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