{chem} [ictce-7.3.5] QuantumESPRESSO 5.1.2 (REVIEW)

[sylmarien]

Hi, here is the latest QuantumESPRESSO built with the latest ictce toolchain.

We have it built on our side on Debian, please tell us if there are things to change.

[hpcugentbot]

Automatic reply from Jenkins: Can I test this?

[boegel]

Jenkins: ok to test

[hpcugentbot]

Refer to this link for build results (access rights to CI server needed):
https://jenkins1.ugent.be/job/easybuild-easyconfigs-pr-builder/3136/
Easyconfigs unit test suite PASSed (see https://jenkins1.ugent.be/job/easybuild-easyconfigs-pr-builder/3136/console for more details).

This pull request is now ready for review/testing.

Please try and find someone who can tackle this; contact @boegel if you're not sure what to do.

[boegel]

Test report by @boegel
FAILED
Build succeeded for 0 out of 1 (1 easyconfigs in this PR)
Linux SL 6.6, Intel(R) Xeon(R) CPU E5-2670 0 @ 2.60GHz, Python 2.6.6
See https://gist.github.com/c0c1ed7c2bdab6086123 for a full test report.

[boegel]

@sylmarien: did the auto-download of sources work for you?

[sylmarien]

I checked the links and they work fine, but indeed the automatic downloads didn't work well. (But I thought it was coming from our network, for various reasons)

[boegel]

In previous QuantumESPRESSO easyconfigs, must of these could be replaced by http://files.qe-forge.org/index.php?file=; trying a download from there that doesn't work properly fails (as opposed to the other links, which spit out some HTML page instead).

[boegel]

@sylmarien: EB tries each of the source URLs iteratively, and downloading atomic fails with the first link, but does produce an HTML page, so EB doesn't see it as a failure...

[sylmarien]

Ok, I see why it doesn't work then.
But the previous solution didn't work here, I didn't find a base url that could work for everything.

[sylmarien]

The only solution I see at the moment would be to have

sources = [
    '753/espresso-%(version)s.tar.gz',
    '752/atomic-%(version)s.tar.gz',
    '760/neb-%(version)s.tar.gz',
    '755/PHonon-%(version)s.tar.gz',
    '756/pwcond-%(version)s.tar.gz',
    '758/tddfpt-%(version)s.tar.gz',
    '757/xspectra-%(version)s.tar.gz',
    '754/GWW-%(version)s.tar.gz',
]

source_urls = [
    'http://www.qe-forge.org/gf/download/frsrelease/185/'
]

But I'm not even sure that would work (the source name is used to untar later, isn't it? So names wouldn't work in this case, right?) and from what you told me earlier in another pull request, that wouldn't fit your guidelines.

Do you see any other solution ?

[boegel]

@sylmarien: it seems like something has changed, since I'm now seeing similar issues with the existing QuantumESPRESSO easyconfig files... I'll see if I can figure out a fix for this.

[boegel]

@sylmarien: a workaround for this issue is available at easybuilders/easybuild-easyblocks#618

I'm not sure how happy I am with this, but it seems like the problem is rather specific to QuantumESPRESSO, or rather, qe-forge.org where the sources are downloaded from.

[boegel]

A significantly shorter list of source_urls is sufficient btw (yet still requires the workaround in the QuantumESPRESSO easyblock), see ULHPC#1

[hpcugentbot]

Refer to this link for build results (access rights to CI server needed):
https://jenkins1.ugent.be/job/easybuild-easyconfigs-pr-builder/3473/
Easyconfigs unit test suite PASSed (see https://jenkins1.ugent.be/job/easybuild-easyconfigs-pr-builder/3473/console for more details).

This pull request is now ready for review/testing.

Please try and find someone who can tackle this; contact @boegel if you're not sure what to do.

[boegel]

(close/open to trigger Travis tests)

[boegel]

Test report by @boegel
FAILED
Build succeeded for 0 out of 1 (1 easyconfigs in this PR)
node2012.delcatty.os - Linux SL 6.7, Intel(R) Xeon(R) CPU E5-2670 0 @ 2.60GHz, Python 2.6.6
See https://gist.github.com/fe121681c1db854f79d71ce270e66a68 for a full test report.

[boegel]

merged via #3444, thanks @sylmarien!