DELL EMC SC9000 Storage Array User Guide
- June 9, 2024
- DELL EMC
Table of Contents
UPTIME BULLETIN
A Newsletter from CEE, a Division of Dell EMC Unity/SC/PowerScale/ME4/ME5
Q4
SC9000 Storage Array
Two very rare issues with certain SLIC models on Dell EMC Unity with OEs <
5.2.1 A rare SLIC issue is discussed in LKB 205324 “Dell Unity: I/O Module
ports may unexpectedly become unresponsive, fail to pass traffic, or report as
down (Dell Correctable)” This KB discusses a very rare Firmware assertion
issues for some SLIC models. Affected hardware includes ClearskyX 4 Port
25GbE/iSCSI Xcede and Lightblade 4-port 10G/25G Optical OCP. This issue can
occur very rarely and unexpectedly. If the issue occurs, ports could become
unexpectedly unresponsive, failing to pass traffic, or reporting Links as
down. Storage objects such as NAS Servers may report errors and be unreachable
by end users. Performance degradation could occur as a result of some ports no
longer responding to I/O requests. Should this very rare issue occur, a reboot
of the storage processor will restore access. Alternatively, affected objects
may be failed over.
This issue is fixed in Dell EMC Unity OE 5.2.1.x. Dell EMC Unity engineering
is still interested in investigating individual instances of this issue sa
customer opening a ticket for this issue may be asked to provide logs and
manual dumps to help isolate the trigger conditions for this very rare issue.
In some instances, customers have been asked to replace the affected hardware
as well. A second SLIC related issue which has been very rarely seen can cause
a customer to lose access to SMB/NFS shares if the NIC FW hangs (asserts). The
workaround is to reboot the SPThis problem occurs due to an application or
hardware peripheral in the SAN sending large packets (greater than 9K) to the
array. The NIC FW has been updated in Dell EMC Unity OE 5.2.x and later to
avoid this issue (drop packet/not assert). This problem has been observed on
the ClearskyX 4 Port 25GbE/iSCSI Xcede IOM and may also be possible on the
Lightblade 4port 10G/25G IOM. If you think you may have encountered this
issue, please gather logs and open a ticket with Dell EMC technical support
for further investigat
SNMP may stop responding after upgrading to the latest OE versions for ME4 and
ME5 SNMP services may stop responding after approximately 24 hours following
an upgrade to GT280R010-01 firmware (for ME4) or ME5 1.0.1.0 (for ME5).
The bug appears to be due to a security fix added to the snmp-net library.
Further investigation is being done by one of our partner/vendors. Currently
the only available work-around is to restart SNMO services daily, either
manually or by script. Restarting both MCs also temporarily fixes the issue.
Alternatively you can use non-SNMP solutions, such as OME. SMI-S is also
impacted, as is the API.
Fixes are in progress and may be available in the next software releases for
each platform. For the latest information on the issue and the expected fix
releases and availability, please see
LKB000203521 for
ME4 or
PSE-18611.
VOLUME 43
December, 2022
Rare Unity SLIC issues seen prior to 5.2.1 1
SNMP may stop working after 24 hours on latest OE 1
versions for ME4 and ME5
Upcoming SRS certificate changes that could impact connectivity in January
2023 2
Some Dell EMC Unity storage processors could reboot after 270 days 2
Protecting your replicas after a cache loss 2 Did you know? 2 Latest code
releases and targets.
3 CloudIQ update. 4
You can always access the latest Uptime Bulletin by Support for Dell EMC Unity
Family | Overview | Dell US, then select “Monthly Support Highlights” within
“Featured
Content.”
UPTIME BULLETIN
Date that customers running Unity OE 5.0.3 or earlier and using eVE will lose SRS connectivity moved to Jan, 2023 In order to maintain SRS/CIQ connectivity after January, 2023:
-
If using a VE (centralized gateway): Customers need to upgrade their VE if they are running less than 3.38 to a supported version.
Note: 3.40 and lower is EOSL. Please upgrade to the latest SRS VE code 3.52. (SRS VE code is independent of the Unity OE version). -
If using the eVE (the SRS component ‘gateway’ that’s part of the Unity OE): Customers will need to upgrade their Dell EMC Unity OE if they are running less than 5.0.4. The eVE uses the VE codebase with some features removed and limits greatly reduced to cut down on memory requirements.
Once connectivity is lost, Dell will be unable to remotely monitor your system or remediate issues. Until the affected system is upgraded, be sure to diligently watch for alerts and open cases as needed. CloudIQ will not be available until the OE is upgraded. See KB 183761 for further information.
Dell EMC Unity XT Storage Processors running Unity OE 5.1.x or 5.2.0.x code
could unexpectedly reboot after ~275 days of uptime
”’Unity XT 480, 680 or 880 SP may experience an unexpected storage processor
reboot after being up for 275 – 300 days. This is due to a known integer
overflow calculation issue leading to a 64 bit result being stored in 32 bit
variable. This issue is most likely to be seen on a Unity XT 480, 680 or 880
array running 5.1.0.0.5.394 and newer code due to changes in that code and SP
hardware used in those models. Older codes and Unity models are much less
likely to see this issue. Please refer to KB 200921 for more information about
this issue which is currently fixed in Unity OS 5.2.1.0.5.013.
Protect your replicas after cache loss
Cache loss scenarios on Dell EMC Unity storage arrays are extremely rare. If
such a situation does occur, synchronous replication can allow for rapid
recovery and restoration of production services to the untainted replica, as
long as the integrity of the replicas are properly protected after the initial
data loss event. One important step that customers sometimes forget is that
all replications should be “admin fractured” before any recovery related
troubleshooting is attempted. The moment a cache loss event is experienced on
a production array, the pool and associated LUNs will go offline and will be
marked for recovery. All replicas will become “system fractured.” What
sometimes gets overlooked is that if support is allowed to perform
“recoveries” on the impacted cache loss LUNs without first “admin fracturing”
the replicas, then replication will automatically resume on those LUNs after
the recovery is completed. If the replication is allowed to re- sume in such a
manner after a recovery, then the replication software will copy over the
changes to the data performed by the recovery operation itself, thereby also
corrupting the replica. If there is the possibility that you may wish to
promote and use the replicas in such a scenario, it is important to “admin
fracture” those replications prior to any recovery operations taking place.
Did you know?
-
You can find answers to common Dell EMC Unity NDU related questions in knowledge base article https://www.dell.com/support/kbdoc/en-us/000204308
-
You can get to both the midrange Uptime Bulletin as well as the Uni- ty monthly newsletter by clicking on this link: Support for Dell EMC
Unity Family | Overview | Dell US, then select “Monthly Support Highlights” within “Featured Content.” -
The allowable correctable errors threshold per day on DIMMs has increased from 5k to 50k in Unity OE 5.2.1.x. This threshold will more closely align with PowerStore and will result in fewer DIMMs failures.
-
IF you have auto-tiering enabled on a Dell EMC Unity storage system, t can be good practice to disable it prior to performing an NDU.
UPTIME BULLETIN
Dell EMC Unity/SC/PowerScale/ME4/ME5 target versions
DELL EMC has established target revisions for each product to promote more
stable and reliable environments. As a best practice, DELL EMC recommends that
you operate at target code levels or above to benefit from the latest
enhancements and fixes available. Search using the term “adoption rates” in
http://support.emc.com for current Dell EMC Unity/SC/Isilon target code
adoption rates.
Dell EMC UNITY OE VERSION | RELEASE DATE | STATUS |
---|---|---|
5.2.1.0.5.013 | 7/28/2022 | Target |
5.2.1.0.5.013 | 7/28/2022 | Latest Release |
SCv20x0, SCv30x0, SC4020, SC5020,
SC7020,SC8000, SC9000 SCOS VERSION| RELEASE DATE| STATUS
07.05.02| 11/17/2021| Target
07.05.10| 11/10/2022| Latest Release
DSM 20.1.10.79| 8/19/2022| Latest Release
PowerScale OE VERSION| RELEASE DATE| STATUS
9.4.0.x| 10/12/2022| Target
9.4.0.10| 12/15/2022| Latest Release
ME4 OE VERSION| RELEASE DATE| STATUS
GT280R009-02| 7/1/2021| Target
GT280R010-01| 7/1/2022| Latest Release
ME5 OE VERSION| RELEASE DATE| STATUS
1.0.0.0| 2/2/2022| Target
1.0.1.0| 8/29/2022| Latest Release
See Product Release Notes for a full list of enhancements per new code
release.
Dell PowerScale OneFS OE enhancements in release 9.4.0.3
-
Continuation of STIG Requirements
-
FED/Security Enablement
-
Data mobility (SmartSync v1)
-
APEX enhancements
-
IB to Ethernet migration
Dell ME5 OE enhancements in release 1.0.1.0 -
Added support for ECO Lot 9 compatibility
-
Optional auto-map and auto-unmap fields are added to the CLI set advanced setting command. These settings are disabled by default
-
When enabled, auto-map will automatically map the host or initiators to the volumes or volume groups that are associated with the host group to which the hosts or nitiators are being added
-
When enabled, auto-unmap will automatically remove the mapping from a host or initiators when they are removed from the host group or initiator. This applies to mappings removed using the CLI or using PowerVault Manager
-
PowerScale latest code releas- es
-
The latest code releases for PowerScale can be found here: https://www.dell.com/support/manuals/en-us/isilon-onefs/ifs_pub_current_releases/powerscale-onefs-current-softwarereleases?guid=guid-dcb04cf4-42db-4cd0a0ae-3be10a4d1910&lang=en-us
-
The Latest Monthly Rollup (RUP) is typically recommended for PowerScale customers trying to stay the most up to date with their code. The latest recommend RUP is the August package SCOS enhancements in release 07.05.10
-
Fixes a bug in 7.5.x where a snapshot could get stuck and might cause a controller reset or volume accessibility issues because of IORPs witha null pointer
-
Fixes a bug in 7.5.x where a snapshot could get stuck and cause replication and volume accessibility issues ostly with Live Volumes when there is an Incomplete- WriteRepair process that is stuck
Recent enhancements in CloudIQ
Consolidated Systems View
We’ve consolidated the various Systems pages into a single view for quicker
navigation. You can now use the Systems page to switch between views for
Health, Inventory, Configuration, and Performance for each of your products.
Your selection within this view will persist between sessions. The current
filter options are also cached to facilitate the navigation between the multi-
systems view and details pages. We’ve also updated the Left Navigation bar to
consoli- date all of the multi-system views into an easy-to -navigate view.
APEX Data Storage Services Capacity views updates
We’ve enhanced the Capacity views for APEX systems to allow you to switch
between subscribed capacity for your APEX Service, or Physical capacity for
actual capacity consumed on the storage system. You can toggle between these
views through a radio button in the system details capacity view. In the
Subscribed capacity view, we have streamlined the capacity metrics to show
Base Used and Available, and where applicable, On-Demand Used. The Physical
ca- pacity view corresponds to the capacity displayed in the Element Manager
for the storage system. ‘
Create Report’ Button Replacing ‘Go To All Metrics’ Button
We’ve replaced the ‘Go To All Metrics’ button in Performance details pages
with the ‘Create Report’ button, which allows you to create a custom report
with detailed performance data. You can quickly create custom performance
reports and manipulate them by adding other content, including tables, and
also schedule and export reports. Existing ‘Go To All Metrics’ buttons have
been replaced, with the exception of PowerVault LUN, XtremIO system, and SC
Series system pages where the buttons have been removed for now.
Dell Inc. believes the information in this publication is accurate as of its
publication date. The information is subject to change without notice.
THE INFORMATION IN THIS PUBLICATION IS PROVIDED “AS IS.” DELL INC. MAKES NO
REPRESENTATIONS OR WARRANTIES OF ANY KIND WITH RESPECT TO THE INFORMATION IN
THIS PUBLICATION, AND SPECIFICALLY DISCLAIMS IMPLIED WARRANTIES OF
MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE.
Use, copying, and distribution of any Dell EMC software described in this
publication requires an applicable software license. Dell EMC and other
trademarks are trademarks of Dell, Inc. or its subsidiar-ies. Other trademarks
may be trademarks of their respective owners. Copyright © 2022 Dell EMC
Corporation. All rights reserved. Published in the USA, December, 2022.
WE’D LIKE YOUR FEEDBACK ABOUT THE UPTIME BULLETIN. SEND US YOUR IDEAS FOR
FUTURE TOPICS
AT : dan.gauthier@dell.com
Unity Customer
Documentation https://www.dell.com/support/home/en-us/product-
support/product/unity-all-flash-family/docs
Documents / Resources
| DELL
EMC SC9000 Storage
Array
[pdf] User Guide
SC9000 Storage Array, SC9000, Storage Array
---|---
References
- Support | Dell US
- Support | Dell US
- Manuals | Dell US
- jira.gtie.dell.com/browse/PSE-18611
- Support for Dell EMC Unity Family |Dell EMC Unity All Flash | Documentation | Dell US
- Support for Dell EMC Unity Family | Overview | Dell US
- Login or Register | Dell US
- Dell Unity: SP may panic after being up for 275 - 300 days (Dell Correctable) | Dell US
- Login or Register | Dell US
- Dell Unity: Answers to common Unity Upgrade (NDU) Related Questions (User Correctable) | Dell US
- Login or Register | Dell US
- PowerScale OneFS Current Software Releases | Dell US