[342] | 1 | |
---|
[363] | 2 | LEGEND f: fixed - c: changed - a: added - r: removed |
---|
[361] | 3 | |
---|
[511] | 4 | 0.4: |
---|
| 5 | |
---|
| 6 | jobmond) |
---|
| 7 | a: SGE support |
---|
| 8 | thanks to: Dave Love - d(d.o.t)love(a.t)liverpool(d.o.t)ac(d.o.t)uk |
---|
| 9 | for writing it! |
---|
| 10 | f: queue selection support is now working |
---|
| 11 | thanks to: Craig West - cwest(a.t)astro(d.o.t)umass(d.o.t)edu |
---|
| 12 | for the patch |
---|
| 13 | web) |
---|
| 14 | a: large graphs link for job report |
---|
| 15 | thanks to: Craig West - cwest(a.t)astro(d.o.t)umass(d.o.t)edu |
---|
| 16 | |
---|
[498] | 17 | 0.3.1: |
---|
| 18 | |
---|
| 19 | other) |
---|
| 20 | f: updated INSTALL since "addons" directory is not included by default anymore in Ganglia |
---|
| 21 | thanks to: Steven DuChene linux(d.a.s.h)clusters(a.t)mindspring(d.o.t)com |
---|
| 22 | for reporting it |
---|
| 23 | |
---|
| 24 | rpm) |
---|
| 25 | f: add "addons" directory since it's not included by default anymore in Ganglia |
---|
[501] | 26 | f: properly rewrite WEBDIR path in %files when rebuilding rpms with Makefile |
---|
[498] | 27 | |
---|
| 28 | web) |
---|
| 29 | f: typo in empty_cpu variable: causing incorrect 'free cpu' count and similar errors |
---|
| 30 | thanks to: Craig West - cwest(a.t)astro(d.o.t)umass(d.o.t)edu |
---|
| 31 | for reporting it |
---|
[502] | 32 | f: changed erroneous domain detection a little |
---|
| 33 | thanks to: Craig West - cwest(a.t)astro(d.o.t)umass(d.o.t)edu |
---|
[498] | 34 | for reporting it |
---|
| 35 | a: now properly detects whether or not to use FQDN or short hostnames w/o domain |
---|
[502] | 36 | thanks to: Craig West - cwest(a.t)astro(d.o.t)umass(d.o.t)edu |
---|
[498] | 37 | thanks to: Jeffrey Sarlo - JSarlo(a.t)Central(d.o.t)UH(d.o.t)EDU |
---|
| 38 | for the many testing and reporting it |
---|
| 39 | |
---|
| 40 | SPECIAL THANKS to the University of Houston for sending me a shirt! |
---|
| 41 | |
---|
[500] | 42 | jobarchived) |
---|
| 43 | f: properly catch postgres exception |
---|
| 44 | f: don't use debug_message while loading config file |
---|
| 45 | |
---|
[452] | 46 | 0.3: |
---|
[342] | 47 | |
---|
| 48 | web) |
---|
| 49 | a: allow per-cluster settings/override options: see CLUSTER_CONFS option |
---|
| 50 | a: clusterimage can now draw nodes at x,y position parsed from hostname |
---|
[427] | 51 | see SORTBY_HOSTNAME for this in clusterconf/example.php |
---|
[342] | 52 | a: clusterimage nodes are now clickable: has link to all jobs from that host |
---|
[427] | 53 | a: clusterimage nodes now have a tooltip: displays hostname and jobids for now |
---|
[345] | 54 | a: jobmonarch logo image |
---|
| 55 | thank to: Robin Day |
---|
| 56 | for the design |
---|
[414] | 57 | a: rrd graph of running/queued jobs to overview |
---|
[460] | 58 | a: per-cluster settings for archive database |
---|
| 59 | thanks to: Alexis Michon - alexis(d.o.t)michon(a.t)ibcp(d.o.t)fr |
---|
| 60 | for the patch |
---|
[342] | 61 | |
---|
[414] | 62 | c: host archive view is now more complete and detailed in the same manner as |
---|
| 63 | Ganglia's own host view |
---|
[427] | 64 | c: host archive view available metric list is now compiled from disk, |
---|
| 65 | so that the detailed archive host view works even when the node is currently down. |
---|
[400] | 66 | c: removed size restrictions from detailed host archive view |
---|
| 67 | |
---|
[465] | 68 | f: compatibility: removed php5 call |
---|
[460] | 69 | thanks to: Alexis Michon - alexis(d.o.t)michon(a.t)ibcp(d.o.t)fr |
---|
| 70 | for the patch |
---|
[458] | 71 | f: prevent negative cpu/node calculation |
---|
| 72 | thanks to: aloga(a.t)ifca(d.o.t)unican(d.o.t)es |
---|
| 73 | for the patch |
---|
[364] | 74 | f: archive search not properly resetting nodes list |
---|
| 75 | thanks to: Alexis Michon - alexis(d.o.t)michon(a.t)ibcp(d.o.t)fr |
---|
| 76 | for the patch |
---|
[400] | 77 | f: detailed host view from jobarchive was broken since hostbased support of 0.2 |
---|
| 78 | now host view is properly set and parsed again |
---|
| 79 | thanks to: Alexis Michon - alexis(d.o.t)michon(a.t)ibcp(d.o.t)fr |
---|
| 80 | for reporting the bug and suggesting a patch |
---|
[403] | 81 | f: bug where jobstart redline indicator in host detail graphs was set incorrectly |
---|
[414] | 82 | or not at all due to a miscalculation in job times |
---|
[427] | 83 | f: bug where hostimage headertext xoffset was miscalculated, causing the column names |
---|
| 84 | to overlap their position when the columnname was longer than the columnvalues |
---|
[364] | 85 | |
---|
[342] | 86 | jobmond) |
---|
| 87 | |
---|
[376] | 88 | a: syslog support |
---|
[427] | 89 | a: report number of running/queued jobs as seperate metrics |
---|
| 90 | a: native gmetric support, much faster and cleaner! |
---|
[361] | 91 | thanks to: Nick Galbreath - nickg(a.t)modp(d.o.t)com |
---|
| 92 | for writing it and allowing inclusion in jobmond |
---|
| 93 | |
---|
[452] | 94 | f: crashing jobmond when multiple nodes amounts are requested in |
---|
| 95 | a queued job: numeric_node variable not initialized properly |
---|
| 96 | thanks to: aloga(a.t)ifca(d.o.t)unican(d.o.t)es |
---|
| 97 | for supplying the patch |
---|
| 98 | and many others for reporting and helping debug this |
---|
[361] | 99 | f: hanging/blocked, increased cpu usage and halted reporting |
---|
| 100 | thanks to: Bas van der Vlies - basv(a.t)sara(d.o.t)nl |
---|
| 101 | for discovering the origin of the bug |
---|
| 102 | thanks to: Mickael Gastineau - gastineau(a.t)imcce(d.o.t)fr |
---|
| 103 | for reporting it and testing the fix |
---|
| 104 | thanks to: Craig West - cwest(a.t)astro(d.o.t)umass(d.o.t)edu |
---|
| 105 | for reporting it and testing the fix |
---|
[342] | 106 | f: uninitialized variable in checkGmetricVersion() |
---|
| 107 | thanks to: Peter Kruse - pk(a.t)q-leap(d.o.t)com |
---|
| 108 | for the patch |
---|
[364] | 109 | f: undefined PBSError |
---|
| 110 | thanks to: Peter Kruse - pk(a.t)q-leap(d.o.t)com |
---|
| 111 | for reporting it |
---|
[342] | 112 | |
---|
[363] | 113 | r: SGE support broken |
---|
| 114 | |
---|
[361] | 115 | jobarchived) |
---|
| 116 | |
---|
[427] | 117 | a: can now use py-rrdtool api instead of pipes, much faster! |
---|
[376] | 118 | install py-rrdtool to use this |
---|
| 119 | backwards compatible fails back to pipes if module not installed |
---|
[367] | 120 | |
---|
[427] | 121 | c: all XML input was uniencoded, which could cause errors, |
---|
| 122 | now all properly converted to normal strings |
---|
| 123 | |
---|
[470] | 124 | f: when XML data source (gmetad) is unavailable parsethread didn't return correctly |
---|
| 125 | which caused a large number of threads to spawn while consuming large amounts of memory |
---|
[376] | 126 | f: autocreate clusterdirs in archivedir |
---|
| 127 | f: unhandled gather exception |
---|
[361] | 128 | f: incorrect stop_timestamping when jobs finished |
---|
| 129 | thanks to: Alexis Michon - alexis(d.o.t)michon(a.t)ibcp(d.o.t)fr |
---|
[376] | 130 | for finding and debugging/testing it |
---|
[361] | 131 | |
---|
[308] | 132 | 0.2: |
---|
| 133 | |
---|
| 134 | web) |
---|
[342] | 135 | f: misc. optimization and bugfixes |
---|
| 136 | f: now fully compatible with latest PHP5 and PHP4 |
---|
[308] | 137 | |
---|
[342] | 138 | c: cluster image now incorporates small text descr. |
---|
| 139 | c: monarch (cluster/host) images no longer displayed |
---|
| 140 | for clusters that are not jobmond enabled |
---|
| 141 | c: pie chart percentages are now cpu-based instead of node-based |
---|
[308] | 142 | |
---|
[342] | 143 | a: host template for Ganglia |
---|
| 144 | adds a extra monarch host image to Ganglia's host overview |
---|
| 145 | which displays/links to the jobs on that host |
---|
| 146 | NOTE!: be sure to copy/install new template from addons/templates |
---|
| 147 | a: (optional) nodes hostnames column |
---|
| 148 | thanks to: Daniel Barthel - daniel(d.o.t)barthel(a.t)nottingham(d.o.t)ac(d.o.t)uk |
---|
| 149 | for the suggestion |
---|
[308] | 150 | |
---|
| 151 | jobmond) |
---|
| 152 | |
---|
[342] | 153 | f: when a job metric is longer than maximum metric length, |
---|
| 154 | the info is split up amongst multiple metrics |
---|
| 155 | f: no longer exit when batch server is unavailable |
---|
| 156 | thanks to: Peter Kruse - pk(a.t)q-leap(d.o.t)com |
---|
| 157 | for the patch |
---|
| 158 | f: fd closure bug causing stderr/stdout to remain open after daemonizing |
---|
[308] | 159 | |
---|
[342] | 160 | c: rearranged code to allow support for other batch systems |
---|
[308] | 161 | |
---|
[342] | 162 | a: (experimental) SGE (Sun Grid Engine) support as batch server |
---|
| 163 | thanks to: Babu Sundaram - babu(a.t)cs(d.o.t)uh(d.o.t)edu |
---|
| 164 | who developed it for a OSCAR's Google-SoC project |
---|
| 165 | a: pidfile support |
---|
| 166 | thanks to: Michael Jeanson - michael(a.t)ccs(d.o.t)usherbrooke(d.o.t)ca |
---|
| 167 | for the patch |
---|
| 168 | a: usage display |
---|
| 169 | thanks to: Michael Jeanson - michael(a.t)ccs(d.o.t)usherbrooke(d.o.t)ca |
---|
| 170 | for the patch |
---|
| 171 | a: queue selection support: ability to specify which QUEUE's to get jobinfo from |
---|
| 172 | thanks to: Michael Jeanson - michael(a.t)ccs(d.o.t)usherbrooke(d.o.t)ca |
---|
| 173 | for the patch |
---|
[308] | 174 | |
---|
| 175 | jobarchived) |
---|
| 176 | |
---|
[342] | 177 | f: XML retrieval for Ganglia version >= 3.0.3 working properly again |
---|
| 178 | f: database storing for Ganglia version >= 3.0.3 working properly again |
---|
| 179 | f: fd closure bug causing stderr/stdout to remain open after daemonizing |
---|
[308] | 180 | |
---|
[342] | 181 | c: misc. bugfixes to optimize XML connections |
---|
| 182 | c: misc. bugfixes for misc. minor issues |
---|
[308] | 183 | |
---|
[342] | 184 | a: cleaning of stale jobs in dbase: see JOB_TIMEOUT option |
---|
[308] | 185 | |
---|
[283] | 186 | 0.1.1: |
---|
[249] | 187 | |
---|
| 188 | web) |
---|
| 189 | |
---|
[342] | 190 | f: misc. layout bugs for overview & search |
---|
| 191 | f: bug that occured when calculating the number of nodes when there |
---|
| 192 | was more than one job running on a machine |
---|
[253] | 193 | |
---|
[342] | 194 | c: column requested memory is now optional through conf.php |
---|
| 195 | c: search and overview tables are now full screen (100%) |
---|
| 196 | c: overview jobnames are now cutoff at max 9 characters |
---|
| 197 | to prevent (layout) scews in the tables |
---|
| 198 | c: overview graphs are no longer downsized |
---|
[253] | 199 | |
---|
[342] | 200 | a: (optional) column 'queued' (since) in overview |
---|
| 201 | a: search results (can) now have a SEARCH_RESULT_LIMIT |
---|
| 202 | this increases performance of the query's significantly! |
---|
| 203 | a: date/time format as displayed is now configurable through conf.php |
---|
[249] | 204 | |
---|
| 205 | jobmond) |
---|
| 206 | |
---|
[342] | 207 | a: now reports 'queued since' (or creation time) of jobs |
---|
[249] | 208 | |
---|
| 209 | documentation) |
---|
| 210 | |
---|
[342] | 211 | f: wrong e-mail adress in INSTALL (doh!) |
---|
[249] | 212 | |
---|
[342] | 213 | 0.1: |
---|
[249] | 214 | |
---|
| 215 | - First public release |
---|