Microsoft Cluster node service failing automatically

[former_member409456]

Hello Expert,

We have Net weaver 7.0 EHP 2 installed on Windows 2008 R2 for EP. It is installed on cluster environment.

We have 2 cluster node Host A and Host B. Also we have 2 services one is for database and another is for SCS. During the failover these 2 services will move to another node.

My problem is SCS cluster service is getting offline automatically which is making my entire EP production server down. As it gets down i manually start cluster service first then app server and my EP system gets start.

Please suggest how can i find the root cause for getting SCS service offline or How we can make it always online?

Regards,

[Former Member]

Hi Tarun,

Please check the cluster event logs , that is the first point you need to check when your application is not stable. if it is taking frequent restarts.

Check with the windows server team if there are any security patches are updated recently.

My point you would be able to get the required information from cluster event logs.

Thanks Regards,

Avinash I

[former_member409456]

HI Avinash,

In event log i can so many cluster failure log which generating continuously and also my server is running fine.

[Former Member]

Do you maintain the cluster or there is any windows team who maintain? If there is are windows specialist then I would suggest you take help.

[Former Member]

Hi Tarun,

You can get few limited Event logs in Cluster manager, As sunil suggested please reach out to the windows team to get the event logs.

Thanks Regards,m

Avinash I

[former_member409456]

HI Sunil,

There is no windows team.

Also i checked windows event log and following events observed with error :

1196 Network name resource
1205 Resource control manager

1579 Network name resource

1069

I am surprise this is generating continuously while EP is running.

[Sriram2009]

Hi

Kindly check the following things

1. During failover from node A to B Cluster disk getting online? (That on SAPEP1 group)
2. Have you installed any Antivirus software? Some time antivirus software will not release the cluster disk
3. Just restart the both systems & before starting the SAP just do the failover from node A to B resources are getting online or offline?

BR

SS

[former_member409456]

HI Sriram,

1. Cluster disk remains same. Only the cluster service for SCS is getting online.

2. We have McAfee installed on windows. Please tell when it not release the cluster disk ie while scanning or it not release after scanning complete??

3. When we manually moves the services it works thing is it is getting offline automatically.

Regards,

Tarun

[Sriram2009]

Hi

1. Cluster disk remains same. Only the cluster service for SCS is getting online.

During failover from node A to B along with service its has to move the Cluster disk, in your case SAP EP1 under this group you may have the IP, Cluster disk & SAP Service those are offline in node A & it should online in Node B, and also could you paste the screen shot of SAP EP1 (Service & application group)

 2. We have McAfee installed on windows. Please tell when it not release the cluster disk ie while scanning or it not release after scanning complete??

Check the Mcafee is having firewall settings & cluster disk virus scanning feature?

3. When we manually moves the services it works thing is it is getting offline automatically.

            Could you check the cluster Event viewer. In both Node SAP EP1 instance numbers are same or different?

Regards

Sriram

[Former Member]

When the SCS goes down you need to look into the logs from SCS work directory. Without logs bit difficult to help whats the issue is. As you have done manual restart the logs may be overwritten but there may be .old logs can you please attach these?

[former_member409456]

HI Sunil,

I checked dev_ms.old file and below is log:

---------------------------------------------------

trc file: "dev_ms", trc level: 1, release: "720"

---------------------------------------------------

[Thr 7224] Fri Mar 21 14:05:02 2014

[Thr 7224] ms/http_max_clients = 500 -> 500

[Thr 7224] MsSSetTrcLog: trc logging active, max size = 52428800 bytes

systemid   562 (PC with Windows NT)

relno      7200

patchlevel 0

patchno    101

intno      20020600

make       multithreaded, Unicode, 64 bit, optimized

pid        9488

[Thr 7224] ***LOG Q01=> MsSInit, MSStart (Msg Server 1 9488) [msxxserv.c   2274]

[Thr 7224] Fri Mar 21 14:05:03 2014

[Thr 7224] load acl file = \\EP1SAPGRP\sapmnt\EP1\SYS\global\ms_acl_info.DAT

[Thr 7224] MsGetOwnIpAddr: my host addresses are :

[Thr 7224]   1 : [IP] HOST (HOSTNAME)

[Thr 7224]   2 : [127.0.0.1] FQDN (LOCALHOST)

[Thr 7224]   3 : [IP] FQDN (NILIST)

[Thr 7224]   4 : [IP] EPCLUSTER (NILIST)

[Thr 7224]   5 : [IP] EP1SAPGRP (NILIST)

[Thr 7224]   6 : [IP] EP1ORAGRP (NILIST)

[Thr 7224]   7 : [IP] FQDN (NILIST)

[Thr 7224]   8 : [IP] FQDN (NILIST)

[Thr 7224] MsHttpInit: full qualified hostname = NODE A

[Thr 7224] HTTP logging is switch off

[Thr 7224] set HTTP state to LISTEN

[Thr 7224] *** HTTP port 8110 state LISTEN ***

[Thr 7224] *** I listen to internal port 3910 (3910) ***

[Thr 7224] *** HTTP port 8110 state LISTEN ***

[Thr 7224] CUSTOMER KEY: ><

[Thr 7224] build version=720.2011.05.04

[Thr 7224] MsJ2EE_CheckLoggedInNode: logged in list is not initialized -> reconnect ok

[Thr 7224] MsJ2EE_CheckDisconnectedNode: node [114836600] is not in the logged in list -> reconnect ok

[Thr 7224] MsJ2EE_AddLoggedInNode: add node [114836600] into logged in list

[Thr 7224] MsJ2EE_CheckLoggedInNode: node [128683700] isn't in the logged in list -> reconnect ok

[Thr 7224] MsJ2EE_CheckDisconnectedNode: node [128683700] is not in the logged in list -> reconnect ok

[Thr 7224] MsJ2EE_AddLoggedInNode: add node [128683700] into logged in list

[Thr 7224] MsJ2EE_CheckLoggedInNode: node [128683751] isn't in the logged in list -> reconnect ok

[Thr 7224] MsJ2EE_CheckDisconnectedNode: node [128683751] is not in the logged in list -> reconnect ok

[Thr 7224] MsJ2EE_AddLoggedInNode: add node [128683751] into logged in list

[Thr 7224] MsJ2EE_CheckLoggedInNode: node [139051900] isn't in the logged in list -> reconnect ok

[Thr 7224] MsJ2EE_CheckDisconnectedNode: node [139051900] is not in the logged in list -> reconnect ok

[Thr 7224] MsJ2EE_AddLoggedInNode: add node [139051900] into logged in list

[Thr 7224] MsJ2EE_CheckLoggedInNode: node [114836650] isn't in the logged in list -> reconnect ok

[Thr 7224] MsJ2EE_CheckDisconnectedNode: node [114836650] is not in the logged in list -> reconnect ok

[Thr 7224] MsJ2EE_AddLoggedInNode: add node [114836650] into logged in list

[Thr 7224] MsJ2EE_CheckLoggedInNode: node [139051951] isn't in the logged in list -> reconnect ok

[Thr 7224] MsJ2EE_CheckDisconnectedNode: node [139051951] is not in the logged in list -> reconnect ok

[Thr 7224] MsJ2EE_AddLoggedInNode: add node [139051951] into logged in list

[Thr 7224] MsJ2EE_CheckLoggedInNode: node [139051950] isn't in the logged in list -> reconnect ok

[Thr 7224] MsJ2EE_CheckDisconnectedNode: node [139051950] is not in the logged in list -> reconnect ok

[Thr 7224] MsJ2EE_AddLoggedInNode: add node [139051950] into logged in list

[Thr 7224] MsJ2EE_CheckLoggedInNode: node [128683750] isn't in the logged in list -> reconnect ok

[Thr 7224] MsJ2EE_CheckDisconnectedNode: node [128683750] is not in the logged in list -> reconnect ok

[Thr 7224] MsJ2EE_AddLoggedInNode: add node [128683750] into logged in list

[Former Member]

can you please attach dev_en*.old files.