[prev in list] [next in list] [prev in thread] [next in thread] 

List:       aix-l
Subject:    Re: HACMP problem/question
From:       Stan Vernaillen <stvernaillen () GE ! COKECCE ! COM>
Date:       2000-05-30 12:27:11
[Download RAW message or body]


Bruce,
you're right !
 tmssa had disappeared.
we reinstalled a node several weeks  ago, this probably caused it's
disappearence and following adapter swap problems
I recreated the ssa network and I'm tempted to believe this should be enough.

thanks again,
Stan Vernaillen
again optimistic HACMP-er




Bruce Zimmer <bzimmer@ALL-PHASE.COM> on 30/05/2000 06:30:24

Please respond to IBM AIX Discussion List <AIX-L@PUCC.PRINCETON.EDU>

To:   AIX-L@PUCC.PRINCETON.EDU
cc:    (bcc: Stan Vernaillen/BE/CCE)
Subject:  Re: HACMP problem/question




Stan,
    I am not sure what went on either, but for TM SSA there still should be
a serial network definition in HACMP.  I am not at my office now (attending
CA Unicenter class) but I will check when I get back as to what the
definition should be, because I didn't see one in the snapshot you sent.  TM
SSA may be set up, but HACMP doesn't appear to be using it, so it is doing
no good.

Bruce


-----Original Message-----
From: IBM AIX Discussion List [mailto:AIX-L@PUCC.PRINCETON.EDU]On Behalf
Of Stan Vernaillen
Sent: Monday, May 29, 2000 3:23 AM
To: AIX-L@PUCC.PRINCETON.EDU
Subject: Re: [AIX-L] HACMP problem/question


Bruce,


>>    During a later post you mentioned "I tried to recreate the problem but
>>this time everything went ok."  Do you mean that it no longer tries to
swap
>>adapters?
Yes, I brought it down , started it again, and everything went as it should,
no
swap at all, that's why I can not figure out what went wrong.

>>  I noticed that in the snapshot resource group rg2 the following was set
>>Highly Available Communication Links         ent0
This is for an SNA link to a mainframe, takeover tests for this one were
fine


>> Do you have a serial network?  None is configured, and this could be
>> part of the problem.
No, we use TM SSA

>>  Are these Network interfaces 10/100 and if so are they fixed at 100MB
>>Full?  There have been lots of problems with  10/100 auto-configured
>>interfaces.
Yes, they are fixed at 100MB Full duplex

Stan Vernaillen
Still Clueless  :)





"Bruce R. Zimmer" <isdfgbrz@ALL-PHASE.COM> on 26/05/2000 18:15:45

Please respond to IBM AIX Discussion List <AIX-L@PUCC.PRINCETON.EDU>

To:   AIX-L@PUCC.PRINCETON.EDU
cc:    (bcc: Stan Vernaillen/BE/CCE)
Subject:  Re: HACMP problem/question




I have a few more questions

    During a later post you mentioned "I tried to recreate the problem but
this time everything went ok."  Do you mean that it no longer tries to swap
adapters?

    I noticed that in the snapshot resource group rg2 the following was set

Highly Available Communication Links         ent0

    This is the interface that the boot and service adapters share ( I
realize that rg2 is not trying to come up, but this is a new feature in
4.3.1 and is associated with CS/AIX and I don't see any other config info
for CS/AIX )

    Do you have a serial network?  None is configured, and this could be
part of the problem.

    Are these Network interfaces 10/100 and if so are they fixed at 100MB
Full?  There have been lots of problems with  10/100 auto-configured
interfaces.

HTH
Bruce Zimmer





-----Original Message-----
From: IBM AIX Discussion List [mailto:AIX-L@PUCC.PRINCETON.EDU]On Behalf
Of Stan Vernaillen
Sent: Thursday, May 25, 2000 2:47 AM
To: AIX-L@PUCC.PRINCETON.EDU
Subject: Re: [AIX-L] HACMP problem/question





Here's a snapshot.
there was no real reconfiguration going n. added some logical volumes etc,
but
that was it.

(See attached file: gb6ecf01.000524.odm)(See attached file:
gb6ecf01.000524.info)




"Bruce R. Zimmer" <isdfgbrz@ALL-PHASE.COM> on 24/05/2000 16:54:05

Please respond to IBM AIX Discussion List <AIX-L@PUCC.PRINCETON.EDU>

To:   AIX-L@PUCC.PRINCETON.EDU
cc:    (bcc: Stan Vernaillen/BE/CCE)
Subject:  Re: HACMP problem/question



This looks like a configuration error.  Has the system been up and running,
or is this a new config or resource group?  Have there been any recent
changes?  Would you be able to attach a snapshot of the config for us to
peruse?

Bruce Zimmer


-----Original Message-----
From: IBM AIX Discussion List [mailto:AIX-L@PUCC.PRINCETON.EDU]On Behalf
Of Stan Vernaillen
Sent: Wednesday, May 24, 2000 10:22 AM
To: AIX-L@PUCC.PRINCETON.EDU
Subject: [AIX-L] HACMP problem/question


Hi all,

Can someone help explain what happened......

we have a 2 node cluster.
HACMP was not running on either node.I started HACMP on the first
node.Everything starts ok, he changes from boot address to service,
gives a NODE_UP_COMPLETE .
So far so good.
But then it tries a swap_adapter between Boot and StandBy??.of course since
Boot
is now service he gives an error that he can not locate the boot interface
and
goes in the ever popular config_too_long.....

any ideas?

Stan Vernaillen
Professional HACMP-problem-having-guy.



10.10.31.188   is the standby address or gb6ecf01s
167.105.146.189   is the boot address or gb6ecf01b
167.105.146.188 is the service addres or gb6ecf01


(/var/adm/cluster.log)


May 24 09:53:38 gb6ecf01 clstrmgr[22480]: CLUSTER MANAGER STARTED
May 24 09:53:48 gb6ecf01 clinfo[23482]: send_snmp_req: Messages in queue got
= 4
read = 1
May 24 09:53:50 gb6ecf01 HACMP for AIX: EVENT START: node_up gb6ecf01
May 24 09:53:50 gb6ecf01 HACMP for AIX: EVENT START: node_up_local
May 24 09:53:51 gb6ecf01 HACMP for AIX: EVENT START: acquire_service_addr
gb6ecf01
May 24 09:53:59 gb6ecf01 HACMP for AIX: EVENT START: acquire_aconn_service
en0
ether146
May 24 09:54:34 gb6ecf01 HACMP for AIX: EVENT START: swap_aconn_protocols
en0
en2
May 24 09:54:34 gb6ecf01 HACMP for AIX: EVENT COMPLETED:
swap_aconn_protocols
en0 en2
May 24 09:54:35 gb6ecf01 HACMP for AIX: EVENT COMPLETED:
acquire_aconn_service
en0 ether146
May 24 09:55:20 gb6ecf01 HACMP for AIX: EVENT COMPLETED:
acquire_service_addr
gb6ecf01
May 24 09:55:20 gb6ecf01 HACMP for AIX: EVENT START: get_disk_vg_fs /oracle
May 24 09:55:35 gb6ecf01 HACMP for AIX: EVENT COMPLETED: get_disk_vg_fs
May 24 09:55:35 gb6ecf01 HACMP for AIX: EVENT COMPLETED: node_up_local
May 24 09:55:35 gb6ecf01 HACMP for AIX: EVENT COMPLETED: node_up gb6ecf01
May 24 09:55:36 gb6ecf01 HACMP for AIX: EVENT START: node_up_complete
gb6ecf01
May 24 09:55:36 gb6ecf01 HACMP for AIX: EVENT START: node_up_local_complete
May 24 09:55:36 gb6ecf01 HACMP for AIX: EVENT START: start_server remedy
maximo
May 24 09:55:36 gb6ecf01 HACMP for AIX: EVENT COMPLETED: start_server remedy
maximo
May 24 09:55:37 gb6ecf01 HACMP for AIX: EVENT COMPLETED:
node_up_local_complete

May 24 09:55:37 gb6ecf01 HACMP for AIX: EVENT COMPLETED: node_up_complete
gb6ecf01

May 24 09:55:38 gb6ecf01 HACMP for AIX: EVENT START: swap_adapter gb6ecf01
ether146 10.10.31.188 167.105.146.189
May 24 09:55:38 gb6ecf01 HACMP for AIX: Interface for 167.105.146.189 is not
found.
May 24 09:55:38 gb6ecf01 HACMP for AIX: EVENT FAILED:1: swap_adapter
gb6ecf01
ether146 10.10.31.188 167.105.146.189
May 24 09:55:39 gb6ecf01 clstrmgr[22480]: gb6ecf01: bad script status 1 for
gb6ecf01
May 24 09:55:39 gb6ecf01 HACMP for AIX: EVENT START: event_error gb6ecf01
swap_adapter gb6ecf01 gb6ecf01b
May 24 09:55:39 gb6ecf01 HACMP for AIX: EVENT COMPLETED: event_error
gb6ecf01
swap_adapter gb6ecf01 gb6ecf01b

[prev in list] [next in list] [prev in thread] [next in thread]