summaryrefslogtreecommitdiffstats
AgeCommit message (Collapse)AuthorFilesLines
2014-09-04Automatic commit of release 0.18.0beaker-0.18.0Dan Callaghan7-8/+8
2014-09-01older curl doesn't have --retryDan Callaghan13-77/+252
Change-Id: I86fdffe2dd663d4ff3f180add9c9b535f75d6c97
2014-09-01busybox writes its --help to stderr, not stdoutDan Callaghan13-25/+25
Fixes the busybox detection in the fetch_wrapper snippet. Change-Id: Ib1da256020411e3d36c6b344e66c84f3c6d9002d
2014-08-29docs: glossary terms should be lower case except for proper nounsDan Callaghan1-15/+15
Change-Id: I62b5a6937859af79871b9b0cb8ee6075ebee1711
2014-08-29docs: s/e-mail/email/Dan Callaghan8-9/+9
I have always preferred the spelling "e-mail" but apparently I am living in the past... IBM says, use "email". Change-Id: I9b6d283eefcffe63a8861478b6fe5f1e15d387c2
2014-08-29remove release note from next/Dan Callaghan1-7/+0
This is already in the 0.18 release notes but I missed deleting the file from next/ because it was merged after I started drafting the notes. Change-Id: I9e6898161f75524da78884cd872ff079a9e6bfe0
2014-08-28Automatic commit of release 0.18.0rc1beaker-0.18.0rc1Dan Callaghan7-9/+9
2014-08-28Merge branch 'release-0.17' into developDan Callaghan8-16/+64
2014-08-28release notes for 0.18Dan Callaghan7-44/+196
Change-Id: Ia3d93cf957dc1d62b4ad709d47de5abc7e2b14fe
2014-08-27Record access policy rules removal upon group deletionrelease-0.17Amit Saha4-5/+44
Bug: 1132730 Change-Id: Ia185b8773fe16c7cab0129453f8f59ff95f0a05c
2014-08-27Rename "Remove" button to "Delete Group" button for deleting groupsAmit Saha5-11/+18
Bug: 1132729 Change-Id: I9cd561f0eb2345973f59bbdc9189e8a62905e1c6
2014-08-27re-use existing .requests_session() methodDan Callaghan1-3/+2
This way we avoid importing requests in the client command itself, which lets us avoid having a BuildRequires on python-requests for the client code (which we currently do not have). Change-Id: I6edcf0bf28bb42bd23bc8360feb8a8b425e1a108
2014-08-27tests: create task library snapshots in basepath.reposDan Callaghan1-1/+3
We can't unconditionally use /tmp for task library snapshots in the tests, because basepath.rpmspath might be on /var which might be a separate filesystem. Specifically, this fixes the following failure when tmpfs is mounted in /tmp: Traceback (most recent call last): File "/usr/lib/python2.6/site-packages/bkr/server/tests/test_model.py", line 193, in test_make_snapshot_repo self.tasklibrary.make_snapshot_repo(recipe_repo_parent) File "/usr/lib/python2.6/site-packages/bkr/server/model/tasklibrary.py", line 175, in make_snapshot_repo self._link_rpms(repo_dir) File "/usr/lib/python2.6/site-packages/bkr/server/model/tasklibrary.py", line 158, in _link_rpms os.link(srcpath, dstpath) OSError: [Errno 18] Invalid cross-device link Change-Id: I2db2d05ca902972de6e9a5f72d1435674a74aa9a
2014-08-26forgot to add beaker-repo-update.8 to the specDan Callaghan1-0/+2
2014-08-26beaker-repo-update: skip symlinked harness reposDan Callaghan1-0/+2
Bug: 1132763 Change-Id: Ia2d6d0c4721a523b6fc22f0f393006f26cc79e63
2014-08-26docs: describe how to handle custom distrosDan Callaghan2-5/+48
Bug: 1132764 Change-Id: I37285ba58d7eb1790066042ac627d6ccea4f572f
2014-08-26docs: man page for beaker-repo-updateDan Callaghan4-5/+76
Change-Id: I1834a7a6179a7e5da1b1d8a1719f5614bd9fbf5a
2014-08-26docs: add root notice to all server man pagesDan Callaghan3-3/+10
Change-Id: I8198fb3ecfcac0540b883944e4696f519517f988
2014-08-26move server man pages to section 8Dan Callaghan3-10/+11
Change-Id: I8f8241241daa9cd4cca002dba4ead0f1f61a21a7
2014-08-22tests: use addCleanup for WebDriver and MailCaptureDan Callaghan55-262/+41
This saves a bit of code in tests, and eliminates the potential for missing tearDown methods which can leak browser processes (I spotted a few when doing this patch). Change-Id: Ic6cb9038e753725e45ba887640376b0857a59259
2014-08-22tests: raise an error for leaked browser processesDan Callaghan1-0/+9
Change-Id: I06a75fee2f8208cff6bd68c2c9498fedcffb7087
2014-08-21drop 'systemd' ksmeta variable entirelyDan Callaghan6-9/+22
In commit 160e4e6b I was intending to keep both 'has_systemd' and 'systemd' for compatibility with Beaker installations where systemd=True is set for custom distros, or where the systemd variable is used in custom snippets. However there is no sane way to obey the 'systemd' variable if it is set while also correctly populating both 'has_systemd' and 'systemd'. It was only possible to set 'systemd' since Beaker 0.17.2, it was never documented, and the main use case (setting systemd=True for custom distros with systemd) is no longer needed because custom distros are assumed to have systemd by default. So it should be safe to just drop the 'systemd' variable entirely. Change-Id: Ifd5e7b1dabe5f48f97a54519a3529e9f665e4b90
2014-08-20workflow-xslt: Fix that incorrect beaker job was generated when using ↵David Sommerseth1-2/+3
--save-internal-xml When using --save-internal-xml, the internal XML document needed for the XSLT processing was deleted, which meant no command line arguments with variables to the XSLT template would be considered. This patch ensures that the internal XML document generated will be saved for future use if --save-internal-xml is used. It can be argued that the internal XML document could be dumped after the XSLT processing, which was the case before commit dd874428d6e90141. The problem with that is that it makes debugging XSLT files much harder, if there are an issue which makes the XSLT processor choke. Having access to the internal document in such moments is valueable, as you can do the deeper XSLT debugging using xsltproc. Change-Id: If15c0214334fe728444e3038e389c85f742b9f47 Signed-off-by: David Sommerseth <davids@redhat.com>
2014-08-19Merge branch 'release-0.17' into developDan Callaghan21-41/+106
2014-08-19remove the newlines when triple-clicking and copying H1 element on the ↵mjia1-3/+1
system page Bug: 1121763 Change-Id: I1cedb5fa0127168e737cc60dc3d3ea162f59ec49
2014-08-19bkr job-logs --size, to print log file sizesDan Callaghan4-38/+97
Bug: 1128048 Change-Id: I1e2b9547b8fff6b0095fdc0a9edf1b412085ec84
2014-08-19merged RHEL6 kickstart into default kickstartDan Callaghan14-165/+105
Bug: 1070597 Change-Id: I197a182d042e694c4a691f0f1a42eca33036550b
2014-08-19merged Fedora and RHEL7 kickstarts into a 'default' kickstartDan Callaghan12-140/+57
Also removed Fedora 17 armhfp hacks which are no longer relevant. Bug: 1070597 Change-Id: Ie3e9eeb62beec9a10fee4e52a42fc4b4013f565c
2014-08-18replaced remaining distro conditionals with feature test variablesDan Callaghan8-19/+33
The variables are named 'has_*' and are expected to be defined when the feature is present, or undefined otherwise. An already-defined variable can be undefined by the usual '!has_autopart_type' syntax. Also renamed the existing 'systemd' feature test variable to 'has_systemd' for consistency with the others. Change-Id: I4129a6fb03a27a38f21de032fe0485c8a3dd9e34
2014-08-18move 'yum' ksmeta to OSMajor.default_install_optionsDan Callaghan15-21/+10
Bug: 1070597 Change-Id: If4cdefc115fd4bef4184e1506482e5d6826b5af5
2014-08-18new OSMajor.default_install_options method for detected ks meta variablesDan Callaghan5-28/+56
The first such variables we already have are 'end' and 'systemd'. This moves the definition of 'end' and 'systemd' to the new method, which now comes at the very start of the install options inheritance chain. We also flip the conditional so that by default we assume the new behaviour, the old behaviour is hardcoded for old distros. That means new custom distros will work without needing to be hardcoded. The variables are also now documented, so that admins can override them if necessary. Change-Id: I41fffc12fc97770cedab96cd99ced9b564d346da
2014-08-18conditionalise snippets based on whether the relevant config file existsDan Callaghan17-36/+114
The readahead_sysconfig and virt_console_post snippets can use a runtime condition on the existence of their respective config files, rather than checking against the distro. Bug: 1070597 Change-Id: I6477ce73703c0942db90c4f9437e6a486fc1c80d
2014-08-18replace distro conditionals for curl/wget with a wrapper functionDan Callaghan22-137/+714
The %pre and %post scriptlets will use a 'fetch' wrapper function, which is conditionally defined at runtime to use either curl or wget depending on which one is found on the system. Bug: 1070597 Change-Id: Idc3af99bc3f8609eb0173c2533dd0cbba6fa6c05
2014-08-18enable chrony services in %post, instead of using 'services' commandDan Callaghan16-10/+19
We're already disabling/enabling clock sync services in %post at the moment, so this just centralises it all to there. This also removes some variation between templates, and additionally allows users/admins to add their own 'services' command to their kickstart without interfering. Bug: 1070597 Change-Id: I6b49ef3eeab2347077c9ebbbb26539c43b0f2768
2014-08-18don't hide output from chkconfigDan Callaghan9-36/+36
Change-Id: Ie5735be27e6caddfb37033131980b504b713b140
2014-08-18tests: negative test for bootloader --leavebootorder optionDan Callaghan1-0/+32
Change-Id: I1b661ad7b739f663f55f9c1c2e0b63d96f0c2c84
2014-08-18config option to disable bumping of priority for single-system recipesDan Callaghan4-3/+25
Bug: 978904 Change-Id: I4eca275fd1afa4aed4330fb477e72aa9530b4e57
2014-08-18fix not-enough-systems logic to correctly abort recipe setsDan Callaghan2-5/+51
Under certain very specific circumstances the not-enough-systems logic would try to move a recipe to Queued after already aborting the recipe set. Bug: 1120052 Change-Id: Id64c8a90454013680185e8f214744aa30147d9d9
2014-08-18handle EFI GRUB config paths when setting grubportDan Callaghan17-33/+30
Also split the grubport handling into its own snippet and documented it. Bug: 1108498 Change-Id: Ic9f365fa991594f56ae5c11babb472ceff8c2fe3
2014-08-14Merge branch 'release-0.17' into developAmit Saha14-29/+164
2014-08-12small tweaks to option descriptions in beaker-usage-reminder(1)Dan Callaghan1-8/+11
Change-Id: Idbefbaed23f398f6f31067f1202686f41bb0c682
2014-08-06Merge branch 'release-0.17' into developDan Callaghan8-13/+63
2014-07-30fix a unittest error when upgrading SQLAlchemy to 0.8.6Matt Jia1-1/+0
In SQLAlchemy 0.8, removing a record from the collection will have the effect of the row being deleted on flush. Change-Id: Id19a1059021e904d487f4d61263a30fa342af013
2014-07-30Merge release-0.17 into developMatt Jia19-52/+179
2014-07-29Reporting query for system utilizationAmit Saha2-0/+42
Bug: 1117681 Change-Id: Ie86e2ad05d21b21892ca4751d75cb916f5ade3eb
2014-07-29fix a unittest typo errorMatt Jia1-1/+1
Change-Id: I1a5879d6b663dfacd5e3efbcb0c9542d30a5eda6
2014-07-23Fix nova API call get_console_outputMatt Jia3-4/+4
Bug: 950903 Change-Id: I73b8214ba0a81b328d1e6a749d26df9876fc3b96
2014-07-18Merge branch 'release-0.17' into developDan Callaghan7-8/+8
2014-07-17Use load_config_or_exit() instead of load_config()Amit Saha1-2/+2
The other server tools were modified as part of fix to BZ#1111491
2014-07-17Merge branch 'release-0.17' into developAmit Saha11-18/+101
Conflicts: Server/bkr/server/tools/nag_email.py