Friday, April 20, 2018

HP drivers compatibility with Red Hat Enterprise Linux 7 kernel

Hi DBAs,

Recently I installed   Red Hat Enterprise Linux  7.3  on HP Hardware and applied latest redhat patches which upgraded the kernel version to 7.4 . I came across issue where I was getting hpdsa HP driver error while booting with latest kernel after applying latest Red Hat patches because the kernel version got updated to latest 7.4 kernel due to this issue Red Hat Linux Operating system was not booting up and was giving error. After doing some research and with assistance of Red hat Support I was able to figure and resolve this issue. Sharing details of the Error and Steps followed to resolve this issue.

Kernel details:


Error:Unable to handle kernel NULL pointer dereference 
                                           Modules linked  hpdsa


Observation and Resolution:

 There is a problem with Raid driver for Gen9 Server . We had installed the driver during installation but the module was close driver from HPE and it needs to be upgraded accordingly. Our updated kernel was not kernel for Red Hat 7.3 but is was for Redhat 7.4


Red Hat Enterprise Linux 7.0 (Maipo), June 10,2014   ; 4 years ago, uses Linux Kernel 3.10.0-123
Red Hat Enterprise Linux 7.1, (Maipo), March 5,2015 ; 3 years ago, uses Linux Kernel 3.10.0-229
Red Hat Enterprise Linux 7.2 (Maipo), Nov 19,2015 ;   3 years ago, uses Linux Kernel 3.10.0-327
Red Hat Enterprise Linux 7.3 (Maipo), Nov   3,2016 ; 2 years ago, uses Linux Kernel 3.10.0-514
Red Hat Enterprise Linux 7.4 (Maipo), Aug  1,2017 ; 1 years ago, uses Linux Kernel 3.10.0-693

Latest driver for HP can be found in below location:

https://downloads.linux.hpe.com/SDR/repo/spp-gen9/2018.03.0/packages



As we can see for 7.4 kernel we have a different driver available

Each Red Hat release has one kernel version and only fixes so from 3.10.0-514 they will make only 3.10.0-514.XYX ,so same kernel module will work but if you upgrade to newer Red Hat Kernel (7.4) than new kernel module is required.

Download the driver module kmod-hpdsa-1.2.10-123,rhel7u4.x86_64.rpm and copy to the server.
Boot the RHEL server with older version of  kernel(Red Hat Enterprise Linux 7.3 (Maipo), Nov   3,2016 ; 2 years ago, uses Linux Kernel 3.10.0-514) which is working fine and install the rpm as below:

rpm –ivh packagename.rpm
rpm -ivh  kmod-hpdsa-1.2.10-123,rhel7u4.x86_64.rpm

After successfully installing RPM verify the kernel version, HP Driver module version and reboot the server. Use below commands for validations.

#rpm -q kernel
#rpm -a|grep -i kernel*
#modinfo hpdsa
# init 6

Choose below kernel to boot the server:

Red Hat Enterprise Linux 7.4 (Maipo), Aug  1,2017 ; 1 years ago, uses Linux Kernel 3.10.0-693

Now it will work and the issue is resolved.

Enjoy Learning and Working on Red Hat Linux ...

Thanks,





Friday, April 13, 2018

clrg comand for Oracle Apps DBA in Oracle SuperCluster Environment

Hi All,

In Oracle SuperCluster Environment we use important command 'clrg' for stopping and Starting of Oracle Web tier and Concurrent Manager Services.

clrg stands for clresourcegroup, which assist us to manage resource groups for Oracle Solaris Cluster data services or  on Oracle SuperCluster Environment

Below are commands we use on Oracle Apps nodes as 'root' Operating system user on Solaris Environment where Oracle Supercluster is running:

In order to:
·         Stop application on only application server 1 to stop all opmn and concurrent manager services on apps server1 and server2 completely.

  # clrg offline opmn-rg
  # clrg offline cm-rg


 Start application on only application server 1:

   # clrg online cm-rg
  #  clrg online opmn-rg

·         Stop application on one application server (i.e Apps Server 1 or Apps Server 2):

 # clrg offline –n nodename opmn-rg
# clrg offline –n nodename cm-rg

·         Start application on one application servers (i.e Apps Server 1 or Apps Server 2):

 # clrg online –n nodename cm-rg
 # clrg online –n nodename opmn-rg

·         Monitor the cluster resources
 # clrg status
 # clrs status

·         Monitor the full cluster staus:

# cluster status

Example :

root@erpclusterapps1:~# cluster status

=== Cluster Resource Groups ===

Group Name     Node Name          Suspended     State
----------     ---------          ---------     -----
scalmnt-rg     erpclusterapps2     No            Online
               erpclusterapps1     No            Online

cm-rg          erpclusterapps2     No            Online
               erpclusterapps1     No            Online

opmn-rg        erpclusterapps2     No            Online
               erpclusterapps1     No            Online


=== Cluster Resources ===

Resource Name     Node Name          State      Status Message
-------------     ---------          -----      --------------
erp-fs-rs         erpclusterapps2     Online     Online
                  erpclusterapps1     Online     Online

cm-rs             erpclusterapps2     Online     Online - Service is online.
                  erpclusterapps1     Online     Online - Service is online.

cmlsr-rs          erpclusterapps2     Online     Online - Service is online.
                  erpclusterapps1     Online     Online - Service is online.

opmn-rs           erpclusterapps2     Online     Online
                  erpclusterapps1     Online     Online

Hope it is useful. Enjoy learning Apps DBA on Exadata / Supercluster Environment


Thanks,

Rafi