Upndu 410
Upndu 410
Upndu 410
00 KB)
8/5/2016 7:49 AM
Last saved by boykot
upndu410_R0013
Note that this procedure also covers loading FRUMON code packages (which are rebootless
packages) using the Hot Fix Wizard, which is part of USM.
.2
Connect to the storage system from a management station on the same LAN as the primary
management ports of the SP. You cannot use the Service LAN port to perform the upgrade (USM
cannot retrieve diagnostic data).
.3
Running USM:
.a When you have loaded USM on a windows server/laptop with access to both SPs from the
customers LAN, start USM. (You cannot use the Service LAN port to perform the upgrade since
USM cannot retrieve diagnostic data)
.b Log into the storage system to be updated using the IP address of one of the SPs.
Note: If you are unable to login to the SP because of error Security Uninitialized refer to knowledge
base solutions emc101373, emc103955, and emc75148.
.c Enter Engineering mode by pressing Ctrl/shift/F12 and entering the password SIR (case
sensitive). This indicates that you are a service person performing the activity and you are allowed
to Override a failed rule if it can be overridden.
Perform a disruptive upgrade. If you select this, then both SPs will be rebooted at the same
time. Select this option only if the customer has arranged for all attached hosts to be off-line.
You will see Engineering Mode displayed.
.d Select Software, System Software, Install Software.
.e Select Custom Install.
(Express Install is offered if you pre-staged software within the last 2 days. It is OK to accept this
option if it is presented.)
.f Check the box labeled Allow Installation of Diagnostic Tools on SPs if the customer permits this.
Note: This utility (1) checks if there is an active session on the SP, and (2) monitors the processor
utilization and offers steps to reduce it if it is too high to allow the NDU to take place. These steps are
based on the recommendations in emc211113. You can choose to perform any, all, or none of the
steps. If the utilization drops to a satisfactory level after a task has been run, then the remaining steps
are skipped. Any change made will be reset at the end of the NDU.
Note: The utility is removed from the SPs at the end of the NDU.
.g A box labeled Enable Offline Upgrade is shown. If you check this box, then both SPs will be
rebooted at the same time.
Select this option only if the customer has arranged for all attached hosts to be off-line.
WARNING:
The only confirmation that this option has been selected is the final screen before the
NDU process begins.
If you are not sure what option you selected, click on the Enable Off-line Upgrade box
again, and repeat the confirmation step.
USM versions 1.0.5.1 and later: the check-box selection will be displayed as expected.
.h The following pop-up is displayed.
Rules Checking
.a If the Rules step Event Log rule takes a long time to complete or an NDU using CLI times out and
returns a failure message: see Knowledgebase solution emc213189 Event log rule takes a long
time to complete when Software Assistant (SA) rules are run
.b If the Rules step appears to hang (takes 20 minutes or longer), it may be that a large event log is
causing the delay. You can perform the following steps:
.a Restart CIMOM this will fail all the rules that have not yet returned so that you can exit the
wizard.
.b Copy off the event logs or run SPCollects to preserve the event log information.
.c Clear the event logs.
If these steps do not resolve the issue, restart CIMOM and use CLI to perform the upgrade.
If the NDU fails to complete, BEFORE performing any attempts to diagnose and address the issue, do
the following:
.a Capture the diagnostic data file created by USM as the very first step in the NDU process. This will
be on the station from which USM was run, in the C:\EMC\repository\DiagnosticData\<serial
number> folder and will be named <date_time>.clariion_Healthcheck_<serial
number>_<date_time>.zip
.b Generate and capture a new diagnostic data file using USM, Diagnostics, Capture Diagnostic
Data. This will contain fresh SPcollects. Again, this will be on the station from which USM was run,
in the C:\EMC\repository\DiagnosticData\<serial number> folder and will be named
<date_time>.clariion_Healthcheck_<serial number>_<date_time>.zip
.c Escalate the issue and provide the 2 diagnostic data files that you captured.
.6
Re-enabling statistics logging if the wizard had you turn it off (if it was enabled prior to the NDU)
Follow the directions given by the utility, including the steps to send the updated system storage
information.
Note: If you do not have Internet connectivity at the time of the upgrade, the file can be uploaded after
the activity by using the Submit Configuration Report option in the left-hand pane and following the
directions given by the Utility. The saved registration file is located at C:\emc\repository\Registration
on the workstation the activity was performed on. If necessary, the file can be copied and sent from a
different computer that has the USM installed and an Internet connection.
.7
Using USM, install the Recovery Image for the FLARE version you have just installed.
This NDU is rebootless.
It does not matter if this has already been loaded; reloading this image will not cause a problem.
.8
.9
View the event logs of both SPs to ensure there are no unexpected events that could signal a problem
that occurred during or immediately after the NDU
.10
Restart all stopped SAN Copy sessions and any other replication software activities that were stopped
prior to this code load
.11
If host operating systems are running native multipath software ask the customer to verify that LUNs
are trespassed back to their default SP owner.
Note: Sancopy source LUNs must be listed to their default owner SP or a Sancopy operation will fail.
.13
If required, upgrade ESRS IP Client for CLARiiON to the version that corresponds to the FLARE that
was upgraded to in this procedure. Even if you are loading a patch version of the bundle, there could
be a new Dial-Home template that must be updated if it has been revised.
.14
Using Unisphere Release Notes and the CLARiiON activity Guide (CAG), check that all host based
and array-based software is at an acceptable/compatible revision. If not, then update as required.
.15
Confirm that there are no array faults (note that the SPS units may still be charging based upon the SP
reboots; this is normal). Write cache will automatically re-enable if it was enabled prior to the NDU as
soon as one SPS is fully charged and ready
.16
Confirm that the cache settings are set properly. If new replication software (layered applications) is
being added, cache settings may not be able to be set as they were previously
.17
Were you instructed earlier to manually disable and zero write cache?