Customer environment: aix 5.2-06 , hacmp 5.1and HACMP APAR
IY66556；customer 's application via NFS，so there are no share disk
and vg/fs needed to configure,so I only configure
cluster、node、resoures（service ip and emulation app1、app2
relation to each mode; /ha5l/1start.sh, /ha5l/1stop.sh is for
app1；/ha5l/2start.sh, /ha5l/2stop.sh is for app2)
now I can start/stop hacmp on each node and ip address is normal, nodes
can take over if I remove app1/app2 definition, but will hang if I add
app1/app2 definition to the resourse group,process clstrmgrES is
persist stopping.Following is the scripts and error messages.


vi 1start
date >> /ha51/1start.log
banner " start app1 " >> /tmp/hacmp.out
vi 1stop
date >> /ha51/1stop.log
banner "stop app1 " >> /tmp/hacmp.out
vi 2start
date >> /ha51/2start.log
banner " start app2 " >> /tmp/hacmp.out
vi 2stop
date >> /ha51/2stop.log
banner "stop app2 " >> /tmp/hacmp.out


error message is as follow：
node_down[264] [[ trade1 = trade1 ]]
:node_down[268] :node_down[268] lsvg -L
:node_down[268] lsvg -L -o
:node_down[268] paste -s -d| -
:node_down[268] grep -v -E rootvg
INACTIVE_VGS=
:node_down[281] [[ trade1 = trade1 ]]
:node_down[281] [[ REAL = EMUL ]]
:node_down[286] rm -f /usr/es/sbin/cluster/etc/comm_link.lck
:node_down[293] [[ trade1 = trade1 ]]
:node_down[293] [[ REAL = EMUL ]]
:node_down[299] clchdaemons -r -d clstrmgr_scripts -t resource_locator
:node_down[306] exit 1
May 25 16:49:48 EVENT FAILED: 1: node_down trade1 1

HACMP Event Summary
Event: node_down trade1
Start time: Wed May 25 16:49:28 2005

End time: Wed May 25 16:49:48 2005

Action: Resource: Script Name:
----------------------------------------------------------------------------

Releasing resource group: trade1_es process_resources
Search on:
Wed.May.25.16:49:30.BEIST.2005.process_resources.t rade1_es.ref
Releasing resource: All_servers stop_server
Search on:
Wed.May.25.16:49:32.BEIST.2005.stop_server.All_ser vers.trade1_es.ref
Error encountered with resource: app1 stop_server
Search on:
Wed.May.25.16:49:33.BEIST.2005.stop_server.app1.tr ade1_es.ref
Resource offline: All_nonerror_servers stop_server
Search on:
Wed.May.25.16:49:34.BEIST.2005.stop_server.All_non error_servers.trade1_es.ref

Error encountered with resource: All_servers process_resources
Search on:
Wed.May.25.16:49:35.BEIST.2005.process_resources.A ll_servers.trade1_es.ref

Releasing resource: All_service_addrs release_service_addr
Search on:
Wed.May.25.16:49:37.BEIST.2005.release_service_add r.All_service_addrs.trade1_es.ref

Resource offline: All_nonerror_service_addrs release_service_addr
Search on:
Wed.May.25.16:49:44.BEIST.2005.release_service_add r.All_nonerror_service_addrs.trade1_es.ref

Error encountered with group: trade1_es process_resources
Search on:
Wed.May.25.16:49:46.BEIST.2005.process_resources.t rade1_es.ref
----------------------------------------------------------------------------

May 25 16:49:48 EVENT START: event_error 1 1_node_down trade1 _2

:event_error[52] [[ high = high ]]
:event_error[52] version=1.10
:event_error[53] :event_error[53] cl_get_path
HA_DIR=es
:event_error[55] EXIT_STATUS=1
:event_error[56] RP_NAME=1 1_node_down trade1 _2
:event_error[59] [ 2 -ne 2 ]
:event_error[65] set -u
:event_error[67] RP_NAME=node_down trade1 _2
:event_error[68] RP_NAME=node_down trade1
:event_error[70] :event_error[70] cllsclstr -c
:event_error[70] grep -v cname
:event_error[70] cut -d : -f2
CLUSTER=hft_trade
:event_error[74] [ -x /usr/lpp/ssp/bin/spget_syspar ]
:event_error[81] echo WARNING: Cluster hft_trade Failed while running
node_down trade1 , exit status was 1
:event_error[81] 1> /dev/console
:event_error[82] echo WARNING: Cluster hft_trade Failed while running
node_down trade1 , exit status was 1
WARNING: Cluster hft_trade Failed while running node_down trade1 , exit
status was 1
:event_error[88] [[ node_down trade1 = reconfig_resource* ]]
May 25 16:49:48 EVENT FAILED: -1: event_error 1 1_node_down trade1 _2
-1

andrewqiu@163.com wrote:
> Customer environment: aix 5.2-06 , hacmp 5.1and HACMP APAR
> IY66556；customer 's application via NFS，so there are no share disk
> and vg/fs needed to configure,so I only configure
> cluster、node、resoures（service ip and emulation app1 、app2
> relation to each mode; /ha5l/1start.sh, /ha5l/1stop.sh is for
> app1；/ha5l/2start.sh, /ha5l/2stop.sh is for app2)
> now I can start/stop hacmp on each node and ip address is normal, nodes
> can take over if I remove app1/app2 definition, but will hang if I add
> app1/app2 definition to the resourse group,process clstrmgrES is
> persist stopping.Following is the scripts and error messages.
>
>
> vi 1start
> date >> /ha51/1start.log
> banner " start app1 " >> /tmp/hacmp.out
> vi 1stop
> date >> /ha51/1stop.log
> banner "stop app1 " >> /tmp/hacmp.out
> vi 2start
> date >> /ha51/2start.log
> banner " start app2 " >> /tmp/hacmp.out
> vi 2stop
> date >> /ha51/2stop.log
> banner "stop app2 " >> /tmp/hacmp.out
>
>
> error message is as follow：
> node_down[264] [[ trade1 = trade1 ]]
> :node_down[268] :node_down[268] lsvg -L
> :node_down[268] lsvg -L -o
> :node_down[268] paste -s -d| -
> :node_down[268] grep -v -E rootvg
> INACTIVE_VGS=
> :node_down[281] [[ trade1 = trade1 ]]
> :node_down[281] [[ REAL = EMUL ]]
> :node_down[286] rm -f /usr/es/sbin/cluster/etc/comm_link.lck
> :node_down[293] [[ trade1 = trade1 ]]
> :node_down[293] [[ REAL = EMUL ]]
> :node_down[299] clchdaemons -r -d clstrmgr_scripts -t resource_locator
> :node_down[306] exit 1
> May 25 16:49:48 EVENT FAILED: 1: node_down trade1 1

<snip>

There's not enough of the /tmp/hacmp.out here to see what is actually
not working. You need to trace back until you find the first non-zero
return code in here.

As a side-issue, I think the redirects are redundant since HACMP may
re-direct script output to the log file. Note that HACMP does certain
things to pre-process stdout to get it linear since some scripts are run
in parallel.

I have solve this problme after modify the scripts.

thanks