ViewVC logotype

Annotation of /nl.nikhef.pdp.dynsched-pbs-plugin/trunk/RELEASE

Parent Directory Parent Directory | Revision Log Revision Log

Revision 2267 - (hide annotations) (download)
Tue Apr 5 13:52:24 2011 UTC (11 years, 1 month ago) by templon
File size: 6349 byte(s)
Merge release x.y.z changes back into trunk
1 templon 2012 This file contains release notes and a change history for the
2 templon 2027 pbs/torque and maui plugins for the lcg-info-dynamic-scheduler information provider.
3     The notes are most recent first.
4 templon 2012
5 templon 2267 Release 2.2.2
7     A bug was fixed in parsing the output of "qstat -f"; the error condition had
8     not been previously reported, and was due to job staging errors being reported
9     via the "sched_hint" attribute. Empty lines in this error output broke the
10     parsing algorithm.
12     The lrmsinfo-pbs previously had the location of the python modules determined at
13     build time and embedded in the program. This has been removed as in EMI, everything
14     will go in system default locations (see also below about Makefile changes). For
15     local (non RPM) installs, you'll need to have PYTHONPATH set accordingly.
17     A testing framework was introduced (available in the svn, not in the RPM) with right now
18     a single test defined; which is whether the above situation can be correctly parsed.
20     The Makefile was changed in order to have the python library path be compliant with system
21     defaults.
23 templon 2253 Release 2.2.1
25     Dependencies changed back to the original set, so that this version's RPM
26     dependencies are compatible with those of previous versions (and also
27     with the lcg-info-dynamic-scheduler-generic package).
29 templon 2012 Release 2.2.0
30 templon 2027 Moved to new repository format and changed RPM name. Rather large changes in two areas. First, in the logFileParser
31     part of the pbsServer.py code. The new "account" field printed in the torque accounting logs broke the old parser,
32     this is fixed in the current release. The second set of changes have to do with how the LiveServer code parses
33     the 'qstat -f' output. When present (recent versions of torque), the code now uses 'startTime' as the start time
34     of the job, and uses the 'euser' and 'egroup' as the user and group of the job. If these fields are not present,
35 templon 2146 the code reverts to the old behavior. Using startTime will give slightly different results, since startTime
36     is present immediately upon job start, whereas walltime (used in the old calculation) is only updated once per
37     mom update interval, meaning there were always a few jobs in the system that were in state "running" but that had
38     no startTime. This will no longer be the case.
39 templon 2012
40 templon 2146 Three user-visible changes: there are two new job attributes in the "Job" object. One is the walltime used by the job,
41 templon 2027 the other is the 'startAnchor' field which tells you whether the code has used the 'startTime' to find the start time,
42 templon 2129 or has done it the old way by subtracting the walltime used from the current time. These data are both printed
43 templon 2146 by lrmsinfo-pbs. Those two changes are in principle user-visible, but only for those users directly using lrmsinfo-pbs.
44     Finally, vomaxjobs-maui accepts an argument '-k' to provide a key-file to the diagnose command. This is needed for
45     e.g. installations where the diagnose client cmd is on RHEL4 and the maui server is on RHEL5 (different build = different key).
46 templon 2027
47     Information below this point refers to older versions of the package, under the older naming
48     scheme.
50 templon 2012 Release 2.1.0
51 templon 2027 - Remove support for specifying hostname on command line (savannah bug 35662)
52 templon 2012
53 templon 2027 Release 2.0.1
54     - fix LICENSE, otherwise no change.
55 templon 2012
56     Release 2.0.0
57 templon 2027 - uses new lrms.py, otherwise no changes.
58 templon 2012
59     Release 1.6.1
60 templon 2027 - update documentation and examples to be consistent with 2.0 release
61 templon 2012
62     Release 1.6.0
64     - some changes to build system (three targets increases aggravation)
65     - some changes to pbsServer classes to assist in debugging.
66     - some changes to vomaxjobs-maui to assist in debugging/testing;
67     also fixed various unreported bugs discovered during testing.
68     - Change mapping of pbs/torque job states in pbs classes; up til now
69     was either queued (Q) or running (any other states). Now we have:
71     From the qstat (torque 2.0.0p4) man page:
73     C - Job is completed after having run (mapped to 'done')
74     E - Job is exiting after having run. (mapped to 'running')
75     H - Job is held. (mapped to 'pending')
76     Q - job is queued, eligible to run or routed. (mapped to 'queued')
77     R - job is running. (mapped to 'running')
78     T - job is being moved to new location. (mapped to 'pending')
79     W - job is waiting for its execution time (mapped to 'queued')
81     Release 1.5.2:
83     pbs package: Fix to vomaxjobs-maui to deal with cases where there is
84     extra 'warning' output near the top of the command output from diagnose -g.
86     Release 1.5.1:
88     fix dependency problems with RPMs.
90     Release 1.5.0
92 templon 2027 in vomaxjobs-maui, adapt to handle MAXPROC specifications like
93     MAXPROC=soft,hard The code reports the 'hard' limit, since
94     this is relevant when the system is not full, and this is when
95     it's needed. Maui uses the soft limit on a full system, but
96     in this case the info provider will drop FreeSlots to zero as
97     soon as jobs remain in the queued state instead of executing
98     immediately.
99 templon 2012
100     Release 1.4.2
102 templon 2027 in pbsServer.py: included Steve Traylen's patch to deal with jobs for which the
103     uid/gid printed by 'qstat' is not listed in the in running machine's
104     pw DB. This can happen when the CE is not the same physical
105     machine as the actual LRMS server.
106 templon 2012
107     Estimated Response Time Info Providers (v 1.4.1)
108     ------------------------------------------------
110     This information provider is new in LCG 2.7.0 and is
111     contained in two RPMs, lcg-info-dynamic-scheduler-generic
112     and lcg-info-dynamic-scheduler-pbs. Sites using torque/pbs
113     as an LRMS and Maui as a scheduler are fully supported by
114     this configuration; those using other schedulers and/or
115     LRMS systems will need to provide the appropriate back-end
116     plugins.
118     For sites meeting the following criteria, the system should
119     work out of the box with no modifications whatsoever:
121     LRMS == torque
122     scheduler == maui
123     vo names == unix group names of that vo's pool accounts
125     Documentation on what to do if this is not the case can be
126     found in the file
128     lcg-info-dynamic-scheduler.txt
130     in the doc directory
132     /opt/lcg/share/doc/lcg-info-dynamic-scheduler
134     There is also documentation in this directory indicating
135     the requirements on the backend commands you will need to
136     provide in the case that you are using a different
137     scheduler or LRMS. Tim Bell at CERN can help for people
138     using LSF.


Name Value
svn:keywords Id URL

ViewVC Help
Powered by ViewVC 1.1.28