SPARC SuperCluster: How to apply idr482.7 to Solaris 11.1 SRU 5.5 LDoms and Zones on a SPARC SuperCluster

Asset ID:	1-71-1547278.1
Update Date:	2013-12-04
Keywords:

Solution Type Technical Instruction Sure

Solution 1547278.1 : SPARC SuperCluster: How to apply idr482.7 to Solaris 11.1 SRU 5.5 LDoms and Zones on a SPARC SuperCluster

Applies to:

SPARC SuperCluster T4-4 Full Rack - Version All Versions to All Versions [Release All Releases]
SPARC SuperCluster T4-4 - Version All Versions to All Versions [Release All Releases]
SPARC SuperCluster T4-4 Half Rack - Version All Versions to All Versions [Release All Releases]
Oracle Solaris on SPARC (64-bit)
This note applies to all SPARC SuperCluster Solaris 11 LDoms and Zones that have been patched to Solaris 11.1 SRU 5.5 .

Goal

SPARC SuperCluster offers a wide variety of configuration and virtualization options. With this in mind it is important to ensure that items such as patching are done in the proper order to ensure operational configuration when complete. This note will walk through the considerations and methodology to apply idr482.7 on top of Solaris 11.1 SRU 5.5 delivered via <Document 1452277.1> SPARC SuperCluster Critical Issues. Reference item SOL_11_7 in that note.

Solution

1) Verify each Solaris LDom and Zone to be patched is at Solaris 11.1 SRU 5.5

# pkg list entire
NAME (PUBLISHER) VERSION IFO
entire (solaris) 0.5.11-0.175.1.5.0.5.0 i--

2) Download the idr482.2.p5p from MOS <Patch 17005302>

3) Prepare the repo for the IDR in each applicable domain (Global Zone). Cleanly shutdown all databases , cluster ready services (CRS) and/or applications per your organizations standard operating procedures.

# pkgrepo create /var/tmp/idrrepo
# pkgrecv -s /var/tmp/idr482.7.p5p -d /var/tmp/idrrepo '*'
# pkg set-publisher -g file:///var/tmp/idrrepo solaris

4) Determine if you have virtualization considerations.

a) Locate and the log into your control domain.

# virtinfo -a

Domain role: LDoms control I/O service root
Domain name: primary
Domain UUID: 3c1874c3-d436-c941-d4f3-8a84cb52aa88
Control domain: orlscdb01
Chassis serial#: 1139BDY8C1

b) Get a list of all LDoms (domains). If you have 2 or fewer proceed to patching of guest domain in step

# ldm list

NAME             STATE      FLAGS   CONS    VCPU MEMORY   UTIL NORM UPTIME
primary          active     -n-cv- UART    128   523776M 3.0% 3.0% 2d 30m
ssccn1-app1      active     -n---- 5001    64    256G     0.1% 0.1% 2d 30m
ssccn1-app2      active     -t--v- 5002    64    256G     1.6% 1.6% 2d 30m

c) Determine which if any of your Solaris 11 LDoms are service domains. You can determine this by looking if the domain has its DISK listed as vdiskx in the name column and are being served via a virtual disk service VDS from the other domains. In this example ssccn1-app1 is a non IO domain. You may have 1 or more of these and they should be patched first. If you do not have any proceed to patching of guest domain in step

# ldm list -o disk

VDS
    NAME             VOLUME         OPTIONS          MPGROUP        DEVICE
    primary-vds0     vol1                                           /dev/dsk/c0t5000C5003BE98F4Bd0s1
------------------------------------------------------------------------------
NAME
ssccn1-app1

DISK
    NAME             VOLUME                      TOUT ID   DEVICE SERVER         MPGROUP
    vdisk1           vol1@primary-vds0                0    disk@0 primary
    vdisk2           vol1@service-vds0           1    1    disk@1 ssccn1-app2

------------------------------------------------------------------------------
NAME
ssccn1-app2

VDS
    NAME             VOLUME         OPTIONS          MPGROUP        DEVICE
    service-vds0     vol1                                           /dev/dsk/c0t5000C50047C18E17d0s1

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

5) Log into the non IO domain(s) and patch it(them). If this LDom has Solaris 11 local zones the patching process will automatically patch those as well as long as they are started and meet the Solaris 11.1 SRU5.5 requirement. Note the be-name chosen was derived form appending my existing be-name with -idr482. So your option for --be-name may vary.

# pkg install --be-name SSCMU_2013.04-idr482 -g /var/tmp/idr482.7.p5p idr482

6) Cleanly shutdown any zones if applicable and then the non IO domain(s)

# shutdown -g 0 -y

7) Exit out to the control domain and shutdown the non IO domain(s)

# ldm stop ssccn1-app1

8) Log into the guest IO domain, in this example ssccn1-app2 and patch it. If this LDom has Solaris 11 local zones the patching process will automatically patch those as well as long as they are started and meet the Solaris 11.1 SRU5.5 requirement. Note the be-name chosen was derived form appending my existing be-name with -idr482. So your option for --be-name may vary.

# pkg install --be-name SSCMU_2013.04-idr482 -g /var/tmp/idr482.7.p5p idr482

9) Cleanly shutdown any zones if applicable and then reboot the guest IO domain

# reboot

10) Exit out to the control (primary) domain and patch it. If this LDom has Solaris 11 local zones the patching process will automatically patch those as well as long as they are started and meet the Solaris 11.1 SRU5.5 requirement. Note the be-name chosen was derived form appending my existing be-name with -idr482. So your option for --be-name may vary.

# pkg install --be-name SSCMU_2013.04-idr482 -g /var/tmp/idr482.7.p5p idr482

11) Cleanly shutdown any zones if applicable and then reboot the control (primary) domain.

# reboot

12) Restart the non IO domain(s) In this example ssccn1-app1.

# ldm start ssccn1-app1

13) Verify in all applicable Solaris 11 domains and zones that the IDR is installed.

# pkg list |grep idr
idr482 (solaris) 2 i--

14) Restart all databases , cluster ready services (CRS) and/or applications per your organizations standard operating procedures.

References

<NOTE:1452277.1> - SuperCluster Critical Issues
<BUG:16409079> - THREADS STUCK IN FED_BREADLINE ARE CAUSING ORACLE DB TIMEOUTS

Attachments

This solution has no attachment