<font size=2 face="sans-serif">All,</font><br><br><font size=2 face="sans-serif">As Bob Oesterlin indicated, it is possible
to define an expel script (see /usr/lpp/mmfs/samples/expelnode.sample)
to control which of the two nodes to get expelled. The script can also
be used to issue alerts, etc.</font><br><br><font size=2 face="sans-serif">The current policy used (before the
script is invoked) when deciding which node to expel is the following:</font><br><br><font size=2 face="sans-serif"> 1. quorum
nodes over non-quorum nodes</font><br><br><font size=2 face="sans-serif"> 2. local
nodes over remote nodes</font><br><br><font size=2 face="sans-serif"> 3. manager-capable
nodes over non-manager-capable nodes</font><br><br><font size=2 face="sans-serif"> 4. nodes
managing more FSs over nodes managing fewer FSs</font><br><br><font size=2 face="sans-serif"> 5. NSD server
over non-NSD server</font><br><br><font size=2 face="sans-serif"> Otherwise,
expel whoever joined the cluster more recently.</font><br><br><br><br><font size=2 face="sans-serif">The statement below from Dr. Uwe Falke
is also correct: addressing the network connectivity is the better long-term
approach, but the callback script could be used to control which node to
expel.</font><br><br><font size=2 face="sans-serif"> Felipe</font><br><br><font size=2 face="sans-serif">----<br>Felipe Knop
knop@us.ibm.com<br>GPFS Development and Security<br>IBM Systems<br>IBM Building 008<br>2455 South Rd, Poughkeepsie, NY 12601<br>(845) 433-9314 T/L 293-9314<br><br></font><br><br><br><br><font size=1 color=#5f5f5f face="sans-serif">From:
</font><font size=1 face="sans-serif">"Uwe Falke"
<UWEFALKE@de.ibm.com></font><br><font size=1 color=#5f5f5f face="sans-serif">To:
</font><font size=1 face="sans-serif">gpfsug main discussion
list <gpfsug-discuss@spectrumscale.org></font><br><font size=1 color=#5f5f5f face="sans-serif">Date:
</font><font size=1 face="sans-serif">10/25/2016 08:32 AM</font><br><font size=1 color=#5f5f5f face="sans-serif">Subject:
</font><font size=1 face="sans-serif">Re: [gpfsug-discuss]
Forcing which node gets expelled?</font><br><font size=1 color=#5f5f5f face="sans-serif">Sent by:
</font><font size=1 face="sans-serif">gpfsug-discuss-bounces@spectrumscale.org</font><br><hr noshade><br><br><br><tt><font size=2>Usually, the cluster mgr, receiving a complaint from
a node about another <br>node being gone, checks its own connection to that other node. If that
is <br>positive it expells the requester, if not it follows the request and <br>expells the other node. <br>AFAIK, there are some more subtle algorithms in place if managers or <br>quorum nodes are affected. Maybe that can be used to protect certain nodes
<br>from getting expelled by assigning some role in the cluster to them. I
do <br>however not know these exactly.<br><br>That means: it is not easily controllable which one gets expelled. <br>It is better to concentrate on fixing your connectivity issues, as GPFS
<br>will not feel comfortable in such a unreliable environment anyway. <br> <br>Mit freundlichen Grüßen / Kind regards<br><br> <br>Dr. Uwe Falke<br> <br>IT Specialist<br>High Performance Computing Services / Integrated Technology Services /
<br>Data Center Services<br>-------------------------------------------------------------------------------------------------------------------------------------------<br>IBM Deutschland<br>Rathausstr. 7<br>09111 Chemnitz<br>Phone: +49 371 6978 2165<br>Mobile: +49 175 575 2877<br>E-Mail: uwefalke@de.ibm.com<br>-------------------------------------------------------------------------------------------------------------------------------------------<br>IBM Deutschland Business & Technology Services GmbH / Geschäftsführung:
<br>Frank Hammer, Thorsten Moehring<br>Sitz der Gesellschaft: Ehningen / Registergericht: Amtsgericht Stuttgart,
<br>HRB 17122 <br><br><br><br><br>From: Matt Thorpe <matt.thorpe@bodleian.ox.ac.uk><br>To: gpfsug main discussion list <gpfsug-discuss@spectrumscale.org><br>Date: 10/25/2016 02:05 PM<br>Subject: [gpfsug-discuss] Forcing which node
gets expelled?<br>Sent by: gpfsug-discuss-bounces@spectrumscale.org<br><br><br><br>Hi,<br><br>We are in the process of diagnosing a networking issue that is causing
2 <br>of our 6 node GPFS cluster to expel each other (it appears they experience
<br>a temporary network connection outage and lose contact with each other).
<br>At present it's not consistent which gets expelled by the cluster manager,
<br>and I wondered if there was any way to always force a specific node to
be <br>expelled in this situation?<br><br>Thanks and best regards,<br>Matt<br><br>--------<br>Matt Thorpe | BDLSS Systems Administrator<br>Bodleian Libraries Osney One Building, Osney Mead, Oxford, OX2 0EW<br>matt.thorpe@bodleian.ox.ac.uk | 01865 (2)80027<br>_______________________________________________<br>gpfsug-discuss mailing list<br>gpfsug-discuss at spectrumscale.org<br></font></tt><a href="http://gpfsug.org/mailman/listinfo/gpfsug-discuss"><tt><font size=2>http://gpfsug.org/mailman/listinfo/gpfsug-discuss</font></tt></a><tt><font size=2><br><br><br><br><br><br>_______________________________________________<br>gpfsug-discuss mailing list<br>gpfsug-discuss at spectrumscale.org<br></font></tt><a href="http://gpfsug.org/mailman/listinfo/gpfsug-discuss"><tt><font size=2>http://gpfsug.org/mailman/listinfo/gpfsug-discuss</font></tt></a><tt><font size=2><br><br></font></tt><br><BR>