not checkout GetComponents, unable to switch
>> to tag . Any existing symlinks to GetComponents will be broken
>> Time Elapsed: 11 minutes, 28 seconds
Those are exactly all git repositories in the thornlist. Can you check
what git version is used? Can you run with --verbose, please?

Unfortunately I still cannot reproduce this, so cannot test this
myself (Erik can though :-) ). However, looking at the pasted in
output "Do you want to update all existing components? yes, no [no] :
no" it seems as if you are not actually doing a fresh checkout but
instead an update since GetComponents sees some thorns already being
present. So: are you trying to update from an old release or trunk to
Gauss?

Your output has errors stating that it could not switch to a tag (and
since it says "switch to tag ." the tag variable is actually empty).
Eventually the tag is generated from information in REPO_BRANCH and
output from "git tag -l" so it would be good to check what happens
around line 1390 of GetComponents.


Unrelated: trying to download GetComponents using curl as on the
download page "curl -O
https://raw.github.com/gridaphobe/CRL/master/GetComponents" I get
"curl: (60) SSL certificate problem, verify that the CA cert is OK."
and have to add the "-k" option.

Further unrelated: the code in GetComponents (line 1390ff) constructs
a list of branches out of REPO_BRANCH by splitting it at ", ". Does
that mean that REPO_BRANCH could list several branches separated by ",
" (but not eg just a comma "," or ",  ")? I am not sure I understand
that reason for having multiple branches listed there. The code also
uses the basic "`" backtics of perl and not GetComponents' run_command
routine so circumvents logging and verbosity options.

Yours,
Roland

- --
My email is as private as my paper mail. I therefore support encrypting
and signing email messages. Get my PGP key from http://keys.gnupg.net.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.12 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAlHMhdcACgkQTiFSTN7SboUJvACcCGb9qdUGLRzByE8/m5vWGRgG
IsEAoL7TUrsQU1gaBltDEa8Dn0XwD+1e
=HJse
-----END PGP SIGNATURE-----
_______________________________________________
Users mailing list
Users@einsteintoolkit.org
http://lists.einsteintoolkit.org/mailman/listinfo/users



--
Erik Schnetter <schnetter@cct.lsu.edu>
http://www.perimeterinstitute.ca/personal/eschnetter/