Oracle Exalytics In-Memory Machine diagnostic data collection script

Asset ID:	1-79-1527167.1
Update Date:	2017-01-12
Keywords:

Solution Type Predictive Self-Healing Sure

Solution 1527167.1 : Oracle Exalytics In-Memory Machine diagnostic data collection script

Applies to:

Exalytics In-Memory Machine X4-4 - Version All Versions to All Versions [Release All Releases]
Exalytics In-Memory Machine X3-4 - Version All Versions to All Versions [Release All Releases]
Exalytics In-Memory Machine X2-4 - Version All Versions to All Versions [Release All Releases]
Exalytics In-Memory Machine X5-4 - Version All Versions to All Versions [Release All Releases]
Exalytics In-Memory Machine X6-4 - Version All Versions to All Versions [Release All Releases]
Information in this document applies to any platform.
Oracle Exalytics In-Memory Machine diagnostic data collection script

Purpose

This document contains details on the information collected by and the usage of the Oracle Exalytics In-Memory Machine diagnostic data collection script

Details

The exalyticsdiag.sh script is a diagnostic collection tool for the X86 based Oracle Exalytics In-Memory Machine (X2-4, X3-4, X4-4, X5-4, X6-4 models).
It collects various files and outputs from the system and packages them into a tar.bz2 file that can be easily uploaded to a support case to assist the support center in troubleshooting issues. This script does not collect any ILOM level data and if that information is required it is recommended to manually collect an ILOM snapshot also.

For detailed steps on collecting an ILOM snapshot please refer to the following document: How to run an ILOM Snapshot on a Sun/Oracle X86 System (Doc ID 1448069.1)

To use this tool download the exalyticsdiag.sh script and save it to the /opt/exalytics/bin directory on your Oracle Exalytics Server. If the system does not already have this directory you should create it with the mkdir command

# mkdir -p /opt/exalytics/bin
# mv exalyticsdiag.sh /opt/exalytics/bin

after downloading make sure the script is executable:

# cd /opt/exalytics/bin
# chmod 755 exalyticsdiag.sh

The script needs to be run as the root user and will collect the data and place the resulting datestamped tar.bz2 file in the /tmp directory.

Usage information:

# ./exalyticsdiag.sh -h

Oracle Exalytics In-Memory Machine - Diagnostic Data Collection Tool
exalyticsdiag.sh
Version: 1.4_20170111

Usage: ./exalyticsdiag.sh [-version | -help]

-V, -version
Output version information

-h, -help
Output this usage information

Example output seen when running the script without any options (output may vary slightly between models):

[root@exalytics-a bin]# /opt/exalytics/bin/exalyticsdiag.sh

Oracle Exalytics In-Memory Machine - Diagnostic Data Collection Tool
System Model: ORACLE SERVER X5-4
Script Version: 1.4_20170111

Collecting logs....
Collecting system configuration data....
Collecting hmp output....
Collecting network info....
Collecting disk & raid info....
Collecting flash data....
Qty: 3 NVMe based Flash Accelerator Cards found.1.2.3.
SUNW-NVME-1
Retrieve crash dump successfully.
Retrieve memory dump successfully.
SUNW-NVME-2
Retrieve crash dump successfully.
Retrieve memory dump successfully.
SUNW-NVME-3
Retrieve crash dump successfully.
Retrieve memory dump successfully.
Generating diagnostic tarball and removing temp directory

=========================================================================
Done.
The report files are bzip2 compressed and located in:
/tmp/exalyticsdiag_exalytics-a_2017_01_12_10_58.tar.bz2
=========================================================================

The following is a list of all the files, command outputs and their locations collected by the script:

/ directory:
 version_exalyticsdiag               information about the version of exalyticsdiag, runtime, and command-line options used

/flash/ directory:
 NO_FLASH_CARDS_FOUND                empty file created if no Flash cards are found
 flash_collect.err                   output of any errors encountered while collecting flash data
 exalytics_CheckFlash.out            /opt/exalytics/bin/exalytics_CheckFlash.sh
 mdstat                              /proc/mdstat
 mdadm.conf                          /etc/mdadm.conf
 mdadm-D.err                         output of any errors encountered while executing mdadm -D
 mdadm-D.out                         /sbin/mdadm -D /dev/md*

*if the system has F40/F80 Flash Accelerator Cards installed the following will be collected
 ddcli-cN-health.out                 /opt/exalytics/flashUtil/ddcli -c N -health (command run for each Flash card found N=card number)
 ddcli-cN-list.out                   /opt/exalytics/flashUtil/ddcli -c N -list (command run for each Flash card found N=card number)
 ddcli-cN-showvpd.out                /opt/exalytics/flashUtil/ddcli -c N -showvpd (command run for each Flash card found N=card number)
 ddcli-listall.out                   /opt/exalytics/flashUtil/ddcli -listall
 ddoemcli-healthall.out              /opt/exalytics/flashUtil/ddoemcli -healthall

*if the system has NVMe based Flash Accelerator Cards installed the following will be collected
 nvmeadm_getlog-h_SUNW-NVME-N.out    /usr/bin/nvmeadm getlog -h SUNW-NVME-N (command run for each Flash card found N=card number)
 nvmeadm_getlog-e_SUNW-NVME-N.out    /usr/bin/nvmeadm getlog -e 100 SUNW-NVME-N (command run for each Flash card found N=card number)
 nvmeadm_list-v.out                  /usr/bin/nvmeadm list -v
 nvmeadm_list-s.out                  /usr/bin/nvmeadm list -s
 nvmeadm_namespace-v.out             /usr/bin/nvmeadm namespace -v

 crashdump_SUNW-NVME-N               output from nvmeadm getlog -s for Samsung based flash cards
 memorydump_SUNW-NVME-N              output from nvmeadm getlog -s for Samsung based flash cards
 eventlog_SUNW-NVME-2                output from nvmeadm getlog -s for Intel based flash cards
 nlog_SUNW-NVME-2                    output from nvmeadm getlog -s for Intel based flash cards

/log/ directory:
 dmesg /bin/dmesg
 /var/log/init-exalytics files       /var/log/init-exalytics/*
 /var/log/exalytics files            /var/log/exalytics/*
 mcelog                              /var/log/mcelog
 all messages logs                   /var/log/messages*
 log_collect.err                     output of any errors encountered while collecting log data

*if system is virtualized the following will be collected
 xm_list.out                         /usr/bin/xm list
 /var/log/ov*                        all ov* files from /var/log
 xend-config.sxp                     /etc/xen/xend-config.sxp
 xen_collect.err                     output of any errors encountered while collecting xen data

/net/ directory:
 iblinkinfo.out                      /usr/sbin/iblinkinfo
 ibstatus.out                        /usr/sbin/ibstatus
 ifconfig-a.out                      /sbin/ifconfig -a
 ip6tables.out                       /sbin/ip6tables-save
 iptables.out                        /sbin/iptables-save
 ls-l_network-scripts.out            /bin/ls -l /etc/sysconfig/network-scripts
 netstat-nr.out                      /bin/netstat -nr
 network                             /etc/sysconfig/network
 net_collect.err                     output of any errors encountered while collecting network data
/net/bonding directory: 
 all bond* files                     /proc/net/bonding/*
/net/network-scripts directory: 
 all ifcfg* files                    /etc/sysconfig/network-scripts/ifcfg*

/raid/ directory:
 fdisk-l.err                         error output from fdisk -l
 fdisk-l.out                         /sbin/fdisk -l
 lsscsi-vv.out                       /usr/bin/lsscsi -vv
 megacli64-AdpAllInfo.out            /opt/MegaRAID/MegaCli/MegaCli64 -AdpAllInfo -aALL
 megacli64-BbuCmd.out                /opt/MegaRAID/MegaCli/MegaCli64 -adpbbucmd -aALL
 megacli64-CfgDsply.out              /opt/MegaRAID/MegaCli/MegaCli64 -cfgdsply -aALL
 megacli64-FwTermLog.out             /opt/MegaRAID/MegaCli/MegaCli64 -fwtermlog -dsply -aALL
 megacli64-GetEvents-all.out         /opt/MegaRAID/MegaCli/MegaCli64 -AdpEventLog -GetEvents -aALL
 megacli64-LdInfo.out                /opt/MegaRAID/MegaCli/MegaCli64 -LDInfo -LALL -aALL
 megacli64-LdPdInfo.out              /opt/MegaRAID/MegaCli/MegaCli64 -LdPdInfo -aALL
 megacli64-PdList_long.out           /opt/MegaRAID/MegaCli/MegaCli64 -PDList -aALL
 megacli64-PdList_short.out a        shortened output of MegaCli64 -PDList -aALL showing disks and their status
 parted.err                          error output from /sbin/parted
 parted-ls.out                       /sbin/parted --script --list

/sysconfig/ directory:
 chkconfig--list.out                 /sbin/chkconfig --list
 cpuinfo                             /proc/cpuinfo
 df-hl.out                           /bin/df -hl
 dmidecode.out                       /usr/bin/dmidecode
 exalytics_CheckHWnFWProfile.out     /opt/exalytics/bin/exalytics_CheckHWnFWProfile (only collected on systems with PS3 or later installed)
 exalytics_CheckSWProfile.out        /opt/exalytics/bin/exalytics_CheckSWProfile (only collected on systems with PS3 or later installed)
 exalytics_imageinfo.out             /opt/exalytics/bin/exalytics_imageinfo (only collected on systems with PS3 or later installed)
 last-ax.out                         /usr/bin/last -ax
 lspci.out                           /sbin/lspci
 lspci-vvv.out                       /sbin/lspci -vvv
 lspci-xxxx.out                      /sbin/lspci -xxxx
 lsusb.out                           /sbin/lsusb
 lsusb-v.out                         /sbin/lsusb -v
 meminfo                             /proc/meminfo
 modprobe.conf                       /etc/modprobe.conf
 ps-aux.out                          /bin/ps aux
 pstree-aApu.out                     /usr/bin/pstree -aApu
 redhat-release                      /etc/redhat-release
 rpm-qa.out                          /bin/rpm -qa
 rpm-qai.out                         /bin/rpm -qai
 sysconfig.err                       output of any errors encountered while collecting sysconfig data
 uname-a.out                         /bin/uname -a
 uptime.out                          /usr/bin/uptime

*if the HMP package is installed the following will be collected
/sysconfig/hmp/ directory:
 fwupdate_list_all.out               /usr/sbin/fwupdate list all
 hmp.err                             output of any errors encountered while collecting hmp data
 hwmgmtcli_list_all-d.out            /usr/sbin/hwmgmtcli list all -d
 lomconfig_list_system-summary.out   /usr/sbin/ilomconfig list system-summary
 raidconfig_list_all.out             /usr/sbin/raidconfig list all
 raidconfig_list_all-v.out           /usr/sbin/raidconfig list all -v

Attachments

This solution has no attachment