From dbf81c476e7d0d52f64a3eeb06db14c929e309c6 Mon Sep 17 00:00:00 2001
From: Moe Jette <jette1@llnl.gov>
Date: Wed, 27 May 2009 21:24:58 +0000
Subject: [PATCH] svn merge -r17609:17615
https://eris.llnl.gov/svn/slurm/branches/slurm-2.0
---
NEWS | 1 +
auxdir/x_ac_readline.m4 | 2 +-
configure | 2 +-
doc/html/power_save.shtml | 30 ++++++++++++++++++++++++++++--
doc/html/publications.shtml | 11 +++++++++--
5 files changed, 40 insertions(+), 6 deletions(-)
diff --git a/NEWS b/NEWS
index 8be03611fd6..840848c24a2 100644
--- a/NEWS
+++ b/NEWS
@@ -349,6 +349,7 @@ documents those changes that are of interest to users and admins.
* Changes in SLURM 1.3.17
=========================
+ -- Fix bug in configure script that can clear user specified LIBS.
* Changes in SLURM 1.3.16
=========================
diff --git a/auxdir/x_ac_readline.m4 b/auxdir/x_ac_readline.m4
index fc6b6389209..d9dd393c486 100644
--- a/auxdir/x_ac_readline.m4
+++ b/auxdir/x_ac_readline.m4
@@ -39,7 +39,7 @@ AC_DEFUN([X_AC_READLINE],
#include <readline/history.h>]], [[
char *line = readline("in:");]])],[AC_DEFINE([HAVE_READLINE], [1],
[Define if you are compiling with readline.])],[READLINE_LIBS=""])
- LIBS="$savedLIBS"
+ LIBS="$saved_LIBS"
fi
AC_SUBST(READLINE_LIBS)
])
diff --git a/configure b/configure
index 512a534695b..0a25a7fcebf 100755
--- a/configure
+++ b/configure
@@ -27102,7 +27102,7 @@ fi
rm -f core conftest.err conftest.$ac_objext conftest_ipa8_conftest.oo \
conftest$ac_exeext conftest.$ac_ext
- LIBS="$savedLIBS"
+ LIBS="$saved_LIBS"
fi
diff --git a/doc/html/power_save.shtml b/doc/html/power_save.shtml
index 27ba012f0f2..aa29216ce51 100644
--- a/doc/html/power_save.shtml
+++ b/doc/html/power_save.shtml
@@ -1,6 +1,7 @@
<!--#include virtual="header.txt"-->
<h1>Power Saving Guide</h1>
+
<p>SLURM provides an integrated power saving mechanism for idle nodes.
Nodes that remain idle for an configurable period of time can be placed
in a power saving mode.
@@ -19,6 +20,7 @@ SLURM's support to increase power demands in a gradual fashion.</p>
<h2>Configuration</h2>
+
<p>A great deal of flexibility is offered in terms of when and
how idle nodes are put into or removed from power save mode.
Note that the SLURM control daemon, <i>slurmctld</i>, must be
@@ -168,7 +170,31 @@ support.
You can also configure SLURM with programs that perform no aciton as <b>SuspendProgram</b> and <b>ResumeProgram</b> to assess the potential
impact of power saving mode before enabling it.</p>
-<h2>Fault tolerance</h2>
+<h2>Use of Allocations</h2>
+
+<p>A resource allocation request will be granted as soon as resources
+are selected for use, possibly before the nodes are all available
+for use.
+The launching of job steps will be delayed until the required nodes
+have been restored to service (it prints a warning about waiting for
+nodes to become available and periodically retries until they are
+available).</p>
+
+<p>In the case of an <i>sbatch</i> command, the batch program will start
+when node zero of the allocation is ready for use and pre-processing can
+be performed as needed before using <i>srun</i> to launch job steps.
+The operation of <i>salloc</i> and <i>srun</i> follow a similar pattern
+of getting an job allocation at one time, but possibly being unable to
+launch job steps until later.
+If <i>ssh</i> or some other tools is used by <i>salloc</i> it may be
+desirable to execute "<i>srun /bin/true</i>" or some other command
+first to insure that all nodes are booted and ready for use.
+We plan to add a job and node state of <i>CONFIGURING</i> in SLURM
+version 2.1, which could be used to prevent salloc from executing
+any processes (including <i>ssh</i>) until all of the nodes are
+ready for use.</p>
+
+<h2>Fault Tolerance</h2>
<p>If the <i>slurmctld</i> daemon is terminated gracefully, it will
wait up to <b>SuspendTimeout</b> or <b>ResumeTimeout</b> (whichever
@@ -193,6 +219,6 @@ In order to minimize this risk, when the <i>slurmctld</i> daemon is
started and node which should be allocated to a job fails to respond,
the <b>ResumeProgram</b> will be executed (possibly for a second time).</p>
-<p style="text-align:center;">Last modified 26 May 2009</p>
+<p style="text-align:center;">Last modified 27 May 2009</p>
<!--#include virtual="footer.txt"-->
diff --git a/doc/html/publications.shtml b/doc/html/publications.shtml
index 1b0d40bf937..49c34f8f758 100644
--- a/doc/html/publications.shtml
+++ b/doc/html/publications.shtml
@@ -40,13 +40,20 @@ M. Jette and M. Grondona,
<i>Proceedings of ClusterWorld Conference and Expo</i>,
San Jose, California, June 2003.</p>
-<b>SLURM: Simple Linux Utility for Resource Management</b>,
+<p><b>SLURM: Simple Linux Utility for Resource Management</b>,
A. Yoo, M. Jette, and M. Grondona,
<i>Job Scheduling Strategies for Parallel Processing</i>,
volume 2862 of <i>Lecture Notes in Computer Science</i>,
pages 44-60,
Springer-Verlag, 2003.</p>
-<p style="text-align:center;">Last modified 19 March 2009</p>
+<h2>Interview</h2>
+
+<p><a href="http://www.rce-cast.com/index.php/Podcast/rce-10-slurm.html">
+RCE 10: SLURM (podcast)</a>:
+Brock Palen and Jeff Squyres speak with Morris Jette and
+Danny Auble of LLNL about SLURM.</p>
+
+<p style="text-align:center;">Last modified 27 May 2009</p>
<!--#include virtual="footer.txt"-->
--
GitLab