summaryrefslogtreecommitdiffstats
AgeCommit message (Collapse)AuthorFilesLines
2015-02-14reproducible: html_notes: print the actual number of categorized packages, ↵Mattia Rizzolo1-2/+4
not all noted packages
2015-02-14reproducible: html_notes: call get_bugs() globallyMattia Rizzolo1-3/+3
2015-02-14reproducible: fix html in reproducible_html_notesMattia Rizzolo1-1/+1
2015-02-13reproducible: TODO: another usertag, 'environment'Holger Levsen1-1/+1
2015-02-11use sparse logical volumes for g-iHelmut Grohne1-11/+2
These are snapshots of the zero target and thus automatically zeroed, so we do not have to overwrite them avoiding the IO load altogether.
2015-02-11reproducible: check notes for validity. Check whether the package actually ↵Mattia Rizzolo1-0/+11
exists and was tested
2015-02-10bin/reproducible_html_notes: let more external URLs escape the iframe tarpitChris Lamb1-5/+5
2015-02-10the new tag is called signature_s_Holger Levsen1-1/+1
2015-02-09reproducible TODO: graph new usertag "signature"Holger Levsen1-0/+1
2015-02-09cleanup FIXMEsHolger Levsen1-1/+2
2015-02-09Revert "whitespace change to force touching of all job configs"Holger Levsen17-28/+28
This reverts commit 124c0ddb89d1c420804a15ee8366b89d7e1371de.
2015-02-09whitespace change to force touching of all job configsHolger Levsen17-28/+28
2015-02-09reproducible: credit Holger in *py scripts, as they are based on his work. + ↵Mattia Rizzolo4-4/+8
change license GPL-2+ → GPL-2
2015-02-09reproducible: UDD is not critical in this infrastucture, let it fail nicely ↵Mattia Rizzolo1-3/+22
as much as possible In particular: * missing network connection is fine * errors on the udd part are fine * quite everything is fine just, every query will cause a fake empty response, and this will be logged as error.
2015-02-09Reverting the last two commitsMattia Rizzolo2-9/+4
2015-02-08reproducible: and you must try and try... and really better except properlyHolger Levsen2-4/+9
2015-02-08reproducible: initialize udd db connection only when also querying uddHolger Levsen1-1/+1
2015-02-08apache: clean the config by using a macroMattia Rizzolo1-50/+20
2015-02-08apache: better commentsMattia Rizzolo1-3/+5
2015-02-07fixup 7ee6af0Holger Levsen1-2/+2
2015-02-07reproducible README: add links to wiki, .json and sqlite db file.Holger Levsen1-3/+9
2015-02-07g-i: download preseed file from 127.0.0.1 from the jenkins job (and from ↵Holger Levsen1-5/+7
10.0.2.1 from inside qemu) but show jenkins.debian.net in output, so the url from the outside world
2015-02-07g-i: do not redirect to https if accessed as 127.0.0.1 or 10.0.2.1 - ↵Holger Levsen1-0/+54
redudant but working configuration
2015-02-07bin/find_dpkg_trigger_cycles.sh: wait 10 seconds before retryjosch1-1/+1
2015-02-07bin/find_dpkg_trigger_cycles.sh: print curl exit code and exit after failurejosch1-1/+1
2015-02-06fix grammar: sb#builded#built#Holger Levsen1-1/+1
2015-02-06chroot-installation: only simulate autoremove, thanks jcristauHolger Levsen1-1/+1
2015-02-06fixupHolger Levsen1-2/+2
2015-02-06chroot-installation: give up for now, revert exception+retry handling...Holger Levsen1-16/+17
2015-02-06chroot-installation: yet more debugHolger Levsen1-0/+3
2015-02-06chroot-installation: use 'set -o pipefail' to really catch all the failures, ↵Holger Levsen1-7/+11
thanks wRAR for the hint
2015-02-05chroot-installation: add more debug outputHolger Levsen1-0/+3
2015-02-05TODO for reproducible: new job+page: "packages different from sid in our ↵Holger Levsen1-0/+2
toolchain"
2015-02-05chroot-installation: another attempt to handle failures correctly...Holger Levsen1-4/+6
2015-02-04Revert "Revert "reproducible: link to tracker.debian.org instead to the old ↵Holger Levsen1-1/+1
PTS, but keep the link title 'PTS', at least for now"" This reverts commit 1b11b4620a3a6b0caf7b6fd7a5f44101c1569f44.
2015-02-04Revert "reproducible: link to tracker.debian.org instead to the old PTS, but ↵Holger Levsen1-1/+1
keep the link title 'PTS', at least for now" This reverts commit 7217f26d12593c07cb7ec51df02f5489f5d77e1f.
2015-02-04reproducible: link to tracker.debian.org instead to the old PTS, but keep ↵Holger Levsen1-1/+1
the link title 'PTS', at least for now
2015-02-03reproducible: also export build_time in the jsonMattia Rizzolo1-3/+4
2015-02-03reproducible: no need to install our version of discount anymore, as #762622 ↵Holger Levsen1-2/+2
has been fixed in sid
2015-02-03bin/find_dpkg_trigger_cycles.sh: print error if curl failsjosch1-1/+1
2015-02-03reproducible: add 'and others' to copyright noticeHolger Levsen2-2/+2
2015-02-02reproducible: rebuild relevant package pages also when we remove a noteMattia Rizzolo1-1/+5
2015-02-02TODO++Holger Levsen1-1/+1
2015-01-31rebootstrap: remove useless combinationsHelmut Grohne1-0/+4
The combination of "supported" and "nobiarch" is doomed to always fail, as nobiarch simply is unsupported. This commit removes the following jenkins jobs: jenkins_hppa_gcc49_nobiarch_supported jenkins_i386_gcc49_nobiarch_supported jenkins_i386_gcc49_nobiarch_supported_debbindiff jenkins_mips_gcc49_nobiarch_supported jenkins_mips_gcc49_nobiarch_supported_debbindiff jenkins_mips64el_gcc49_nobiarch_supported jenkins_mipsel_gcc49_nobiarch_supported jenkins_mipsel_gcc49_nobiarch_supported_debbindiff jenkins_powerpc_gcc49_nobiarch_supported jenkins_powerpc_gcc49_nobiarch_supported_debbindiff jenkins_ppc64_gcc49_nobiarch_supported jenkins_s390x_gcc49_nobiarch_supported jenkins_s390x_gcc49_nobiarch_supported_debbindiff jenkins_sparc_gcc49_nobiarch_supported jenkins_sparc64_gcc49_nobiarch_supported jenkins_x32_gcc49_nobiarch_supported
2015-01-30reproducible: rewrite the reproducible_json.sh in python, and replace itMattia Rizzolo3-44/+34
2015-01-29chroot-installation: set -e explicitly. this doesnt make sense, but thats ↵Holger Levsen1-0/+1
the way it is atm
2015-01-29moreHolger Levsen1-0/+3
2015-01-28chroot-installation: try harder to exit correctlyHolger Levsen1-4/+4
2015-01-28rebootstrap: throttle 4 concurrent jobs max.Holger Levsen1-2/+2
2015-01-28bin/find_dpkg_trigger_cycles.sh: make sure the activated path is either ↵josch1-1/+1
followed by a slash or the line end