diff --git a/doc/html/ibm-pe.shtml b/doc/html/ibm-pe.shtml index 7604aeb3b0ceb48d41663549bccb04a6bfcbd553..0bd1433a1e759ac6195d7cff58f4e48fd41e3bd8 100644 --- a/doc/html/ibm-pe.shtml +++ b/doc/html/ibm-pe.shtml @@ -476,6 +476,23 @@ Units (CAU) is disabled and an error message to that effect will be generated in response to such a request. In addition, version 1200 or higher of IBM's NRT API is required to support this functionality.</p> +<h3>Cold Starting</h3> + +<p>If the slurmctld daemon is cold started (without saved state), then +information about previously allocated network resources is lost. +Slurm will release those resources to the best of its ability based upon +information available from the Network Resource Table (NRT) library functions. +These function provide sufficient information to release all resources +<b>except</b> for CAU on a Torrent network (e.g. a PERCS system). +In order to release CAU, it is necessary to cold start the Protocol Network +Services Daemon (PNSD) on compute nodes following the sequence shown below.</p> +<pre> +Stop Slurm daemons: /etc/init.d/rc slurm stop +Stop PNSD: stopsrc -s pnsd +Start PNSD clean: startsrc -s pnsd -a -c +Start Slurm daemons clean: /etc/init.d/rc slurm startclean +</pre> + <h3>Design Notes</h3> <p>It is necessary for all nodes that can be used for scheduling a single job @@ -550,6 +567,6 @@ startsrc -s pnsd -a -D <p class="footer"><a href="#top">top</a></p> -<p style="text-align:center;">Last modified 29 May 2013</p></td> +<p style="text-align:center;">Last modified 22 July 2013</p></td> <!--#include virtual="footer.txt"-->