HPUX 11V3 HPVM 4.3 and ServiceGuard 11.20 can not execute hpvmsg_package consistency error

[vandevoort]

Hi ,

tried to setup HPVM Guest as SG PACKAGE.

using hpvmsg_package -P testguest :

This is the HP Virtual Machine Serviceguard Toolkit package creation

script.

This script will assist the user to develop and manage a set of

Serviceguard package configuration and associated toolkit template

files.

The templates generated by these scripts will handle many guest

configurations, but they are only templates and may not be

appropriate for your particular configuration needs. You are

encouraged to review and modify these template files, as needed,

for your particular environment.

Do you wish to continue? (y/n):y

[Virtual Machine Details]

Virtual Machine Name VM #  OS Type State

==================== ===== ======= ========

testguest                   5 HPUX    On (EFI)

[Storage Interface Details]

Guest                                 Physical

Device  Adaptor    Bus Dev Ftn Tgt Lun Storage   Device

======= ========== === === === === === ========= =========================

disk    avio_stor    0   0   0   0   0 disk      /dev/rdisk/disk260

[Network Interface Details]

Interface Adaptor    Name/Num   PortNum Bus Dev Ftn Mac Address

========= ========== ========== ======= === === === =================

Package the Virtual Machine Summarized above? (y/n):y

Checking the virtual machine and cluster configuration

Analysis of the hpvmstatus -m command on each node indicates that the

HPVM multi-server environment is not configured correctly.  Before

hpvmsg_package can assist in creating the package configuration template files

the HPVM multi-server environment must be configured.

Configure Serviceguard nodes into the HPVM multi-server environment now? (y/n):y

ERROR : hpvmsg_package : The multi-server environment is not configured consistently on each node

 ssh ahp009 /opt/hpvm/bin/hpvmdevmgmt -l env
HPVM_MSE_GROUP_ENTRY:CONFIG=env,EXIST=NO,DEVTYPE=UNKNOWN,SHARE=NO,UUID=df118256-8d35-11e0-a00c-0017a4779c0c,GROUPNAME=HPVM-SG-cl_iris1::WWID_NULL
root@ahp007:/opt/cmcluster/toolkit/hpvm
#  /opt/hpvm/bin/hpvmdevmgmt -l env
HPVM_MSE_GROUP_ENTRY:CONFIG=env,EXIST=NO,DEVTYPE=UNKNOWN,SHARE=NO,UUID=df118256-8d35-11e0-a00c-0017a4779c0c,GROUPNAME=HPVM-SG-cl_iris1::WWID_NULL
root@ahp007:/opt/cmcluster/toolkit/hpvm
#  /opt/hpvm/bin/hpvmdevmgmt -l server
ahp007:CONFIG=server,SERVERADDR=141.93.115.20,SERVERID=007,Server_physical_uuid=cbd59d59-1557-4277-92e9-c39055454107::WWID_NULL
ahp009:CONFIG=server,SERVERADDR=141.93.115.23,SERVERID=009,Server_physical_uuid=fc3d6504-47b3-4f69-8f59-692e2f9ecc1d::WWID_NULL
root@ahp007:/opt/cmcluster/toolkit/hpvm
# ssh ahp009 /opt/hpvm/bin/hpvmdevmgmt -l server
ahp007:CONFIG=server,SERVERADDR=141.93.115.20,SERVERID=007,Server_physical_uuid=cbd59d59-1557-4277-92e9-c39055454107::WWID_NULL
ahp009:CONFIG=server,SERVERADDR=141.93.115.23,SERVERID=009,Server_physical_uuid=fc3d6504-47b3-4f69-8f59-692e2f9ecc1d::WWID_NULL
root@ahp007:/opt/cmcluster/toolkit/hpvm
# hpvmstatus -m
[HPVM Multi-server environment]

        Server group name       : HPVM-SG-cl_iris1
        Server group uuid       : df118256-8d35-11e0-a00c-0017a4779c0c
        This server's identifier: 7

        Server_id               : 7
        Server_state            : 1
        Server_ipaddr           : 141.93.115.20
        Server_hostname         : ahp007
        Server_uuid             :
        Server_physical_uuid    :

        Server_id               : 9
        Server_state            : 1
        Server_ipaddr           : 141.93.115.23
        Server_hostname         : ahp009
        Server_uuid             :
        Server_physical_uuid    :

root@ahp007:/opt/cmcluster/toolkit/hpvm
# ssh ahp009 /opt/hpvm/bin/hpvmstatus -m
[HPVM Multi-server environment]

        Server group name       : HPVM-SG-cl_iris1
        Server group uuid       : df118256-8d35-11e0-a00c-0017a4779c0c
        This server's identifier: 9

        Server_id               : 7
        Server_state            : 1
        Server_ipaddr           : 141.93.115.20
        Server_hostname         : ahp007
        Server_uuid             :
        Server_physical_uuid    :

        Server_id               : 9
        Server_state            : 1
        Server_ipaddr           : 141.93.115.23
        Server_hostname         : ahp009
        Server_uuid             :
        Server_physical_uuid    :

root@ahp007:/opt/cmcluster/toolkit/hpvm

does someone has any idea what's wrong ??

thanks

Wim van de Voort

[Steven E. Protter]

Shalom,

Please post the link to the cookbook you are using. Check that you are following it correctly.

After that, I'd say you might want to contact HP and see if you have uncovered a bug.

SEP

[vandevoort]

Hi Steven

we do not use a cookbook .

only from the HP Integrity Virtual Machines 4.3: Installation, Configuration, and Administration Manual chapter 10.4 .

only we created the Hosts and Guest a while ago  they are up and running now we want to install Service Guesrd and set these up .

kind regards

Wim van de voort.

[Eric SAUBIGNAC]

Bonjour,

Not sure but strange things in uuid. 2 points :

- Where does come 'Server_physical_uuid' in output of hpvmdevmgmt command ? As far as i know It should be libelled PHYSUUID. (it was in HPVM 4.20)

- Fore each server, you miss the UUID wich is the same value than the PHYSUUID

An other thing is that hpvmstatus -m doesn't show any server's ID. I guess that it's due to this "'Server_physical_uuid' " that doesn't mean anything with hpvmdevmgmt

I did a post in the good old forum on a similar problem; I can't find it in the new forum grrrr ... But here is the google cached version : http://webcache.googleusercontent.com/search?q=cache:fmfI2UO2pEEJ:forums11.itrc.hp.com/service/forums/questionanswer.do%3FthreadId%3D1448014+hpvmdevmgmt+uuid+Eric+SAUBIGNAC&cd=1&hl=en&ct=clnk&source=www.google.com

Eric

[vandevoort]

Hi I added  Server_uuid

hpvmdevmgmt -m server:ahp009:attr:Server_uuid=fc3d6504-47b3-4f69-8f59-692e2f9ecc1d

and

hpvmdevmgmt -m server:ahp007:attr:Server_uuid=cbd59d59-1557-4277-92e9-c39055454107

on both systems :

 No change :

from 1st Guest ahp007:

hpvmdevmgmt -l env -V; hpvmdevmgmt -l server
HPVM_MSE_GROUP_ENTRY:CONFIG=env,EXIST=NO,DEVTYPE=UNKNOWN,SHARE=NO,UUID=df118256-8d35-11e0-a00c-0017a4779c0c,GROUPNAME=HPVM-SG-cl_iris1::WWID_NULL
ahp007:CONFIG=server,SERVERADDR=141.93.115.20,SERVERID=007,Server_physical_uuid=cbd59d59-1557-4277-92e9-c39055454107,Server_uuid=cbd59d59-1557-4277-92e9-c39055454107::WWID_NULL
ahp009:CONFIG=server,SERVERADDR=141.93.115.23,SERVERID=009,Server_physical_uuid=fc3d6504-47b3-4f69-8f59-692e2f9ecc1d,Server_uuid=fc3d6504-47b3-4f69-8f59-692e2f9ecc1d::WWID_NULL
root@ahp007:/
# hpvmstatus -m
[HPVM Multi-server environment]

        Server group name       : HPVM-SG-cl_iris1
        Server group uuid       : df118256-8d35-11e0-a00c-0017a4779c0c
        This server's identifier: 7

        Server_id               : 7
        Server_state            : 1
        Server_ipaddr           : 141.93.115.20
        Server_hostname         : ahp007
        Server_uuid             :
        Server_physical_uuid    :

        Server_id               : 9
        Server_state            : 1
        Server_ipaddr           : 141.93.115.23
        Server_hostname         : ahp009
        Server_uuid             :
        Server_physical_uuid    :

seconf Guest ( ahp009)

  hpvmdevmgmt -l env -V; hpvmdevmgmt -l server
HPVM_MSE_GROUP_ENTRY:CONFIG=env,EXIST=NO,DEVTYPE=UNKNOWN,SHARE=NO,UUID=df118256-8d35-11e0-a00c-0017a4779c0c,GROUPNAME=HPVM-SG-cl_iris1::WWID_NULL
ahp007:CONFIG=server,SERVERADDR=141.93.115.20,SERVERID=007,Server_physical_uuid=cbd59d59-1557-4277-92e9-c39055454107,Server_uuid=cbd59d59-1557-4277-92e9-c39055454107::WWID_NULL
ahp009:CONFIG=server,SERVERADDR=141.93.115.23,SERVERID=009,Server_physical_uuid=fc3d6504-47b3-4f69-8f59-692e2f9ecc1d,Server_uuid=fc3d6504-47b3-4f69-8f59-692e2f9ecc1d::WWID_NULL
root@ahp009:/
# hpvmstatus -m
[HPVM Multi-server environment]

        Server group name       : HPVM-SG-cl_iris1
        Server group uuid       : df118256-8d35-11e0-a00c-0017a4779c0c
        This server's identifier: 9

        Server_id               : 7
        Server_state            : 1
        Server_ipaddr           : 141.93.115.20
        Server_hostname         : ahp007
        Server_uuid             :
        Server_physical_uuid    :

        Server_id               : 9
        Server_state            : 1
        Server_ipaddr           : 141.93.115.23
        Server_hostname         : ahp009
        Server_uuid             :
        Server_physical_uuid   

I wonder why the Server_UUID and Server_physical_uuid in hpvmstatus -m is not filled in ??

But anyway :

still the same problem :

# /opt/cmcluster/toolkit/hpvm/hpvmsg_package -s -P tgast1

This is the HP Virtual Machine Serviceguard Toolkit package creation
script.

This script will assist the user to develop and manage a set of
Serviceguard package configuration and associated toolkit template
files.

The templates generated by these scripts will handle many guest
configurations, but they are only templates and may not be
appropriate for your particular configuration needs. You are
encouraged to review and modify these template files, as needed,
for your particular environment.

Do you wish to continue? (y/n):y

[Virtual Machine Details]
Virtual Machine Name VM #  OS Type State
==================== ===== ======= ========
tgast1                   5 HPUX    On (EFI)
[Storage Interface Details]
Guest                                 Physical
Device  Adaptor    Bus Dev Ftn Tgt Lun Storage   Device
======= ========== === === === === === ========= =========================
disk    avio_stor    0   0   0   0   0 disk      /dev/rdisk/disk260
[Network Interface Details]
Interface Adaptor    Name/Num   PortNum Bus Dev Ftn Mac Address
========= ========== ========== ======= === === === =================

Package the Virtual Machine Summarized above? (y/n):y

Checking the virtual machine and cluster configuration

Analysis of the hpvmstatus -m command on each node indicates that the
HPVM multi-server environment is not configured correctly.  Before
hpvmsg_package can assist in creating the package configuration template files
the HPVM multi-server environment must be configured.

ERROR : hpvmsg_package : The multi-server environment is not configured consistently on each node

 even checked  Both output with diff and chksum they are exact the same on both systems ..

does anyone has any Idea ??

Thanks ..

[vandevoort]

Hi .

I did not know the exact parameters :

changed the settings on both systems :

hpvmdevmgmt -d server:ahp007:attr:Server_uuid
hpvmdevmgmt -d server:ahp009:attr:Server_uuid
hpvmdevmgmt -d server:ahp007:attr:Server_physical_uuid
hpvmdevmgmt -d server:ahp009:attr:Server_physical_uuid

and added PHYSUUID

 hpvmdevmgmt -m server:ahp007:attr: PHYSUUID=cbd59d59-1557-4277-92e9-c39055454107

and

hpvmdevmgmt -m server:ahp009:attr: PHYSUUID=fc3d6504-47b3-4f69-8f59-692e2f9ecc1d

still the same problem ..

any ideas ??

[Bhadresh]

Hi,

Can you please provide following output from both the nodes:

#hpvmstatus -m

Regards,

Bhadresh

[vandevoort]

Hi :

I added UID with hpvmdevmgmt  command :

# hpvmdevmgmt -m server:ahp007:attr:UUID=df118256-8d35-11e0-a00c-0017a4779c0c
# hpvmdevmgmt -m server:ahp009:attr:UUID=fc3d6504-47b3-4f69-8f59-692e2f9ecc1d
On Both Nodes:

hpvmstatus -m  is now  completely filled and shows like this :

# hpvmstatus -m
[HPVM Multi-server environment]

        Server group name       : HPVM-SG-cl_iris1
        Server group uuid       : df118256-8d35-11e0-a00c-0017a4779c0c
        This server's identifier: 1

        Server_id               : 1
        Server_state            : 1
        Server_ipaddr           : 141.93.115.20
        Server_hostname         : ahp007
        Server_uuid             : df118256-8d35-11e0-a00c-0017a4779c0c
        Server_physical_uuid    : df118256-8d35-11e0-a00c-0017a4779c0c

        Server_id               : 2
        Server_state            : 1
        Server_ipaddr           : 141.93.115.23
        Server_hostname         : ahp009
        Server_uuid             : fc3d6504-47b3-4f69-8f59-692e2f9ecc1d
        Server_physical_uuid    : fc3d6504-47b3-4f69-8f59-692e2f9ecc1d

On the Other system :

# hpvmstatus -m
[HPVM Multi-server environment]

        Server group name       : HPVM-SG-cl_iris1
        Server group uuid       : df118256-8d35-11e0-a00c-0017a4779c0c
        This server's identifier: 2

        Server_id               : 1
        Server_state            : 1
        Server_ipaddr           : 141.93.115.20
        Server_hostname         : ahp007
        Server_uuid             : df118256-8d35-11e0-a00c-0017a4779c0c
        Server_physical_uuid    : df118256-8d35-11e0-a00c-0017a4779c0c

        Server_id               : 2
        Server_state            : 1
        Server_ipaddr           : 141.93.115.23
        Server_hostname         : ahp009
        Server_uuid             : fc3d6504-47b3-4f69-8f59-692e2f9ecc1d
        Server_physical_uuid    : fc3d6504-47b3-4f69-8f59-692e2f9ecc1d

created a simple Guest  with hpvmcreate -P testaap -O hpux :

# hpvmsg_package -P testaap

This is the HP Virtual Machine Serviceguard Toolkit package creation
script.

This script will assist the user to develop and manage a set of
Serviceguard package configuration and associated toolkit template
files.

The templates generated by these scripts will handle many guest
configurations, but they are only templates and may not be
appropriate for your particular configuration needs. You are
encouraged to review and modify these template files, as needed,
for your particular environment.

Do you wish to continue? (y/n):y

[Virtual Machine Details]
Virtual Machine Name VM #  OS Type State
==================== ===== ======= ========
testaap                  9 HPUX    Off
[Storage Interface Details]
Guest                                 Physical
Device  Adaptor    Bus Dev Ftn Tgt Lun Storage   Device
======= ========== === === === === === ========= =========================
[Network Interface Details]
Interface Adaptor    Name/Num   PortNum Bus Dev Ftn Mac Address
========= ========== ========== ======= === === === =================

Package the Virtual Machine Summarized above? (y/n):y

Checking the virtual machine and cluster configuration

Analysis of the hpvmstatus -m command on each node indicates that the
HPVM multi-server environment is not configured correctly.  Before
hpvmsg_package can assist in creating the package configuration template files
the HPVM multi-server environment must be configured.

Configure Serviceguard nodes into the HPVM multi-server environment now? (y/n):y

ERROR : hpvmsg_package : The multi-server environment is not configured consistently on each node

still the same error ..

[StanM]

When adding server entry uuid using hpvmdevmgmt pls use UUID and PHYSUUID as attribute names

(not 'server_uuid' ...).

You can get their values from hpvminfo -S.

Make sure to configure all these values on both sides of the cluster so that hpvmstatus -m shows

the same values for cluster and for both servers on both nodes.

[Eric SAUBIGNAC]

Bonjour,

Sorry for this late answer : I had some days off last week.

Some points :

- In one of your posts, servers'id were 7 and 9. In the latest post they are now 1 and 2.  Obviously you have done more changes than those announced in your posts. Could you post the latest hpvmdevmgmt -l server and -l env ?

- Did you tried hpvmsg_package with the "sanity check" option : -s . When used, the script will try to correct the multi-environment if necessary ... and possible ;-)

- Here are the checks done by the script. (extract from HPVM 4.20, not sure they are exactly the same in 4.30)  :

 # 1. Only the SG nodes are configured
 # 2. They are all configured the same
 # 3. Each node entry includes all name/value pairs:  PHYSUUID SERVERADDR SERVERID UUID

So, could you have an issue at the cluster side ? For example, more nodes than those configured in the multi_environment ... Also post output of cmviewcl.

Eric

[vandevoort]

hi Eric .

we did  clean up the config:

hpvmdevmgmt -d server:ahp008:attr:UUID
hpvmdevmgmt -d server:ahp008:attr:PHYSUUID
hpvmdevmgmt -d server:ahp010:attr:PHYSUUID
hpvmdevmgmt -d server:ahp010:attr:UUID
hpvmdevmgmt -m server:ahp008:attr:UUID=b3a1138c-1eba-4f18-8035-408f17f13528
hpvmdevmgmt -m server:ahp008:attr:PHYSUUID=1b5bb8a6-71a1-11e0-b5ea-4d6b12350ba1
hpvmdevmgmt -m server:ahp010:attr:UUID=1882ec5c-f79c-41d0-aa83-f46b2363c44c
hpvmdevmgmt -m server:ahp010:attr:PHYSUUID=0e35e3ca-71a1-11e0-8c57-4f841a320a9d

on both environments hpvmsg_package works now . without any error  messages .

onfly on one two node cluster we do not get the hpvmstatus -m info :

# hpvmstatus -m
No HPVM multi-server environment configured

on both Nodes !!!

One node  ahp008:

root@ahp008.ah.nl:/
# hpvmdevmgmt -l env
HPVM_MSE_GROUP_ENTRY:CONFIG=env,EXIST=NO,DEVTYPE=UNKNOWN,SHARE=NO,UUID=f11abdfe-aee7-11e0-a301-0017a4778c10,GROUPNAME=HPVM-SG-ahp010_cluster::WWID_NULL
root@ahp008.ah.nl:/
# hpvmdevmgmt -l server
ahp008:CONFIG=server,SERVERID=8,SERVERADDR=141.93.115.26,UUID=b3a1138c-1eba-4f18-8035-408f17f13528,PHYSUUID=1b5bb8a6-71a1-11e0-b5ea-4d6b12350ba1::WWID_NULL
ahp010:CONFIG=server,SERVERID=10,SERVERADDR=141.93.115.29,UUID=1882ec5c-f79c-41d0-aa83-f46b2363c44c,PHYSUUID=0e35e3ca-71a1-11e0-8c57-4f841a320a9d::WWID_NULL

Second node ahp0010

# hpvmdevmgmt -l env
HPVM_MSE_GROUP_ENTRY:CONFIG=env,EXIST=NO,DEVTYPE=UNKNOWN,SHARE=NO,UUID=f11abdfe-aee7-11e0-a301-0017a4778c10,GROUPNAME=HPVM-SG-ahp010_cluster::WWID_NULL
root@ahp010.ah.nl:/
# hpvmdevmgmt -l server
ahp008:CONFIG=server,SERVERID=8,SERVERADDR=141.93.115.26,UUID=b3a1138c-1eba-4f18-8035-408f17f13528,PHYSUUID=1b5bb8a6-71a1-11e0-b5ea-4d6b12350ba1::WWID_NULL
ahp010:CONFIG=server,SERVERID=10,SERVERADDR=141.93.115.29,UUID=1882ec5c-f79c-41d0-aa83-f46b2363c44c,PHYSUUID=0e35e3ca-71a1-11e0-8c57-4f841a320a9d::WWID_NULL
root@ahp010.ah.nl:/

we did the same on the other Cluster environment ahp007 and ahp010

this all works fine now  ??

[Eric SAUBIGNAC]

OK,

It's now OK with nodes ahp007 and ahp009 in cluster cl_iris1 . Right ?

With nodes ahp008 and ahp010 in cluster ahp010_cluster it doesn't work. Right ?

Some remarks :

- for both server, UUID has not the same value than PHYSUUID. For example, ahp008 has UUID b3a1138c-1eba-4f18-8035-408f17f13528 and PHYSUUID 1b5bb8a6-71a1-11e0-b5ea-4d6b12350ba1. Which one is the good one ? It should be the same value than output of #getconf CS_MACHINE_IDENT

- is the name of the MC/ServiceGuard really ahp010_cluster ? Is this cluster up and running ?

- well, could be consider as a  supersition, but SERVERID are 8 and 10. Never tried values other than 1, 2, 3, ... but programmers sometimes have stupid ideas. So maybe it doesn't work if you don't start server numbering at 1 ?

Eric

[vandevoort]

Hi Eric , 

It's now OK with nodes ahp007 and ahp009 in cluster cl_iris1 . Right ? yes all looks ok now for this Cluster

With nodes ahp008 and ahp010 in cluster ahp010_cluster it doesn't work. Right ? also I can set the nodes into ServiceGuard without any problems .

Some remarks :

- for both server, UUID has not the same value than PHYSUUID. For example, ahp008 has UUID b3a1138c-1eba-4f18-8035-408f17f13528 and PHYSUUID 1b5bb8a6-71a1-11e0-b5ea-4d6b12350ba1. Which one is the good one ? It should be the same value than output of #getconf CS_MACHINE_IDENT

Used machinfo :

and hpinfo -S:

HPVM Server information
Version: HPVM B.04.30 ccipf opt Tue Jan 11 2011 08h48m13s PST
My partition ident: 1882ec5c-f79c-41d0-aa83-f46b2363c44c
My physical ident: 0e35e3ca-71a1-11e0-8c57-4f841a320a9d

see a differetn in PHYSUUID and UUID.

changed bot to the same address shown with machinfo , on both systems no results ..

- is the name of the MC/ServiceGuard really ahp010_cluster ? Is this cluster up and running ? Yes it's up and running, and configured now with 4 Guests .

# cmviewcl

CLUSTER          STATUS
ahp010_cluster   up

  NODE           STATUS       STATE
  ahp008         up           running

    PACKAGE        STATUS           STATE            AUTO_RUN    NODE
    ahv008         up               running          enabled     ahp008
    ahv010         up               running          enabled     ahp008

  NODE           STATUS       STATE
  ahp010         up           running

    PACKAGE        STATUS           STATE            AUTO_RUN    NODE
    ahv012         up               running          enabled     ahp010
    ahv014         up               running          enabled     ahp010

- well, could be consider as a  supersition, but SERVERID are 8 and 10. Never tried values other than 1, 2, 3, ... but programmers sometimes have stupid ideas. So maybe it doesn't work if you don't start server numbering at 1 ?

we caled the systems ID same as the Server ahp009 = SererID=9 , wil be created more server systems in the Future ..  als ahp001  etc ..

Kind regards

Eric

[Eric SAUBIGNAC]

"see a differetn in PHYSUUID and UUID."

OK, but are ahp008 and ahp010 the same hardware than ahp007 and ahp009 ? If yes, we can see in your post of Friday that UUID = PHYSUUID for ahp007 and ahp009 . So why does it work properly with ahp007 and ahap009 ?

"we caled the systems ID same as the Server ahp009 = SererID=9 , wil be created more server systems in the Future ..  als ahp001  etc .."

I can understand that, but suppose one moment, that in a given HPVM_MSE_GROUP, the servers must be numbered from 1 ?

Also note that in your same post of Friday, we can see that server instanciation begins at 1, though the concerned servers are ahp007 and ahp009.

[vandevoort]

Hi Eric .

Like I told already , we cleared the whole configurationn and started from scratch again friday afternoon .

The pressure of getting the systems deliverd was to big , we could not hold them any longer ..

we deleted all hpvmdevmgmt , values and recreated them , looks like that did do the job .

In the beginning we made also some mistakes I assume with UUID's setting .

the overview I send this morning is the last one , after recreating everything .

, we are at this moment not in a position to test anything , on these systems , they will be available in 1 or two weeks again to change the config if neccesary .

[vandevoort]

Hi all .

sorry It took some time before I response but we are in the middle of a holiday season :

At this moment I still got one problem with one of the Service Guard Clusters :

all seems to work fine except

hpvmsg_move a Guest  to other hosts. ( in this case to ahp010)

hpvmsg_move: ERROR (hpvmsg_move): Node 'ahp010' is not up and running or it is not in the same cluster as this node.

and hpvmstatus -m   shows still :

No HPVM multi-server environment configured

The two Physical systems are ahp008 and ahp010 :

we also set the whole environment again :

ahp008:# hpvminfo -S
HPVM Server information
Version: HPVM B.04.30 ccipf opt Tue Jan 11 2011 08h48m13s PST
My partition ident: b3a1138c-1eba-4f18-8035-408f17f13528
My physical ident: 1b5bb8a6-71a1-11e0-b5ea-4d6b12350ba1

# hpvmdevmgmt -l env
HPVM_MSE_GROUP_ENTRY:CONFIG=env,EXIST=NO,DEVTYPE=UNKNOWN,SHARE=NO,UUID=f11abdfe-aee7-11e0-a301-0017a4778c10,GROUPNAME=HPVM-SG-ahp010_cluster::WWID_NULL

# hpvmdevmgmt -l server
ahp008:CONFIG=server,SERVERID=1,SERVERADDR=141.93.115.26,UUID=b3a1138c-1eba-4f18-8035-408f17f13528,PHYSUUID=1b5bb8a6-71a1-11e0-b5ea-4d6b12350ba1::WWID_NULL
ahp010:CONFIG=server,SERVERID=2,SERVERADDR=141.93.115.29,UUID=1882ec5c-f79c-41d0-aa83-f46b2363c44c,PHYSUUID=0e35e3ca-71a1-11e0-8c57-4f841a320a9d::WWID_NULL

ahp010:

# hpvminfo -S
HPVM Server information
Version: HPVM B.04.30 ccipf opt Tue Jan 11 2011 08h48m13s PST
My partition ident: 1882ec5c-f79c-41d0-aa83-f46b2363c44c
My physical ident: 0e35e3ca-71a1-11e0-8c57-4f841a320a9d

hpvmdevmgmt -l env
HPVM_MSE_GROUP_ENTRY:CONFIG=env,EXIST=NO,DEVTYPE=UNKNOWN,SHARE=NO,UUID=f11abdfe-aee7-11e0-a301-0017a4778c10,GROUPNAME=HPVM-SG-ahp010_cluster::WWID_NULL

# hpvmdevmgmt -l server
ahp008:CONFIG=server,SERVERID=1,SERVERADDR=141.93.115.26,UUID=b3a1138c-1eba-4f18-8035-408f17f13528,PHYSUUID=1b5bb8a6-71a1-11e0-b5ea-4d6b12350ba1::WWID_NULL
ahp010:CONFIG=server,SERVERID=2,SERVERADDR=141.93.115.29,UUID=1882ec5c-f79c-41d0-aa83-f46b2363c44c,PHYSUUID=0e35e3ca-71a1-11e0-8c57-4f841a320a9d::WWID_NULL

 cmviewcl

CLUSTER          STATUS
ahp010_cluster   up

  NODE           STATUS       STATE
  ahp008         up           running

    PACKAGE        STATUS           STATE            AUTO_RUN    NODE
    ahv008         up               running          enabled     ahp008
    ahv010         up               running          enabled     ahp008

  NODE           STATUS       STATE
  ahp010         up           running

    PACKAGE        STATUS           STATE            AUTO_RUN    NODE
    ahv012         up               running          enabled     ahp010
    ahv014         up               running          enabled     ahp010

we did a cmcheckconf on cluster and packages on each system they are identical to . and no errors ..

Off line Migration of the Guest works fine via Service Guard cmhaltpkg and cmrunkpg -n ahp010 <Guest>

Does anyone got an Idea ?

[Eric SAUBIGNAC]

Bonjour,

1) hpvmsg_move: ERROR (hpvmsg_move): Node 'ahp010' is not up and running or it is not in the same cluster as this node.

I took a look at the 4.3 script. The only thing that could produce this error, is if the result of the command cmwiewcl -l node is not exactly like this :

#cmviewcl -l node

  NODE           STATUS       STATE
  ahp008         up           running
  ahp010         up           running

Is it ? Well, I guess that if you run the command from a root shell, it works correctly.

So how do you use hpvmsg_move ? Through a remote shell (remsh) ? In the background from a daemonized process, like cron or a MCSG package ? From a user that has no right on the cluster ? Anything that could explain that the envrionment is not the very same than when you use directly a root shell ...

Maybe you could also try to debug. Just copy hpvmsg_move in a corner, and modify the copied version. For example lines 183 and 184 :

typeset nodeinfo=`cmviewcl -l node | awk '/NODE/{n=1;next};n{print $1,$2,$3}' OFS=:`
typeset nodeargs=`echo $nodeinfo | sed 's/:[^[:space:]*]*//g;s/[^[:space:]][^[:space:]]*/-n &/g'`

could be modified in :

cmviewcl -l node >/tmp/mydebug.txt 2>&1

typeset nodeinfo=`cmviewcl -l node | awk '/NODE/{n=1;next};n{print $1,$2,$3}' OFS=:`

echo "$nodeinfo" >>/tmp/mydebug.txt 2>&1
typeset nodeargs=`echo $nodeinfo | sed 's/:[^[:space:]*]*//g;s/[^[:space:]][^[:space:]]*/-n &/g'`

echo "$nodeargs"  >>/tmp/mydebug.txt 2>&1

/tmp/mydebug.txt should contain something like :

  NODE           STATUS       STATE
  ahp008         up           running
  ahp010         up           running
ahp008:up:running
ahp010:up:running
-n ahp008 -n ahp010

Then, use the modified version in the same way you used hpvmsg_move ..

2) and hpvmstatus -m   shows still : No HPVM multi-server environment configured

Well not sure it has a relation with the hpvmsg_move's problem : hpvmsg_move doesn't check the multi environment with hpvmstatus. But ... ?

In fact, everything seems OK from hpvmdevmgmt point of view. So 2 leads, to be taken ASIS :

- could you have a name resolution, and a reverse IP to name resolution, problem. I think that no, because MC/ServiceGuard is very sensitive to this kind of missconfiguration. But ... ?

- the second one is closed to superstition :-(((  : name of a node is ahp010 and name of the cluster is ahp010_cluster. Could we imagine that the code in hpvmstatus doesn't handle correctly the situation where the cluster and a node have the name beginning with the same string ? In this perspective, can you rename the cluster ? If you can, don't forget to modify the HPVM_MSE_GROUP entry.

Eric

[vandevoort]

Hallo Eric ..

cmviewcl -l node :

root@ahp008.ah.nl:/
# cmviewcl -l node

  NODE           STATUS       STATE
  ahp008         up           running
  ahp010         up           running

on both systems :

executed hpvmsg_move ( with debug)

named the modified executable  hpvmsg_move.debug

hpvmsg_move.debug -v -h ahp008 -P ahv012
hpvmsg_move.debug: ERROR (hpvmsg_move.debug): Node 'ahp008' is not up and running or it is not in the same cluster as this node.

  NODE           STATUS       STATE
  ahp008         up           running
  ahp010         up           running

Usage:

hpvmsg_move [-v] [-n|-h] <target-node> [-P] <package>

The hpvmsg_move command initiates an online migration (move) of a
virtual machine which has previously been associated with a
Serviceguard packages. The command will move both virtual machine and
package named in the <package> argument from the current running node
to the node to the node specified with the -n or -h option argument.

Where:
  -v verbose
  <target-node> is the Serviceguard node where the package should migrate to
  <package> Is the name of a packaged virtual machine to be migrated

cat /tmp/mydebug.txt

  NODE           STATUS       STATE
  ahp008         up           running
  ahp010         up           running
ahp008:up:running
ahp010:up:running
-n ahp008 -n ahp010

2)

renaming of the Cluster  is not so easy at this moment , some of these Guests running are claimed by development of application , I not allowed today to stop or start these Guests ..

Kind regards .

[Eric SAUBIGNAC]

a little be stuck ...

Still looking @ the script. Is the result of command hostname on ahp008, exactly ahp008 or something different like a FQDN ? If so the problem could be there

[vandevoort]

Hi Eric ,

the host name are FQDN .

But that is also for another server which works fine .

I can change in the Cluster Configuration into the FQDN ..

The FQDN  is neede  for the  application  they told me ..

will let you know the results tomorrow .

Thanks

[vandevoort · Accepted solution]

Hi Eric .

Thanks that did the job .

Of the Host systems renamed hostname to ahp008 and ahp010 no FQDN .

serviceGuard does not allwed to use FQDN .

I removed the whole service Guard configuration and build Service Guard up from scratch , during all Guest were down added them with hpvmsg_package into service Guard . they all woked fine now .

on line migratetion hpvmsg_move and off line .

even hpvmstatus  shows expected info :

 hpvmstatus -m
[HPVM Multi-server environment]

        Server group name       : HPVM-SG-cl_iris2
        Server group uuid       : f11abdfe-aee7-11e0-a301-0017a4778c10
        This server's identifier: 1

        Server_id               : 1
        Server_state            : 1
        Server_ipaddr           : 141.93.115.26
        Server_hostname         : ahp008
        Server_uuid             : b3a1138c-1eba-4f18-8035-408f17f13528
        Server_physical_uuid    : 1b5bb8a6-71a1-11e0-b5ea-4d6b12350ba1

        Server_id               : 2
        Server_state            : 1
        Server_ipaddr           : 141.93.115.29
        Server_hostname         : ahp010
        Server_uuid             : 1882ec5c-f79c-41d0-aa83-f46b2363c44c
        Server_physical_uuid    : 0e35e3ca-71a1-11e0-8c57-4f841a320a9d

Thanks for your support.