1 A 5 Minute Guide to Orcehstra
2 -----------------------------
7 Orchestra is a series of tools for Getting Shit Run.
9 It consists of a Conductor, which is the coordinating process, and
10 Players, which are the actual daemons running on nodes to do the work.
12 To prevent arbitrary execution of code, Players can only execute
13 predefined scores which have to be installed on them seperately. You
14 can use puppet, cfengine or other configuration management system to
17 Canonically, entities requesting work to be done are known as the
20 Please read the Orchestra paper (in doc/) for more information.
25 Copyright (c) 2011, Anchor Systems Pty Ltd
28 Redistribution and use in source and binary forms, with or without
29 modification, are permitted provided that the following conditions are met:
30 * Redistributions of source code must retain the above copyright
31 notice, this list of conditions and the following disclaimer.
32 * Redistributions in binary form must reproduce the above copyright
33 notice, this list of conditions and the following disclaimer in the
34 documentation and/or other materials provided with the distribution.
35 * Neither the name of Anchor Systems Pty Ltd nor the
36 names of its contributors may be used to endorse or promote products
37 derived from this software without specific prior written permission.
39 THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
40 ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
41 WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
42 DISCLAIMED. IN NO EVENT SHALL <COPYRIGHT HOLDER> BE LIABLE FOR ANY
43 DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
44 (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
45 LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
46 ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
47 (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
48 SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
54 Install the Go Release 59 compiler. Prior to compilation, apply the
55 patches in go-patches/ in order to fix a few package bugs.
57 From the top level directory, run make. It will use goinstall to
58 build the binaries in bin/
63 src/ -- All the go sources for the conductor, player, and the
64 submitjob and getstatus sample implementations.
66 doc/ -- Documentation about Orchestra and it's implementation.
68 samples/ -- Sample configuration files.
70 clientlibs/ -- Sample client libraries for communicating with the
71 Conductor as the Audience.
73 go-patches/ -- Patches against the Release version of Go to fix
74 critical bugs or missing features encountered when
75 implementing Orchestra
80 * json-unmarshal-immediate.diff : Fix for Go Bug #2170
81 * syslog-auto-reconnect.diff : Fix for Go Bug #2264
87 * BUGFIX: Fixed conductor ignoring the last_id checkpoint file after
89 * FEATUREFIX: Fix the exported fieldnames in the audience interface
90 so they no longer contain capitals. Refactor slightly to reuse
91 state types defined for persistence.
92 * Separation of some of the more esoteric shared code from the common
94 * Conductor Queue Persistence.
95 * Patches against the Go standard packages. :(
98 * First public release.
103 * There is no clean up of job data, or persistance of results at this
104 time. This is intended to be implemented as soon as time permits.
106 * getstatus gets back a lot more information than it displays.
108 * No efficient 'wait for job' interface yet. You have to poll the
109 audience interface for completion results for now. (The polls are,
110 however, stupidly cheap)
112 * Disconnect/Reconnect behaviour for players is not particualrly well
113 tested. Annecdotal evidence suggests that this is relatively
116 * Jobs will be left dangling if a player is removed from the
117 conductor's configuration and the conductor HUP'd whilst there is
118 still work pending for that player.
120 * Some of the more advanced score scheduling ideas that we've had
121 remain unimplemented, resulting in Orchestra looking a lot blander
122 than it really is meant to be.
124 * There is no support for CRLs yet.