On 2014-12-12, Dima Pasechnik <dimp...@gmail.com> wrote: > On 2014-12-12, Dima Pasechnik <dimp...@gmail.com> wrote: >> On 2014-12-12, kcrisman <kcris...@gmail.com> wrote: >>> ------=_Part_154_1466710874.1418353858850 >>> Content-Type: multipart/alternative; >>> boundary="----=_Part_155_44875405.1418353858850" >>> >>> ------=_Part_155_44875405.1418353858850 >>> Content-Type: text/plain; charset=UTF-8 >>> >>> >>> >>>> Sage (6.4.1) notebook build failed on 32 bit CentOS 5.11. Version of >>>> sagenb was 0.11.1. To reproduce, build Sage from source (e.g., ./make from >>>> Sage installation directory). CPU was "CPU1: Intel(R) Core(TM)2 Duo CPU >>>> >>>> E4400 @ 2.00GHz stepping 0d" per dmesg. Sage runs, but ptest or building >>>> pdf docs replays the error messages, as shown below. The tar file for >>>> sagenb could be unpacked manually OK (in off-tree location). Checksums on >>>> Sage source download tarball matched up OK. What is the best course of >>>> action to get this to build properly? Is there a reason why it failed? >>>> I previously posted a similar item, but it did not appear in the >>>> sage-support list of items, so I am trying it again. >>>> >>>> >>> I wonder if it's because of the bsd tar I used to tar it (I built that >> don't you have 'gtar' on your machine too? >> >>> package, and have a Mac, and this has been known to cause occasional >>> warning messages, though I don't know about build errors). See >>> e.g. >>> http://superuser.com/questions/318809/linux-os-x-tar-incompatibility-tarballs-created-on-os-x-give-errors-when-unt >> >> surely enough, I see the same warnings on a 64-bit Linux: >> $ tar tf upstream/sagenb-0.11.1.tar | less >> tar: Ignoring unknown extended header keyword `LIBARCHIVE.creationtime' >> tar: Ignoring unknown extended header keyword `SCHILY.dev' >> tar: Ignoring unknown extended header keyword `SCHILY.ino' >> .... >> >> Probably my tar is newer and thus the install can live with this. >> IMHO this tarball should be replaced by something glatt kosher... >> >> Please open a ticket and cc me on it. > > I also noticed that the instructions to build the package in SPKG.txt > are outdated. Perhaps this ticket should fix these too... including a reference to this https://github.com/sagemath/sagenb/blob/master/ReleaseInstr.md will do.
> > >> >>> >>> What you could do is to retar the directory yourself, and replace the old >>> tar file with that one; then things should work fine. >> no, this is not enough, as this will surely change the tarball >> checksum; you need to follow the instructions >> in sage developer guide to recreate them. >> >> http://sagemath.org/doc/developer/packaging.html#checksums >> >> Best, >> Dima >> >>> The old release >>> scripts repackaged all our spkg files anyway, but I think the new layout >>> release style does not do that, as it takes tar balls directly from >>> upstream. In this case, an upstream who uses Mac :( >>> >> > -- You received this message because you are subscribed to the Google Groups "sage-support" group. To unsubscribe from this group and stop receiving emails from it, send an email to sage-support+unsubscr...@googlegroups.com. To post to this group, send email to sage-support@googlegroups.com. Visit this group at http://groups.google.com/group/sage-support. For more options, visit https://groups.google.com/d/optout.
