Do you have /opt/xcat/xdsh/Context directory with two files DSH.pm and XCAT.pm? Also see if DSH_CONTEXT is exported. It should not be, but you should have gotten an error, if you did. That is a strange error message "Error: Invalid context specified: DSH"
Is the Management Node and the service nodes still at xCAT 2.3? It will be really had to support that level, it has been out of service for a long time. Lissa K. Valletta 8-3/B10 Poughkeepsie, NY 12601 (tie 293) 433-3102 From: Russell Jones <[email protected]> To: [email protected] Date: 12/07/2012 03:49 PM Subject: Re: [xcat-user] Cannot find master for the node $node Also forgot to mention, we are *not* using the syslog postscript. Each node is responsible for their own logs. On 12/7/2012 2:45 PM, Russell Jones wrote: I am not having much luck finding an xdsh command to run that succeeds. I get "permission denied" on the mgmt node for any command I try to run as root. On the service nodes, I am trying to follow the xdsh man page to just run a simple uptime and am getting: [root@service01 ~]# xdsh c26n24 "uptime" Error: Invalid context specified: DSH Error: Failed to dispatch command to any of the following service nodes: service01,service02,service03,service04 From the management node (trying to use the trace option to see errors): [root@mgmt1 ~]# xdsh c26n24 -T "uptime" Error: Permission denied for request Logs show: Dec 7 14:36:13 mgmt1 xCAT: xCAT: Allowing xdsh to c26n24 for root from localhost Policy table for root: "1","root",,,,,,"allow",, Not seeing anything in the compute nodes messages or secure that would lead me to believe an attempt to login via SSH was ever made. psh, on the other hand, works fine. the remoteshell script does run on each node at boot time. On 12/7/2012 1:42 PM, Ling Gao wrote: The error is given because the nodes are not in the same subnet as the service node, the script did not check for routing. The code uses this function to set up MASTER environmental variable for the node when calling postscripts. If it is not set up correctly some of the postscripts such as syslog will fail. When you say "it works fine even with that error". Can you xdsh to the nodes that has the error? Ling Ling Gao Poughkeepsie Unix Development Lab IBM Systems and Technology Group Internal: T/L 293-5692 External: [email protected], 845-433-5692 "I never worry about the future. It comes soon enough." --- Albert Einstein From: Russell Jones <[email protected]> To: [email protected] Date: 12/07/2012 02:06 PM Subject: Re: [xcat-user] Cannot find master for the node $node Thanks! DNS is working properly, both forward and reverse... this seems to occur mostly when entire clusters are booted at the same time. The nodes are not within the same subnet, all of these nodes are on different subnets from the service nodes (and routed of course). The other strange thing is again, it works fine even with that error: [root@service03-hc log]# grep c25n37 messages Dec 7 09:05:02 service03 xCAT: xCAT: Allowing getpostscript from c25n37 Dec 7 09:05:03 service03 xCAT: Cannot find master for the node c25n37 Dec 7 09:05:14 service03 xCAT: xCAT: Allowing getcredentials from c25n37 Dec 7 09:05:24 service03 xCAT: xCAT: Allowing getcredentials from c25n37 Dec 7 09:05:32 service03 xCAT: xCAT: Allowing getcredentials from c25n37 Could this message manifest itself if a service node is very busy? IE, not really an error, just took too long to respond to the request? Or is this literally sending the error because the compute nodes are not on the same subnet as the service nodes? On 12/7/2012 12:18 PM, Ling Gao wrote: Hi, The error is from getFacingIP function. This function takes a node name as an input. Then it find out the ip of the given node. Then it call "ifconfig" on the local host. Then it try to see if the node ip and local host is within the same subnet or not. (Please see the code below). The error usually happens when the name resolution on the local host cannot resolve the given node. Hope it helps. #------------------------------------------------------------------------------- =head3 getFacingIP Gets the ip address of the adapter of the localhost that is facing the the given node. Arguments: The name of the node that is facing the localhost. Returns: The ip address of the adapter that faces the node. =cut #------------------------------------------------------------------------------- sub getFacingIP { my ($class, $node) = @_; my $ip; my $cmd; my @ipaddress; my $nodeip = inet_ntoa(inet_aton($node)); unless ($nodeip =~ /\d+\.\d+\.\d+\.\d+/) { return 0; #Not supporting IPv6 here IPV6TODO } $cmd = "ifconfig" . " -a"; $cmd = $cmd . "| grep \"inet \""; my @result = xCAT::Utils->runcmd($cmd, 0); if ($::RUNCMD_RC != 0) { xCAT::MsgUtils->message("S", "Error from $cmd\n"); exit $::RUNCMD_RC; } # split node address my ($n1, $n2, $n3, $n4) = split('\.', $nodeip); foreach my $addr (@result) { my $ip; my $mask; if (xCAT::Utils->isLinux()) { my ($inet, $addr1, $Bcast, $Mask) = split(" ", $addr); if ((!$addr1) || (!$Mask)) { next; } my @ips = split(":", $addr1); my @masks = split(":", $Mask); $ip = $ips[1]; $mask = $masks[1]; } else { #AIX my ($inet, $addr1, $netmask, $mask1, $Bcast, $bcastaddr) = split(" ", $addr); if ((!$addr1) && (!$mask1)) { next; } $ip = $addr1; $mask1 =~ s/0x//; $mask = `printf "%d.%d.%d.%d" \$(echo "$mask1" | sed 's/../0x& /g')`; } if ($ip && $mask) { # split interface IP my ($h1, $h2, $h3, $h4) = split('\.', $ip); # split mask my ($m1, $m2, $m3, $m4) = split('\.', $mask); # AND this interface IP with the netmask of the network my $a1 = ((int $h1) & (int $m1)); my $a2 = ((int $h2) & (int $m2)); my $a3 = ((int $h3) & (int $m3)); my $a4 = ((int $h4) & (int $m4)); # AND node IP with the netmask of the network my $b1 = ((int $n1) & (int $m1)); my $b2 = ((int $n2) & (int $m2)); my $b3 = ((int $n3) & (int $m3)); my $b4 = ((int $n4) & (int $m4)); if (($b1 == $a1) && ($b2 == $a2) && ($b3 == $a3) && ($b4 == $a4)) { return $ip; } } } xCAT::MsgUtils->message("S", "Cannot find master for the node $node\n"); return 0; } Ling Gao Poughkeepsie Unix Development Lab IBM Systems and Technology Group Internal: T/L 293-5692 External: [email protected], 845-433-5692 "I never worry about the future. It comes soon enough." --- Albert Einstein From: Russell Jones <[email protected]> To: xCAT Users Mailing list <[email protected]> Date: 12/07/2012 11:05 AM Subject: [xcat-user] Cannot find master for the node $node Hi all, What circumstances have to be present for an xCAT 2.3 service node (old I know, but upgrade is not an option at this time) to write to the logs: Dec 7 09:16:33 service03 xCAT: Cannot find master for the node c25n25 All of our service nodes have been doing this for at least over a month now, and we have just never noticed it before as nodes are netbooting/installing fine. Just curious what the logic is in the code that has this message being written out, just so that if it turns out to be something we need to track down we know where to start. Thanks! ------------------------------------------------------------------------------ LogMeIn Rescue: Anywhere, Anytime Remote support for IT. Free Trial Remotely access PCs and mobile devices and provide instant support Improve your efficiency, and focus on delivering more value-add services Discover what IT Professionals Know. Rescue delivers http://p.sf.net/sfu/logmein_12329d2d _______________________________________________ xCAT-user mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/xcat-user ------------------------------------------------------------------------------ LogMeIn Rescue: Anywhere, Anytime Remote support for IT. Free Trial Remotely access PCs and mobile devices and provide instant support Improve your efficiency, and focus on delivering more value-add services Discover what IT Professionals Know. Rescue delivers http://p.sf.net/sfu/logmein_12329d2d _______________________________________________ xCAT-user mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/xcat-user ------------------------------------------------------------------------------ LogMeIn Rescue: Anywhere, Anytime Remote support for IT. Free Trial Remotely access PCs and mobile devices and provide instant support Improve your efficiency, and focus on delivering more value-add services Discover what IT Professionals Know. Rescue delivers http://p.sf.net/sfu/logmein_12329d2d _______________________________________________ xCAT-user mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/xcat-user ------------------------------------------------------------------------------ LogMeIn Rescue: Anywhere, Anytime Remote support for IT. Free Trial Remotely access PCs and mobile devices and provide instant support Improve your efficiency, and focus on delivering more value-add services Discover what IT Professionals Know. Rescue delivers http://p.sf.net/sfu/logmein_12329d2d _______________________________________________ xCAT-user mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/xcat-user ------------------------------------------------------------------------------ LogMeIn Rescue: Anywhere, Anytime Remote support for IT. Free Trial Remotely access PCs and mobile devices and provide instant support Improve your efficiency, and focus on delivering more value-add services Discover what IT Professionals Know. Rescue delivers http://p.sf.net/sfu/logmein_12329d2d _______________________________________________ xCAT-user mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/xcat-user ------------------------------------------------------------------------------ LogMeIn Rescue: Anywhere, Anytime Remote support for IT. Free Trial Remotely access PCs and mobile devices and provide instant support Improve your efficiency, and focus on delivering more value-add services Discover what IT Professionals Know. Rescue delivers http://p.sf.net/sfu/logmein_12329d2d _______________________________________________ xCAT-user mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/xcat-user
<<inline: graycol.gif>>
------------------------------------------------------------------------------ LogMeIn Rescue: Anywhere, Anytime Remote support for IT. Free Trial Remotely access PCs and mobile devices and provide instant support Improve your efficiency, and focus on delivering more value-add services Discover what IT Professionals Know. Rescue delivers http://p.sf.net/sfu/logmein_12329d2d
_______________________________________________ xCAT-user mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/xcat-user
