Changes between Version 53 and Version 54 of WikiStart
- Timestamp:
- 11/06/12 18:09:21 (9 years ago)
Legend:
- Unmodified
- Added
- Removed
- Modified
-
WikiStart
v53 v54 83 83 84 84 In other repos, `.jar` files have mostly been removed (especially as 85 some of them were from copyleft source), and the directories for 86 support of obsolete versions have been removed from `nbgui`. This may 87 affect building Java sources. 85 some of them were from unavailable copyleft source), and the directories for 86 support of obsolete versions have been removed from `nbgui`. 88 87 89 88 The repositories are maintained under … … 111 110 snapshots currently aren't made automatically, so it may not be so up-to-date. 112 111 113 You can access the repos with Mercurial (`hg`) at the same URLs as with `darcs`, e.g. {{{hg clone http://arc.liv.ac.uk/repos/darcs/sge/}}}. (They should probably be aliased to avoid confusion.) You can also pull with `git`, but the relevant URL is the darcs one with `/.git` appended, e.g. {{{git clone http://arc.liv.ac.uk/repos/darcs/sge/.git}}}. Servers aren't currently available for hg or git, so pulling may be relatively slow Also note that pulls with git or hg leave scripts not executable — see the amended build instructions. 112 You can access the repos with Mercurial (`hg`) at the same URLs as 113 with `darcs`, 114 e.g. {{{hg clone http://arc.liv.ac.uk/repos/darcs/sge/}}}. (They should probably be aliased to avoid confusion.) You can also pull with `git`, but the relevant URL is the darcs one with `/.git` appended, e.g. {{{git clone http://arc.liv.ac.uk/repos/darcs/sge/.git}}}. 115 Native servers aren't currently available for hg or git, so pulling 116 may be relatively slow. Also note that pulls with git or hg leave 117 scripts not executable, but the initial `aimk` fixes that. 114 118 115 119 == Building == 116 120 117 Building the source is currently rather a pain, 118 which is something to improve urgently. With appropriate development 119 packages installed — they need documenting (but see #1305) — it should build on GNU/Linux 121 Building the source is rather a pain, 122 which is something to improve at high priority, although there is a 123 spec file for RPM-based systems which may need adjusting for non-RHEL ones. With appropriate development 124 packages installed — the list in README.BUILD below probably still needs expanding (but see #1305) — it should build on GNU/Linux 120 125 in Bourne shell with something like this: 121 126 {{{ 122 127 cd source 123 csh aimk -only-depend && scripts/zerodepend && ./aimk depend &&128 csh -f aimk -only-depend && scripts/zerodepend && ./aimk depend && 124 129 ./aimk 125 130 }}} 126 (The aimk arguments are set up to default to {{{-system-libs -no-gui-inst}}}.)127 To avoid Java stuff, try:131 (The aimk arguments are set up to default to 132 {{{-system-libs -no-gui-inst}}} in `aimk.private`.) To avoid Java stuff, try: 128 133 {{{ 129 134 cd source 130 csh aimk -only-depend && scripts/zerodepend &&135 csh -f aimk -only-depend && scripts/zerodepend && 131 136 ./aimk -no-java -no-jni depend && 132 137 ./aimk -system-libs -no-gui-inst -no-java -no-jni … … 135 140 necessary executable permissions, which will be missing if you pulled 136 141 with git or hg, or didn't use `--set-scripts-executable` with darcs. 137 See ticket #821 about the `--no-gui-inst`.138 142 139 143 For further information, see … … 141 145 [source:/sge/source/README.aimk README.aimk] in the source tree, and 142 146 ignore any inappropriate references to the sunsource web site, at 143 least — those documents need review in the absence of a better build147 least — those documents need further review in the absence of a better build 144 148 system. 145 149 … … 197 201 fragmentary apart from more recent items. We have a partial 198 202 [http://arc.liv.ac.uk/pipermail/gridengine-users/ archive] of the old -users and -dev 199 lists (but mixed up, non-threaded, and currently not indexed) pending supply of 200 the complete archive from sunsource. Beware of referring to items in that archive, 201 as the URLs won't be stable until we get the complete archives. 203 lists (but mixed up, non-threaded, and currently not indexed). Beware of referring to items in that archive, 204 as the URLs won't be stable unless we get the complete archives from Oracle. 202 205 203 206 == Copyright and Naming == … … 213 216 obviously under the original (rather odd!) 214 217 [http://arc.liv.ac.uk/repos/darcs/sge/LICENCES/SISSL SISSL], 215 except for some more-or-less ‘third party’ stuff like `qmake`. 218 except for some more-or-less ‘third party’ stuff like `qmake` and 219 various library code. See the 220 [http://arc.liv.ac.uk/repos/darcs/sge/LICENCES/README licences README] 221 for a non-lawyer's summary. 216 222 217 223 The appendix to the SISSL requires a conformance test to use names … … 232 238 == Related == 233 239 234 As well as Univa's public repo, there's another project,240 As well as Univa's now-dead public repo, there's another project, 235 241 [http://sourceforge.net/projects/gridscheduler/ Open Grid Scheduler], 236 or ‘gridscheduler’, 237 which is also continuing work on gridengine but has 242 or ‘gridscheduler’, which is also continuing work on gridengine but has 238 243 different aims. Unlike this one, its repository doesn't use a distributed 239 244 revision control system, doesn't include history from 240 245 before the gridengine v6.2u5 release, existing issues, or sources from outside the gridengine repository. It would 241 be better if there was only one project , and there is now hope of242 being able to work together somewhat.246 be better if there was only one project but, unfortunately, there now 247 seems no prospect of that. 243 248 244 249 The `sge` source here is essentially a large superset of