Here's simplest example I have so far: https://travis-ci.org/greghendershott/rackjure/builds/14292882
I've been testing this package against 5.3.2, 5.3.5, 5.3.6. Building fine against all. Last night I thought, oh, I should add HEAD to the matrix. It fails. In this case, as a result of just a `raco make main.rkt`. But with similar error: ~~~ $ /usr/racket/bin/raco make main.rkt open-output-file: cannot open output file path: /usr/racket/share/pkgs/rackunit-lib/rackunit/private/compiled/location_rkt.zo system error: Permission denied; errno=13 context...: /usr/racket/collects/compiler/cm.rkt:221:4 /usr/racket/collects/compiler/cm.rkt:510:0: maybe-compile-zo /usr/racket/collects/compiler/cm.rkt:622:2: do-check /usr/racket/collects/compiler/cm.rkt:661:15 /usr/racket/collects/compiler/../racket/private/map.rkt:113:23: loop /usr/racket/collects/compiler/cm.rkt:622:2: do-check /usr/racket/collects/compiler/cm.rkt:661:15 /usr/racket/collects/compiler/../racket/private/map.rkt:113:23: loop /usr/racket/collects/compiler/cm.rkt:622:2: do-check /usr/racket/collects/compiler/cm.rkt:661:15 /usr/racket/collects/compiler/../racket/private/map.rkt:113:23: loop /usr/racket/collects/compiler/cm.rkt:622:2: do-check /usr/racket/collects/compiler/cm.rkt:661:15 /usr/racket/collects/compiler/../racket/private/map.rkt:113:23: loop /usr/racket/collects/compiler/cm.rkt:622:2: do-check /usr/racket/collects/compiler/cm.rkt:732:4: compilation-manager-load-handler... The command "/usr/racket/bin/raco make main.rkt" exited with 1. $ /usr/racket/bin/raco test -x . raco test: (submod "./conditionals.rkt" test) raco test: (submod "./dict.rkt" test) raco test: (submod "./str.rkt" test) raco test: (submod "./test.rkt" test) raco test: (submod "./threading.rkt" test) 34 tests passed The command "/usr/racket/bin/raco test -x ." exited with 0. Done. Your build exited with 1. ~~~ On Thu, Nov 21, 2013 at 10:18 AM, Greg Hendershott <greghendersh...@gmail.com> wrote: > p.s. In the gist I linked to, I had deleted some stuff from the > beginning and end, because I thought it was N/A and the thing is > already very long. > > In case I was wrong, the full raw log is here: > > https://s3.amazonaws.com/archive.travis-ci.org/jobs/14278903/log.txt > > > On Thu, Nov 21, 2013 at 10:13 AM, Greg Hendershott > <greghendersh...@gmail.com> wrote: >>> Do you have any ideas on why file timestamps might be sync in the >>> installation? >> >> The transcript I linked to is from a Travis CI session. It's a fairly >> clean slate. >> >> - VM with Ubuntu image and gcc tools. >> - Travis git clones the repo (being tested) source >> - Racke nightly build is downloaded, Racket install sh script is run. >> - raco pkg install --link <x> >> >> Hmm. As I type this, occurs to me: What if the nightly build file >> timestamps are Utah time, but the Travis server is running in a >> different time zone? Could that explain it? _________________________ Racket Developers list: http://lists.racket-lang.org/dev
