piranha load balancer failover Post: 302576085

Sponsored Content

Top Forums UNIX for Advanced & Expert Users piranha load balancer failover Post 302576085 by robo on Wednesday 23rd of November 2011 02:54:01 PM

11-23-2011

Registered User

piranha load balancer failover

we use piranha load balancer with two nodes

even the primary node is running fine and up failover happend to secondary node
this happend quite few times ehy node2 cannot talk to node1

what logs are to be checked and investigate why failover occured

Code:

  pulse[7340]: partner dead: activating lvs
Nov 21 10:53:05 node2 pulse[7340]: partner active: deactivating lvs
Nov 21 10:53:11 node2 pulse[27219]: gratuitous lvs arps finished
Nov 21 16:25:53 node2 pulse[7340]: Terminating due to signal 15
Nov 21 16:25:53 node2 pulse: pulse shutdown succeeded
Nov 23 14:15:11 node2 pulse: pulse shutdown failed
Nov 23 14:15:11 node2 pulse: SIOCGIFADDR failed: Cannot assign requested address
Nov 23 14:15:11 node2 pulse[5552]: STARTING PULSE AS BACKUP
Nov 23 14:15:11 node2 pulse: pulse startup succeeded
Nov 13 20:07:17 node2 pulse[7340]: partner dead: activating lvs
Nov 13 20:07:20 node2 pulse[7340]: partner active: deactivating lvs
Nov 13 20:07:22 node2 pulse[2103]: gratuitous lvs arps finished
Oct 30 20:09:59 node2 pulse[7340]: partner dead: activating lvs
Oct 30 20:10:06 node2 pulse[24269]: gratuitous lvs arps finished
Oct 30 20:11:17 node2 pulse[7340]: partner active: deactivating lvs
Oct 30 20:39:47 node2 pulse[7340]: partner dead: activating lvs
Oct 30 20:39:48 node2 pulse[7340]: partner active: deactivating lvs
Oct 30 20:39:52 node2 pulse[25055]: gratuitous lvs arps finished
Nov  5 22:40:16 node2 pulse[7340]: partner dead: activating lvs
Nov  5 22:40:21 node2 pulse[26812]: gratuitous lvs arps finished

there failover occured whan node1 is aactive on the network

can some one help me to find why this happned

Last edited by robo; 11-23-2011 at 03:59 PM..

robo

View Public Profile for robo

Find all posts by robo

5 More Discussions You Might Find Interesting

1. Programming

Load Balancer

Halo mates, I m going to write a load balancer with C. Does anybody know some good reference on this? Cheers, Elton

2. AIX

hacmp ip load balancer failover

Hi All, How do I failover on the ip load balancer (back and forth)? It involves first to load a new config on the passive ip. If success, load the new config on the ip active (which is now passive). Any idea, please. Thanks in advance.

3. Red Hat

What is the Best Load Balancer for Linux?

Hi, What's the best load balancer for Linux (CentOS, SuSE) according to your personal experience? Linux Virtual Server (LVS) is a famous one, but their download site has not been updated since 2007. Their web and mailing list are so quiet. Is the Ultra Monkey project including LVS...

4. Linux

Server Load balancer

Hello Guys, Hope you all doing well . :) I was checking load balance command (uptime)on VM server and got below output. # uptime 07:08:40 up 52 min, 2 users,a load average: 0.45, 0.11, 0.03 A :How we can calculate load average? Thank you in advance !! Cheers:) Dont forget...

5. Virtualization and Cloud Computing

A load balancer for Nomachine NX

Hello, in case somebody has a NoMachine NX cluster, and is suffering from its dumb round-robin dispatcher, here is a solution: nxpub (NX Pluggable User Balancer). It should run on all LUnix OS. Scripts for install/uninstall are supplied. While tested with NX 3 (NX 3.5 is the latest), it might...

LEARN ABOUT DEBIAN

orte-clean

orte-clean(1)							     Open MPI							     orte-clean(1)

NAME

       orte-clean - Cleans up any stale processes and files leftover from Open MPI jobs.

SYNOPSIS

       orte-clean [--verbose]
       mpirun --pernode [--host | --hostfile file] orte-clean [--verbose]

OPTIONS

       [-v | --verbose] This argument will run the command in verbose mode and print out the universes that are getting cleaned up as well as pro-
       cesses that are being killed.

DESCRIPTION

       orte-clean attempts to clean up any processes and files left over from Open MPI jobs that were run in the past as  well	as  any  currently
       running jobs.  This includes OMPI infrastructure and helper commands, any processes that were spawned as part of the job, and any temporary
       files.  orte-clean will only act upon processes and files that belong to the user running the orte-clean command.  If run as root, it  will
       kill off processes belonging to any users.

       When run from the command line, orte-clean will attempt to clean up the local node it is run from.  When launched via mpirun, it will clean
       up the nodes selected by mpirun.

EXAMPLES

       Example 1: Clean up local node only.

       example% orte-clean

       Example 2: To clean up on a specific set of nodes specified on command line, it is recommended to use the pernode option.   This  will  run
       one orte-clean for each node.

       example% mpirun --pernode --host node1,node2,node3 orte-clean

       To clean up on a specific set of nodes from a file.

       example% mpirun --pernode --hostfile nodes_file orte-clean

       Example 3: Within a resource managed environment like N1GE, SLURM, or Torque.  The following example is from N1GE.

       First, we see that we have two nodes with two CPUs each.

       example% qsh -pe orte 4
       example% mpirun -np 4 hostname
       node1
       node1
       node2
       node2

       Clean up all the nodes in the cluster.

       example% mpirun --pernode orte-clean

       Clean up a subset of the nodes in the cluster.

       example% mpirun --pernode --host node1 orte-clean

SEE ALSO

       orterun(1), orte-ps(1)

1.4.5								   Feb 10, 2012 						     orte-clean(1)