hydra-users
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: low-hassle build-output excerpt retrieval


From: Rob Vermaas
Subject: Re: low-hassle build-output excerpt retrieval
Date: Sun, 30 Nov 2014 16:35:34 +0100

Hi,

it seems the feature to directly access the failed build's output is broken at the moment. I will look into this. For now this is the contents of the file you mentioned.

# cat /nix/store/f6ql2q10pdxyac9sk44s5fq6y5fg3fjc-rcs-5.9.3-4-g65a50ae/nix-build-rcs-5.9.3-4-g65a50ae.drv-0/rcs-UNKNOWN/tests/t999.d/out
3,5c3,5
< $Header: /private/tmp/nix-build-rcs-5.9.3-4-g65a50ae.drv-0/rcs-UNKNOWN/tests/t999.d/RCS/a.c 2.1 2014/11/29 11:11:05 w seriously-spurious _appleevents $
< $Id: a.c 2.1 2014/11/29 11:11:05 w seriously-spurious _appleevents $
< $Locker: _appleevents $
---
> $Header: /private/tmp/nix-build-rcs-5.9.3-4-g65a50ae.drv-0/rcs-UNKNOWN/tests/t999.d/RCS/a.c 2.1 2014/11/29 11:11:05 w seriously-spurious nixbld1 $
> $Id: a.c 2.1 2014/11/29 11:11:05 w seriously-spurious nixbld1 $
> $Locker: nixbld1 $
#rcsdiff -kkvl failed

Cheers,
Rob

On Sat, Nov 29, 2014 at 5:55 PM, Thien-Thi Nguyen <address@hidden> wrote:
GNU RCS currently fails for ‘x86_64-darwin’:

 http://hydra.nixos.org/build/17633689

with one failure (t999, the last test in the test suite):

 http://hydra.nixos.org/build/17633689/nixlog/2/raw

The "failed build produced output" page:

 http://hydra.nixos.org/build/17633689/contents/1

shows these files, among the others (w/ common prefix
"./nix-build-rcs-5.9.3-4-g65a50ae.drv-0/rcs-UNKNOWN/tests/"
and some constant-content fields from ls(1) snipped):

 dr-xr-xr-x 3   4096  t999.d
 -r--r--r-- 1    525  t999.d/a.0
 -r--r--r-- 1    525  t999.d/a.1
 -r--r--r-- 1      4  t999.d/a.11
 -r--r--r-- 1      4  t999.d/a.12
 -r--r--r-- 1      8  t999.d/a.3x1
 -r--r--r-- 1    480  t999.d/a.b
 -r--r--r-- 1    503  t999.d/a.c
 -r--r--r-- 1    478  t999.d/a.e
 -r--r--r-- 1    429  t999.d/a.i
 -r--r--r-- 1    489  t999.d/a.i1
 -r--r--r-- 1    429  t999.d/a.i2
 -r--r--r-- 1    141  t999.d/a.k
 -r--r--r-- 1    480  t999.d/a.kv
 -r--r--r-- 1    503  t999.d/a.kvl
 -r--r--r-- 1    479  t999.d/a.n
 -r--r--r-- 1    480  t999.d/a.o
 -r--r--r-- 1    159  t999.d/a.oo
 -r--r--r-- 1      4  t999.d/a.t
 -r--r--r-- 1    368  t999.d/a.v
 -r--r--r-- 1      0  t999.d/err
 -r--r--r-- 1    521  t999.d/out
 dr-xr-xr-x 2   4096  t999.d/RCS
 -r--r--r-- 1   2234  t999.d/RCS/a.c

Of these, i am most interested in in t999.d/out, which should give
some hint about where the test failed.  Is there any easy way to
retrieve that file (or better yet, t999.d/ in its entirety, as a
tarball), without having to obtain a "hydra account"?  I'm not
really opposed to getting a hydra account (just ignorant, lazy,
and pressed for time :-D), so if that's the only way forward, then
please consider this a request for one.

--
Thien-Thi Nguyen
   GPG key: 4C807502
   (if you're human and you know it)
      read my lisp: (responsep (questions 'technical)
                               (not (via 'mailing-list)))
                     => nil



--
Rob Vermaas

[email] address@hidden

reply via email to

[Prev in Thread] Current Thread [Next in Thread]