summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
* SYSV: change instances of simplename to simplefileMichael Fenn2014-11-062-8/+7
| | | | | | | | | | | This is to better match the schema since simplefile already exists. The previous simplename attribute would fail validation. Since pkgmgr already helpfully constructs url for you if simplefile exists, the tool no longer needs to do the concatenation itself. Given the low usage rate of SYSV.py and that the original functionality was introduced in a late 1.3 release, changing the name w/o providing backwards compatiblity seems reasonable.
* doc: Fix typosSol Jerome2014-11-031-8/+8
| | | | Signed-off-by: Sol Jerome <sol.jerome@gmail.com>
* doc: Add note about Decisions fix for 1.3.6Sol Jerome2014-10-311-0/+1
| | | | Signed-off-by: Sol Jerome <sol.jerome@gmail.com>
* testsuite: Fix tools __init__.py unit testAlexander Sulfrian2014-10-301-0/+2
| | | | Signed-off-by: Sol Jerome <sol.jerome@gmail.com>
* Tools: Fix install of non-whitelisted servicesSol Jerome2014-10-292-21/+23
| | | | | | | Previously, a service restart would occur even when running in whitelist mode without the service being present in the whitelist. Signed-off-by: Sol Jerome <sol.jerome@gmail.com>
* doc: Add 1.3.6 release notesSol Jerome2014-10-281-0/+26
| | | | Signed-off-by: Sol Jerome <sol.jerome@gmail.com>
* Merge pull request #223 from AlexanderS/fix-keyboard-interruptChris St. Pierre2014-10-272-1/+3
|\ | | | | fix keyboard interrupt during intial event handling
| * fix keyboard interrupt during intial event handlingAlexander Sulfrian2014-10-262-1/+3
|/ | | | | | | This removes some wildcard except handler because this drops some KeyboardInterrupt exceptions (for example previously a KeyboardInterrupt during the loading of the cache for Packages resulted in a fallback to file read).
* Merge pull request #222 from AlexanderS/fix-keyboard-interruptChris St. Pierre2014-10-241-5/+5
|\ | | | | Server/Core: also shutdown on exceptions during fam blocking
| * Server/Core: also shutdown on exceptions during fam blockingAlexander Sulfrian2014-10-251-5/+5
|/ | | | | | | Previously the server got stuck, if a keyboard interrupt occured during block_for_fam_events. The KeyboardInterrupt exception was only handled in the executable and it does not call shutdown for the Core. So the running fam thread does not get killed and the main thread waits for it.
* Merge branch 'cleanup_pidfile' of https://github.com/mattikus/bcfg2 into maintSol Jerome2014-10-232-23/+41
|\
| * Remove PIDFileError as it does not always exist in the package and is rarely ↵Matt Kemp2014-10-021-5/+0
| | | | | | | | used.
| * Catch possible typeerror resulting from None being returned when reading the ↵Matt Kemp2014-10-022-2/+2
| | | | | | | | pid.
| * Fixes to ensure pidfile can be opened or broken if stale.Matt Kemp2014-10-012-19/+15
| |
| * pylint fixes.Matt Kemp2014-10-011-3/+4
| |
| * Attempt to break the pid lock during startup.Matt Kemp2014-10-012-15/+41
| | | | | | | | | | | | | | | | This commit attempts to break the pidfilelock during startup in cases where the process may have exited without successfully cleaning up the lockfile. It also attempts to grab the lock before opening the context. Also applied to the Collector module, which may have been looking for the wrong exception since it does not rely on a timeout.
* | Merge branch 'reporting-fix-filter' of https://github.com/AlexanderS/bcfg2 ↵Sol Jerome2014-10-232-39/+50
|\ \ | | | | | | | | | into maint
| * | Reporting: better exception handlingAlexander Sulfrian2014-10-201-33/+44
| | | | | | | | | | | | Try to keep the try-except-blocks as small as possible.
| * | Reporting: fix filter urlsAlexander Sulfrian2014-10-121-6/+6
| | | | | | | | | | | | | | | | | | The regex match for the filter urls were to strict. They disallowed some charaters, that are valid in group names and so the django reverse mechanism for building urls failed.
* | | Merge branch 'add-name-for-sources' of https://github.com/AlexanderS/bcfg2 ↵Sol Jerome2014-10-212-2/+19
|\ \ \ | | | | | | | | | | | | into maint
| * | | Packages: add name to additional_data for SourcesAlexander Sulfrian2014-10-181-0/+1
| | | |
| * | | Packages: add name to sourcesAlexander Sulfrian2014-10-182-2/+18
| |/ /
* | | Merge branch 'ssl-protocol-fix' of github.com:solj/bcfg2 into maintSol Jerome2014-10-213-5/+10
|\ \ \
| * | | Proxy.py: Pass through SSL protocol optionSol Jerome2014-10-153-5/+10
| |/ / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Previously we were not passing through the SSL protocol specified in the client's bcfg2.conf which caused it to unconditionally be set to xmlrpc/ssl. While this appears to automagically work with newer versions of openssl, the version in e.g. centos5 will fail if the server is set to use TLSv1. This commit passes through the setting from the client's bcfg2.conf so that older clients can talk to servers which are set to TLSv1 (in order to mitigate the effects of POODLE). Signed-off-by: Sol Jerome <sol.jerome@gmail.com>
* | | Merge pull request #215 from jsbillings/yum-pluginChris St. Pierre2014-10-202-0/+22
|\ \ \ | |/ / |/| | YUM: Add options to enable and disable Yum plugins
| * | YUM: Add options to enable and disable Yum pluginsJonathan Billings2014-10-172-0/+22
|/ / | | | | | | | | | | | | | | | | | | | | Adds two options you can define: * disabled_plugins: A comma-separated list of plugins to disable * enabled_plugins: A comma-separated list of plugins to enable This allows you to run bcfg2 with certain plugins enabled or disabled when they're not set that way in the yum configuration. This is useful because the Bcfg2 YUM plugin is initialized before it can read in any files that might overwrite yum plugin configuration.
* | doc: Clarify debian bcfg2 versionsSol Jerome2014-10-101-3/+3
| | | | | | | | Signed-off-by: Sol Jerome <sol.jerome@gmail.com>
* | strike possibly incorrect docs about genshi/cheetah priorityChris St. Pierre2014-09-111-9/+0
| |
* | Merge branch 'bcfg2-1.3.5-python2.4-compat' of ↵Sol Jerome2014-09-103-6/+7
|\ \ | | | | | | | | | https://github.com/fennm/bcfg2 into maint
| * | use Bcfg2.Compat to get any()Michael Fenn2014-09-101-1/+1
| | |
| * | Use older nested try syntax for finally for python 2.4 compatMichael Fenn2014-09-101-4/+5
| | |
| * | Use the older Thread.getName() interface for python 2.4 compatMichael Fenn2014-09-101-1/+1
| | |
* | | doc: Include the 1.3.5 release notesSol Jerome2014-09-101-0/+1
|/ / | | | | | | Signed-off-by: Sol Jerome <sol.jerome@gmail.com>
* | Version bump to 1.3.5v1.3.5Sol Jerome2014-09-0516-17/+23
| | | | | | | | Signed-off-by: Sol Jerome <sol.jerome@gmail.com>
* | doc: Add 1.3.5 release notesSol Jerome2014-09-041-0/+33
| | | | | | | | Signed-off-by: Sol Jerome <sol.jerome@gmail.com>
* | Merge branch 'sysv-http' of https://github.com/nathanolla/bcfg2 into maintSol Jerome2014-08-264-4/+69
|\ \
| * | Log when downloading packages via HTTPNathan Olla2014-08-191-0/+2
| | |
| * | Remove group tag from examplesNathan Olla2014-07-161-4/+0
| | |
| * | Check for origpkgtool attribute to prevent things that subclass SYSV from ↵Nathan Olla2014-07-161-6/+8
| | | | | | | | | | | | breaking
| * | Add some basic documentation for SYSV packages over HTTPNathan Olla2014-07-161-1/+26
| | |
| * | Remove unused import of copyNathan Olla2014-07-161-1/+0
| | |
| * | Implement _get_package_command and append _sysv_pkg_path attributeNathan Olla2014-07-161-33/+27
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Instead of doing a partially complete Install() method for SYSV, implements a custom _get_package_command that will use the _sysv_pkg_path attribute added by the pkgmogrify call. This will allow the installs to complete. Unfortunately, the single-pass install will still fail if there are any packages with an http:// URL. The pkgadd invocation for 'device' sources doesn't take multiple packages and the 'datastream' invocation doesn't handle packages with an HTTP URL. Finally, there is no reliable standard naming convention for SYSV datastream files, so the simplename attribute is re-used. There is a known issue with this patch - if any packages specified in the PackageList have an http url, the single-pass install will produce an error like: Trying single pass package install for pkgtype sysv pkgadd: ERROR: Failure occurred with http(s) negotiation: <'Peername' doesn't match 'host' or no matching entry> pkgadd: ERROR: unable to download package datastream from <http://install1.d.stor.en.desres.deshaw.com/jumpstart10U10/packages>. Single Pass Failed because the command that results isn't valid syntax for pkgadd. A workaround would be to add code to skip the single-pass install if any packages had the simplename attribute, or by checking the url for the presence of 'http'. I'm not sure if that should be fixed or if this is reasonable in this case.
| * | Fix indentingNathan Olla2014-07-141-5/+4
| | |
| * | Fix indent.Nathan Olla2014-07-141-1/+1
| | |
| * | Add urlretrieve to Compat and documentNathan Olla2014-07-143-3/+7
| | |
| * | SYSV: Implement downloading and installing SYSV packages from HTTPNathan Olla2014-07-141-2/+46
| | | | | | | | | | | | | | | | | | | | | | | | pkgadd has different syntax for different sources (datastream and file system format) which makes using a single pkgtool variable difficult. Also, SYSV packages in datastream format don't necessarily have uniform names. Therefore, use the existing 'simplename' attribute to specify the datastream file name.
* | | debian: Clean up PID fileSol Jerome2014-08-151-11/+12
| | | | | | | | | | | | Signed-off-by: Sol Jerome <sol.jerome@gmail.com>
* | | Reporting: Remove duplicate methodSol Jerome2014-08-141-3/+0
| | | | | | | | | | | | Signed-off-by: Sol Jerome <sol.jerome@gmail.com>
* | | doc: Fix typoSol Jerome2014-07-281-1/+1
| | | | | | | | | | | | Signed-off-by: Sol Jerome <sol.jerome@gmail.com>
* | | Packages/Apt: Essential could be "no"Alexander Sulfrian2014-07-281-1/+2
| | | | | | | | | | | | | | | | | | The "Essential" field in the package control fields could be "yes" or "no". Only yes sould define the package as essential. The value "no" sould be handled same as not having the field at all.