Isilon

OneFS
Version 8.1.2

Release Notes
Copyright © 2018 Dell Inc. or its subsidiaries. All rights reserved.

Published August 2018

Dell 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 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 SOFTWARE DESCRIBED
IN THIS PUBLICATION REQUIRES AN APPLICABLE SOFTWARE LICENSE.

Dell, EMC, and other trademarks are trademarks of Dell Inc. or its subsidiaries. Other trademarks may be the property of their respective owners.
Published in the USA.

EMC Corporation
Hopkinton, Massachusetts 01748-9103
1-508-435-1000 In North America 1-866-464-7381
www.EMC.com

2 OneFS 8.1.2 Release Notes

CONTENTS

Chapter 1 OneFS release and upgrade information 5

OneFS 8.1.2 release notes............................................................................ 6
EMC technical advisories, security advisories, and OneFS patches.............. 6
Upgrading OneFS......................................................................................... 6
Supported upgrade paths for OneFS 8.1.2.................................................... 7

Chapter 2 Modifications and enhancements 11

Backup, recovery, and snapshots................................................................ 12
Diagnostic Tools..........................................................................................12
Hardware.................................................................................................... 12
HDFS.......................................................................................................... 12
Upgrade...................................................................................................... 13

Chapter 3 Resolved issues 15

Antivirus......................................................................................................16
Authentication............................................................................................ 16
Backup, recovery, and snapshots................................................................ 17
CLI.............................................................................................................. 18
Cluster configuration.................................................................................. 19
Diagnostics................................................................................................. 19
Events and alerts........................................................................................ 19
File System.................................................................................................20
File Transfer................................................................................................ 21
Hardware.................................................................................................... 21
HDFS..........................................................................................................22
Job engine.................................................................................................. 23
Networking.................................................................................................23
NFS............................................................................................................ 24
OneFS API..................................................................................................25
OneFS web administration interface...........................................................25
SMB........................................................................................................... 26
Security...................................................................................................... 27
Upgrade......................................................................................................28

Chapter 4 Known issues 29

Backup, recovery, and snapshots............................................................... 30
Events, alerts, and cluster monitoring known issues...................................30
File system................................................................................................. 30
Hardware....................................................................................................30
HDFS.......................................................................................................... 31
Security...................................................................................................... 31

Chapter 5 OneFS Release Resources 33

OneFS information and documentation...................................................... 34
Where to go for support............................................................................. 35
Provide feedback about this document...................................................... 35

OneFS 8.1.2 Release Notes 3

CONTENTS

4 OneFS 8.1.2 Release Notes

CHAPTER 1
OneFS release and upgrade information

This chapter provides an overview of the information covered in the OneFS 8.1.2
release notes. This chapter also describes the OneFS upgrades and supported upgrade
paths.

l OneFS 8.1.2 release notes....................................................................................6

l EMC technical advisories, security advisories, and OneFS patches......................6
l Upgrading OneFS.................................................................................................6
l Supported upgrade paths for OneFS 8.1.2........................................................... 7

OneFS release and upgrade information 5

OneFS release and upgrade information

OneFS 8.1.2 release notes

This document contains descriptions of the new features, modifications,
enhancements, resolved issues, and known issues that are included in OneFS 8.1.2.
The new features, modifications, enhancements, resolved issues, and known issues
listed in the release notes are categorized by functional area, such as Authentication,
Events and alerts, File system, and Networking.
For a list of available OneFS releases, and for information about Target Code releases
and general availability (GA) releases, see Current Isilon Software Releases on the
EMC Online Support site.

EMC technical advisories, security advisories, and OneFS

patches
EMC technical advisories (ETAs), EMC security advisories (ESAs), and OneFS
patches are made available on the EMC Online Support site to provide important
information about and solutions to issues that affect the OneFS operating system.
Technical advisories
For the most up-to-date list of Isilon ETAs, see the EMC Technical Advisories (ETAs)
for Isilon OneFS page on the EMC Isilon Community Network site.
To subscribe to receive email notifications about ETAs, visit the EMC Isilon OneFS
product page on the EMC Isilon Support site and then click the Manage Advisory
Subscriptions link under Advisories.
Security advisories
For the most up-to-date list of Isilon ESAs, see the EMC Isilon Security Advisories
(ESAs) page on the EMC Isilon Community Network site.
To subscribe to receive email notifications about ETAs, visit the EMC Isilon OneFS
product page on the EMC Isilon Support site and then click the Manage Advisory
Subscriptions link under Advisories.
OneFS patches
For the most up-to-date list of patches that are available for the version of OneFS
running on your cluster, see the Current Patches document on the EMC Online
Support site.

Upgrading OneFS
OneFS upgrades comprise a full operating system upgrade and require that the Isilon
cluster be rebooted. Depending on the version of OneFS from which you are
upgrading, the upgrade might require a simultaneous reboot of the cluster.
Before upgrading OneFS, review the Supported upgrade paths section of this
document to verify that the cluster can be upgraded from your current version of
OneFS directly to this release.
To download the installer for this release, see the OneFS Downloads page on the EMC
Online Support site.

6 OneFS 8.1.2 Release Notes

 OneFS release and upgrade information

Upgrade resources
For more information about simultaneous and rolling upgrades and other important
details about the OneFS upgrade process—including recommended pre-upgrade and
post-upgrade tasks—see the OneFS Upgrades - Isilon Info Hub on the EMC.com web
site.

Supported upgrade paths for OneFS 8.1.2

To ensure that the version of OneFS you are upgrading to contains all the fixes in the
version of OneFS you are upgrading from, upgrades are supported only from specified
versions of OneFS.

Note

Following an upgrade to OneFS 8.1.0.0 or later, you must update your OneFS license
by generating a license activation file and submitting it to Software Licensing Central
(SLC). For instructions on how to create a new license file for your OneFS cluster,
refer to the Licensing section of the OneFS Administration Guide.

Upgrades from OneFS 8.1.2 to later versions of OneFS 8.1.2

The supported upgrade paths from earlier OneFS 8.1.2 releases to later OneFS 8.1.2
releases are provided in the following table.
Rolling and simultaneous upgrades are supported for all upgrades from OneFS 8.1.2 to
later versions of OneFS 8.1.2. Rolling upgrades are performed by default.

If your cluster is running this version of You can upgrade to this version of
OneFS OneFS
8.1.2.0 future releases

Upgrades from OneFS 8.1.1 to OneFS 8.1.2

The supported upgrade paths from OneFS 8.1.1 to OneFS 8.1.2 releases are provided
in the following table.
Rolling and simultaneous upgrades are supported for all upgrades from OneFS 8.1.1 to
OneFS 8.1.2. Rolling upgrades are performed by default.

If your cluster is running this version of You can upgrade to this version of
OneFS OneFS
8.1.1.0, 8.1.1.1 8.1.2.0

Upgrades from OneFS 8.1.0 to OneFS 8.1.2

The supported upgrade paths from OneFS 8.1.0 to OneFS 8.1.2 releases are provided
in the following table.
Rolling and simultaneous upgrades are supported for all upgrades from OneFS 8.1.0 to
OneFS 8.1.2. Rolling upgrades are performed by default.

If your cluster is running this version of You can upgrade to this version of
OneFS OneFS
8.1.0.0 through 8.1.0.4 8.1.2.0

Upgrades from OneFS 8.0.1 to OneFS 8.1.2

The supported upgrade paths from OneFS 8.0.1 to OneFS 8.1.2 releases are provided
in the following table.

Supported upgrade paths for OneFS 8.1.2 7

OneFS release and upgrade information

Rolling and simultaneous upgrades are supported for all upgrades from OneFS 8.0.1 to
OneFS 8.1.2. Rolling upgrades are performed by default.

If your cluster is running this version of You can upgrade to this version of
OneFS OneFS
8.0.1.0 through 8.0.1.2 8.1.2.0

Upgrades from OneFS 8.0.0 to OneFS 8.1.2

The supported upgrade paths from OneFS 8.0.0 to OneFS 8.1.2 are provided in the
following table.
Rolling and simultaneous upgrades are supported for all upgrades from OneFS 8.0.0 to
OneFS 8.1.2. Rolling upgrades are performed by default.

If your cluster is running this version of You can upgrade to this version of
OneFS OneFS
8.0.0.0 through 8.0.0.7 8.1.2.0

Upgrades from OneFS 7.2.1 to OneFS 8.1.2

The supported upgrade paths from OneFS 7.2.1 to OneFS 8.1.2 are provided in the
following table.
Only simultaneous upgrades are supported for upgrades from OneFS 7.2.1 to OneFS
8.1.2.

If your cluster is running this version of You can upgrade to this version of
OneFS OneFS
7.2.1.0 through 7.2.1.6 8.1.2.0

Upgrades from OneFS 7.2.0 to OneFS 8.1.2

The supported upgrade paths from OneFS 7.2.0 to OneFS 8.1.2 are provided in the
following table.
Only simultaneous upgrades are supported for upgrades from OneFS 7.2.0 to OneFS
8.1.2. Direct upgrades to OneFS 8.1.2 are not supported from OneFS 7.2.0.0 or OneFS
7.2.0.1

If your cluster is running this version of You can upgrade to this version of
OneFS OneFS
7.2.0.2 through 7.2.0.5 8.1.2.0

Upgrades from OneFS 7.1.1 to OneFS 8.1.2

The supported upgrade paths from OneFS 7.1.1 to OneFS 8.1.2 are provided in the
following table.
Only simultaneous upgrades are supported for upgrades from OneFS 7.1.1 to OneFS
8.1.2. Direct upgrades to OneFS 8.1.2 are not supported from OneFS 7.1.1.0 through
OneFS 7.1.1.10.

If your cluster is running this version of You can upgrade to this version of
OneFS OneFS
7.1.1.11 8.1.2.0

8 OneFS 8.1.2 Release Notes

 OneFS release and upgrade information

Upgrades from OneFS 7.1.0 or OneFS 7.0.x to OneFS 8.1.2

Direct upgrades to OneFS 8.1.2 are not supported from OneFS 7.0.x or OneFS 7.1.0.x.
To upgrade to OneFS 8.1.2 from these OneFS versions, you must upgrade twice. First
upgrade to a version of OneFS that supports upgrades to OneFS 8.1.2, and then
upgrade to the target version of OneFS 8.1.2. For example, if the cluster is running
OneFS 7.1.0.1, you can upgrade to OneFS 7.1.1.11, and then upgrade to OneFS 8.1.2.0.

Note

Intermediate upgrades must be treated like full upgrades, with all the recommended
planning and preparations that apply. If you must upgrade to an intermediate version
of OneFS before you can upgrade to the target version, complete the pre-upgrade and
post-upgrade steps in the OneFS Upgrade Planning and Process Guide first for the
intermediate upgrade, and then complete the pre-upgrade and post-upgrade steps
again for the upgrade to the target version. Follow the instructions in the version of
the OneFS Upgrade Planning and Process Guide that matches the OneFS version that
you are upgrading from. For more information, see the OneFS Upgrades - Isilon Info
Hub.

Supported upgrade paths for OneFS 8.1.2 9

OneFS release and upgrade information

10 OneFS 8.1.2 Release Notes

CHAPTER 2
Modifications and enhancements

This chapter describes the modifications and enhancements to OneFS for this release.
Modifications and enhancements include:
l Changes to the formatting or syntax of an existing command.
l The addition of parameters to an existing command.
l Changes to underlying functionality to improve performance.
l Updates to integrated OneFS components such as OpenSSL and GNU bash.
l Changes to enable existing functionality to be configured or controlled through
additional interfaces or modes.
l Changes to remove support for deprecated protocols and embedded components.

l Backup, recovery, and snapshots....................................................................... 12

l Diagnostic Tools................................................................................................. 12
l Hardware............................................................................................................ 12
l HDFS.................................................................................................................. 12
l Upgrade..............................................................................................................13

Modifications and enhancements 11

Modifications and enhancements

Backup, recovery, and snapshots

226775
This bug introduces a new SyncIQ global setting (--use-workers-per-node,
Boolean). The setting is on by default on upgrades from versions of OneFS
previous to 8.0. For new installations or existing 8.0 and later installation, the
setting is off by default. When the status is on, the SyncIQ considers the workers
per node setting in the policy configuration as part of the worker distribution
model. On startup, a SyncIQ job compares the computed total number of workers
using the -workers-per-node settings with the worker allocation that is computed
across all SyncIQ jobs and uses the lesser value.

Diagnostic Tools
216078
The default mechanism to upload the output of the isi_gather_info
command is now EMC Secure Remote Services (ESRS). This mechanism
provides better security and upload speed. You can suppress this behavior with
the --noesrs flag.

Hardware
213544
A new --refresh flag is introduced to forcibly update the cached firmware that
is returned by the isi upgrade firmware devices command. You can use
this flag when there is discrepancy in the cached output of isi upgrade
firmware devices command due to either hot-swapping of the power supply
or malfunctioning of some device.

HDFS
Support for Hadoop 3
Isilon OneFS 8.1.2 is compatible with HDFS, Hortonworks Data Platform (HDP),
Cloudera Distribution including Apache Hadoop (CDH), and Apache stacks with all
currently supported workloads, including MapReduce, Spark, and Hive.

Support for the Isilon Ambari Management Pack

The Isilon Ambari Management Pack simplifies managing your Isilon storage and
cluster options using an agentless integration approach. The Isilon Ambari
Management pack is not backward compatible with HDP 2.x with Ambari.

Support for WebHDFS

WebHDFS provides a RESTFUL API for accessing HDFS datasets on Isilon
clusters. WebHDFS support enables the use of Knox with Isilon OneFS 8.1.2 with
native WebHDFS integration and enables the use of Ambari Views with Isilon in a
Kerberized setup. The use of Knox and Ambari Views in Kerberized clusters are
key security and management features for Hortonworks.

12 OneFS 8.1.2 Release Notes

 Modifications and enhancements

Support for LLAP

Live Long and Process (LLAP, also Low Latency Analytical Processing) enables
HDP customers to run interactive SQL queries with Hive2. Isilon OneFS 8.1.2
supports reference by inode, which enables LLAP to work with the HDFS
filesystem. Reference by inode allows the HDFS client to get the inode (Hadoop
LIN) for a file or directory and then use that inode in future queries to ensure that
the same resource is accessed regardless of renaming, moves, and so on.

Support for SSL with Ranger

Isilon OneFS 8.1.2 adds support for SSL/HTTPS to enable securing all
communication between the Ranger policy manager and the plugin with the use
of 1-way SSL.

Upgrade
215871
You can now use the isi upgrade cluster rolling-reboot command to
reboot nodes in a rolling manner, one node at a time.
The isi upgrade cluster rolling-reboot command includes the
following options, where x and y are node numbers:
--nodes=x,y
This option is used to perform a rolling reboot of the nodes listed, in the order
listed. Each node must be separated by a comma.

--nodes=x-y
This option is used to perform a rolling reboot of all nodes numerically
between the first node listed and the last node listed.

--nodes=all
This option is used to perform a rolling reboot of all nodes in the cluster.

Upgrade 13
Modifications and enhancements

14 OneFS 8.1.2 Release Notes

CHAPTER 3
Resolved issues

This chapter contains descriptions of issues that are resolved in this release.

l Antivirus............................................................................................................. 16
l Authentication.................................................................................................... 16
l Backup, recovery, and snapshots........................................................................17
l CLI......................................................................................................................18
l Cluster configuration.......................................................................................... 19
l Diagnostics......................................................................................................... 19
l Events and alerts................................................................................................ 19
l File System........................................................................................................ 20
l File Transfer....................................................................................................... 21
l Hardware............................................................................................................ 21
l HDFS................................................................................................................. 22
l Job engine..........................................................................................................23
l Networking........................................................................................................ 23
l NFS....................................................................................................................24
l OneFS API......................................................................................................... 25
l OneFS web administration interface.................................................................. 25
l SMB...................................................................................................................26
l Security..............................................................................................................27
l Upgrade............................................................................................................. 28

Resolved issues 15
Resolved issues

Antivirus
214825
If an AVScan job was in progress, then the scheduled antivirus policy that ran
during the same timeframe would go into a Pending status. The antivirus policy
would not resume even when the former AVScan job was finished.
206825
If the AVScan report code ran when the date in the OneFS cluster was set to
around 2038 or later, AVScan would truncate timestamps in its SQLite database.

206557
You might have received false Clusterwide Event Log (CELOG) event alerts for
the antivirus scan-on-close policy.

Authentication
216822
If you created a home directory share, but that was targeted to a read-only file
system, then you could not manage shares through the Windows Computer
Management. As a result, the following error message appeared:

Error 31: A device attached to the system is not functioning

215732
When the OneFS cluster operated in a multi-zone environment with two zones
having Microsoft Domain Controllers at different patch levels for the MS15-027
patch, the OneFS cluster might not have allowed you to map shares.

213624
If you used passwordless Secure Shell (SSH) to access the cluster, the
ISI_PRIV_IFS_BACKUP or ISI_PRIV_IFS_RESTORE privileges did not work. As a
result of not having these privileges, you were unable to access the files.

212977
When listing the access zones, if for some reason, the user lookup timed out (for
example, domain was offline), then the entire operation failed with no information
regarding what went wrong.

204699
If you tried to create an SMB share and browse to the Select Persona window to
add a user, incorrect Active Directory (AD) user search results were found. This
issue occurred only when both the User Name and User Description fields were
populated.

195592
When over 400 clients tried to connect to a OneFS node through FTP, additional
clients failed to connect, and the following error message was logged to
the /var/log/messages file:

pam_sm_authenticate error [login:root][error code:40185]

16 OneFS 8.1.2 Release Notes

 Resolved issues

Backup, recovery, and snapshots

218698
If a SyncIQ secondary worker disconnected from its peer, the secondary worker
might have continued to process work resulting in a race condition with the other
SyncIQ secondary workers. To address this issue, work items will now be locked
to prevent the race condition.

215880
If a SyncIQ job restarted in the WORK_CC_DIR_CHGS phase and a SyncIQ job
processed deleted directories, then the added or deleted files could be missed
during the WORK_CC_DIR_CHGS phase. This would cause the SyncIQ job to later
fail in a link transfer phase. If deleted files were missed, then a message similar to
the following appeared in the /var/log/isi_migrate.log file: Error :
Unable to open Lin [LIN]: No such file or directory. If added
files were missed, then a message similar to the following appeared in
the /var/log/isi_migrate.log file: FAILED ASSERTION found ==
true.

215632
SyncIQ used a single root temporary directory to handle the temporary files
during SyncIQ jobs. This could have led to the filesystem contention on the root
temporary directory during the SyncIQ job link phases.

215342
The incremental backup sessions slowed down and the log messages were
swamped with failed - message too long error messages.

214886
If a SyncIQ job with subnet or pool rules was paused for more than 5 minutes, the
job would fail.

213306
A SnapshotDelete job was interrupted by a MediaScan job, leaving some STF
delta blocks in an inconsistent state. If you tried to manually delete the snapshots,
the following error message appeared:

Stale NFS file handle

213021
Depending on the timing, inconsistent duration values were calculated for
snapshot schedules. The durations were typically off by one second, although in
some cases it could have been more.

212899
If a policy reset or an upgrade sync was initiated for a SyncIQ policy after the
resync preparation, then the subsequent SyncIQ job could fail. When this

Backup, recovery, and snapshots 17

Resolved issues

occurred the following message appeared in /var/log/messages

and /var/log/isi_migrate.log files:

FAILED ASSERTION !!(global->rep_mirror_ctx) ==!!(global-

>cpss_mirror_ctx)

208880
Using CloudPools, if you tried to back up large files that were smartlinked files,
the backup throughput might have been extremely low.
205071
Duplicate entries for snapshot schedules appeared in the output of the isi
snapshot schedules list command. The OneFS web administration
interface showed individual schedules but the schedule details were duplicated.

Note

You must manually delete all the duplicate snapshot schedules that might have
been previously created.

203936
During the SyncIQ failover or failback operation, the clients could get
disconnected unexpectedly.

201601
It was possible for a successful DomainMark job to leave a SyncIQ domain or a
Snaprevert domain incomplete. If this occurred, the SnapRevert job, which might
run during the SyncIQ Prepare Resync job phase failed and the following status
message appeared in the SyncIQ job report:

Snapshot restore domain is not ready (unrunnable).

197240
During SyncIQ jobs, temporary directories were created and removed for each
quota root directory nested below the SyncIQ target root path, even when there
were no changes within a quota path. This operation could have caused the jobs
to run longer in the cleanup phase.

194659
Specifying an expiration date that extends beyond January 19, 2038 no longer
overflows the allowable range or wraps around to the minimum date of January 1,
1970 UTC.

CLI
201152
The atime command under the isi cluster command branch was missing. As a
result, you could not set or view the access time grace period values through the
command-line interface.

18 OneFS 8.1.2 Release Notes

 Resolved issues

Cluster configuration
218549
An alert that the cluster is not licensed for SED drives is no longer generated after
the license file is added.
217022
During an upgrade from OneFS 8.0.0.0 to 8.1.0.0 or later builds, the drive
repurposing daemon might have inadvertently tried an L3 migration on the drives
that were already correctly formatted.

216768
When background tasks and time-based operations were performed on files that
were deleted after the task was queued, the CloudPool daemon log files were
spammed.

216410
Inefficient checkpoint implementation in the CloudPool10 writeback might have
caused performance issues.
215937
An unrecoverable isi_cpool_d failure occurred when a copy writeback job,
resuming from a checkpoint came across an uninitialized pointer.

210679
The command-line interface (CLI) command for listing files of a particular
CloudPools job could handle only the first 100000 files. The file count restriction
is now removed.

Diagnostics
215162
The DomainMark job failed if the target directory contained the WORM Ancestor
attribute for the deleted domain.

Events and alerts

217932
Clusterwide Event Log (CELOG) sent Simple Network Management Protocol
(SNMP) traps no longer has theeventKbUrl varBind set to Unavailable.
Although, eventKbUrl Unavailable varBinds might occur if no documentation is
available for the event ID that is listed as part of the causes.

210971
If you tried to dump entries from the split changelist, the isi_changelist_mod
command failed and the following error message appeared:
Changelist metadata not found.

Cluster configuration 19
Resolved issues

207825
Multiple Simple Network Management Protocol (SNMP) traps and inform
varBind, Object Identifier (OID) numbering, trap severity level, trap conformance
group, and Clusterwide Event Log (CELOG) associated issues have now been
addressed. Additionally, location information to the unsupported boot disk event
(100010041) and quota domain with quota threshold violation alerts (500010001)
were added.

194358
Domain quotas are now enforced, and temporary celog and event files and
directories no longer accumulate.

147392
The audit_protocol.log file displayed inaccurate timestamps for audit
logging that did not match the actual audit events' time.

File System
219862
The SYS_DISK_SSDFULL event is now disabled to avoid excessive spamming
when the disks in all the nodes of the cluster are exclusively solid-state drives
(SSDs).

217654
The clusters with Brazil time zones would have had incorrect time beginning from
21st October, 2018. To avoid the time zone error, the data is updated.

217644
The nodes had insufficient number of worker threads to process Remote Block
Management (RBM) messages and slowed down the RBM message processing.
This issue caused ping timeouts and disconnects, which resulted in node splits.

216394
The SMB permission changes that were made using the Windows File Manager
were significantly slower in OneFS 8.0.1.1 than those made using versions of
OneFS earlier than OneFS 8.0.0.x.

215812
An underlying issue of isi_drive_d added preformatted drive on booting and
caused Isilon Data Integrity (IDI) failure on Node State Block (NSB). As a result,
you could get an Isilon Data Integrity (IDI) alert under the /var/log/idi.log
file, such as the one shown here:
2018-01-05T13:58:21+11:00 <29.3> isilon-27 /boot/
kernel.amd64/kernel: [ | IDI_VERIFY=idi_error_msg:1307 |
COND=idi_error | MSG=Detected IDI failure (type Node State
Block), attempting DSR | BADDR=166,35,172032:8192 | OWNER=
| ]

214175
The isi_hangdump command did not have enough failover information. An
improvement is made to dump more lock information.

20 OneFS 8.1.2 Release Notes

 Resolved issues

211063
If 18 or more files were concurrently written to one SMB share by one or more
SMB clients, OneFS might send an unexpectedly large number of change
notifications to the Requester(s). If this issue occurred, any workflow relying on
the change notifications for proper processing might have failed and required a
restart.

208497
Rotated versions of a single log file were arbitrarily compressed or uncompressed
using the wrong scheme.

202334
Although there were no errors, the isi_gather_info --esrs command
resulted in a large number of * SSLv2, Unknown (23) * warning messages.

201480
The Generation 6 nodes failed to boot and displayed the following error message:

panic: No bootable partitions found!

File Transfer
217266
When the listener on Apache HTTP server did not shut down completely, the PID
file was removed but the service was still up and accepting requests. It created a
problem between different services where the logs no longer rotated. The HTTP
service could continually run commands and it added to the logs. The boot drive
became completely filled up and the logging throughout the system failed.

190375
FTPS clients now correctly negotiate TLSv1.1 connections.

Hardware
218817
If the Generation 6 node initiating a firmware update was not in chassis slot 1, the
firmware updates to the fpinffw and fpioc devices failed.

217228
After a power supply replacement, the output of isi upgrade firmware
devices might have listed the old supply that is no longer present.

216765
Adding a node to the cluster might have produced a socket error and failed to
complete the add process. The underlying processes for the socket use are
improved to prevent the error.

216269
When a drive containing a boot mirror failed, it created a
SYS_DISK_BOOT_MIRROR_FAILURE event even though the mirror was
successfully rebalanced a moment later.

File Transfer 21
Resolved issues

215557
The HW_INFINITY_NTB_LINK_OUTAGE event occurred during the expected
maintenance scenarios where the node rebooted. The
HW_INFINITY_NTB_LINK_OUTAGE event is now disabled and the
HW_INFINITY_JM_STATE_ERROR is created when a Generation 6 peer node is
down for more than 4 hours.

215384
The F800 nodes failed to boot after replacing the 10GbE with 40GbE and
displayed the following error message:

Verify Hardware exited with an error

Validate the backend interfaces against the intended configuration. If there is a

mismatch, the boot process stops. In all such cases, contact Technical support
for assistance.

214156
The hwmon did not correctly decode the memory Error Correction Code (ECC)
events on Generation 5 models. The related Clusterwide Event Log (CELOG)
events on Generation 5 now correctly identify the location of the problem in dual
inline memory module (DIMM).

211315
Clariion Message Interface (CMI) watchdog occurrence is no longer logged if the
Non -Transparent Bridge (NTB) link is down with the node pair.

204914
To reduce flash wear and extend the usability of the boot flash drives used in
Generation 4 and Generation 5 storage nodes, the var partitions are now
dynamically migrated onto the data drives.

181665
A more descriptive console message with the drive info, and the bootdisk ID has
been added to describe mirror boot failures.

HDFS
221832
HDFS block reads and writes are now only logged when DEBUG is set to ON to
limit log file size.

211207
Running multiple Spark threads from a client resulted in MAC address
mismatches and Spark job failures if HDFS wire encryption was enabled.

190069
The Ambari Agent failed with unhandled exclusive lock exception errors in the
destructor __clang_call_terminate operation if multiple access zones
were configured running multiple Ambari agent fiber tasks.

22 OneFS 8.1.2 Release Notes

 Resolved issues

Job engine
214129
The isi job reports and the OneFS web interface contained incorrect values
for processed logical i-nodes (LINs) and shadow identification numbers (SINs).
These incorrect values however, did not affect the job functions.

210726
The deduplication job failed with the following error message:
isi_job_d[68839]: [MAJOR] dedupe
mli_index_level_load_bucket: Could not flush untouched
buckets to checkpoint: Unsupported Operation

210388
OneFS provides reserved storage blocks for jobs that must run even when a
cluster is nearly full (that is jobs that create net free space). However, all jobs
could access these blocks, thus impacting the user performance. Access to
reserved blocks is now restricted to the following jobs:
l Collect
l SnapshotDelete
l TreeDelete
l ShadowStoreDelete

205476
Files that had the SF_APPEND flag set, could not have had the
dedupe_was_sampled flag set on them. As a result, the deduplication job
reported errors for each file:

Dedupe[1373] Node 3 (3) task 4-33: Failed to open file 1:4a2a:

09f0: Operation not permitted

197854
Job Engine did not clean up old checkpoint directories. Every time that a job
engine job ran, it created a directory and subdirectories under the /
ifs/.ifsvar/modules/jobengine/cp file. Over a period of time, the
checkpoint directory got filled with many subdirectories. While the number of
subdirectories present did not pose a problem to normal cluster operation, they
could delay isi_gather_info collections.

Networking
222433
When some ports of a lagg interface in Link Aggregation Control Protocol (LACP)
mode went down, OneFS displayed No Carrier status on lagg interfaces,
although there were other ports that were still active. This error might have
resulted in losing connection to the node from outside of the cluster.

215106
The isi_eth_mixer_d daemon uses ping to check state. If pinging a backend
interface resulted in an error, this daemon could have missed that error and not

Job engine 23
Resolved issues

set the node to use the other backend interface. As a result, you could have faced
traffic and daemon failures.

211952
The default values that were previously set for the Sockets Direct Protocol's
(SDP's) time_wait_timeout and fin_wait_timeout parameters were
excessive for a low-latency InfiniBand fabric. In the event of a cluster split, these
values caused connection resources to be held by closed connections for a long
period, and in extreme cases, prevented new connections from being created.

210689
If you tried to use the lagg failover mode, neither Gratuitous ARP (IPv4) nor
Unsolicited Neighbor Advertisements (IPv6) were sent to notify the other nodes
that the MAC address might have moved. This issue was caused as the master
port became unavailable and the next active port got stimulated. Ultimately, due
to OneFS not sending Gratuitous ARP (IPv4) or Unsolicited Neighbor
Advertisements (IPv6), you could temporarily lose connection to the OneFS node
when the failover occurred on the aggregation (lagg) port.

207937
This issue has been fixed to enable UCD-SNMP-MIB value monitoring after
upgrade.
201989
The nodes with Intel 10GbE network interface cards could have encountered
system failure while reconfiguring the network parameters when the 802.1q VLAN
tagging was enabled.

NFS
217441
If an error occurred while opening a file over NFSv4, the NFS service might not
have incremented the sequence ID and then might have rejected requests to
close a file. If this happened, a state ID table could have filled up and caused data
to become unavailable, with the following message in the /var/log/nfs.log
file:
[nfs] Nfs4 stateid table full!

216566
When attempting to move a file from a snapshot, the following assertion appeared
in the nfs.log file and the NFS container restarted indefinitely while the client
retried the operation.

[lwio] ASSERTION FAILED: Expression = (pWork-

>rFhaFile1.rInfo.Fileno < pWork->rFhaFile2.rInfo.Fileno).

209840
When you mount newer NFS clients like Centos7.4 to the Isilon cluster without
specifying the version, it could fail.

24 OneFS 8.1.2 Release Notes

 Resolved issues

197448
When listing a large number of aliases, the isi nfs alias list --check
command returned an internal STATUS_UNSUCCESSFUL error.

OneFS API
215624
The snapshot alias API now returns 18446744073709551615 instead of -1 to the
live filesystem against an alias.

OneFS web administration interface

217191
When you tried to upload an upgrade tar file that was stored in a special character
directory (for example, /ifs/ /'"~!@#$%^&*()_+-=[]{}.,<>?\|), the Select a
Path to a File dialog box displayed a RAN error:
Unable to list directory contents
Unable to open object ""~!@' in store 'ifs' -- not found.

216405
Running a query to get the number of TCP connections using the isi_stats_d
command, showed more number of connections than what was displayed in the
OneFS web interface. The OneFS web interface showed only the established TCP
connections while the isi_stats_d command showed all the TCP connections.

215122
If you navigated to the Groups page in the OneFS web interface and tried to
delete a group on the second page of non-System zones, the group did not get
deleted. An error message similar to the following was displayed:

Group was not deleted

The group did not delete due to the following error:
Failed to find group for 'GROUP:testgroup6': No such group

214983
If you tried to create a subnet pool using the Create Pool page in the OneFS web
interface before creating access zones for newly created groupnet, the Access
Zone drop-down list did not work. As a result, the pool creation failed.

214372
The Patches and Firmware page in the OneFS web interface displayed that all
nodes ran the same firmware even if they did not. As a result, you could not get
reliable data and check the firmware status from the OneFS web interface.

209935
The SMB share creation form in the OneFS web administration interface ignored
the default SMB share settings for file filtering. However, if you used command-
line interface (CLI) to create SMB shares, the file filter settings were set
appropriately.

OneFS API 25
Resolved issues

209483
If you had the ISI_PRIV_SMB privilege, the simple user picker window did not
appear when you tried to modify SMB shares' personas.

207099
If you selected any option from the Notification Batch Mode drop-down list and
saved the change, the setting did not display correctly on the email alert channel
from the OneFS web interface. The following message appeared even though
batching was configured:
No batching

197466
The soft and hard storage quota limits in the OneFS web interface were restricted
to only integer values and decimal values were not allowed.

SMB
215800
While accessing the share through SMB protocol, you could not change a file
name from lowercase text to uppercase text or to any other non-English
characters. As a result, the application might have failed due to its case-sensitive
property.

215687
If a combination of Apple extensions were enabled and the server-side copy
(SSC) was disabled, the Mac OS file that was copied through Finder failed. When
you tried to copy a file from one directory to another, the following error message
might have appeared:
The operation can't be completed because an unexpected
error occurred (error code -50).

209461
The new remote procedure call (RPC) error code: RPC_S_BUFFER_TOO_SMALL
(0x0000007A) is now implemented in the RpcMgmtInqServerPrincipalName
function of the Witness server.

208617
OneFS might have incorrectly modified the atime and mtime on files in Mac OS.

204465
If you used a limit argument with isi smb openfiles and isi smb
sessions PAPI calls, a resume argument was returned, allowing pagination. This
resume argument was returned incorrectly, and caused the last object on a page
to be duplicated on the next page.

201274
In transparent failover scenarios, where many witness request cancellations
accompanied the witness protocol traffic, the bug in the cancellation handling
routines would render the witness server unresponsive. However, the server
falsely appeared as being up and running. Due to the lack of immediate response
from the witness server, the failover took about 20 s resulting in non-transparent
failover.

26 OneFS 8.1.2 Release Notes

 Resolved issues

197783
Mac OS clients were unable to view files within SMB shares when Apple
Extensions support was enabled (EnableAppleExtensions = 1). Additionally, they
were unable to drag (copy) files to the directories that were affected.

Security
233800
NTP was updated to version 4.2.8p11.

229966
TLS v1.0 was disabled in the OneFS Apache configuration.

223697
Updates Libevent to version 2-2.1.8_2.

222536
The version of OpenSSL installed on the cluster was updated to OpenSSL 1.0.2o-
fips. This version addresses multiple CVEs and bugs. For more information about
this Common Vulnerability and Exposure (CVE), see ESA-2017-147 on the
Customer Support website.

213123
OneFS was updated to mitigate the following known variant of speculative
execution side-channel vulnerability: Spectre variant 1 (CVE-2017-5753: Bounds
check bypass. For more information about this Common Vulnerability and
Exposure (CVE), see DSA-2018-098 on the Customer Support website.

209355
The following curl issues have been resolved in OneFS:
l FILE buffer was reading out of bounds. If curl requested a file from a file://
URL, it was possible for the wrong buffer to be sent to the user, which could
have led to private data to be inadvertently displayed.
l TFTP was sending more than buffer size. During a TFTP transfer, if curl was
given a URL that contained a very long file name (greater than 515 bytes), the
file name was truncated to fit within the buffer boundaries. If this issue
occurred, it was possible for a malicious HTTP(S) server to redirect a
vulnerable client and collect private data to a remote server over UDP.
l URL globbing was reading out of bounds. Due to an omission in the globbing
function, curl was reading a byte beyond the end of the URL. As a result, the
URL could have been made to wrong read something else instead of
restarting.

219688
OneFS was modified to disable TLS_RSA ciphers in the Apache configuration to
provide a more secure set of ciphers, and to remediate a ROBOT vulnerability
that scanners would potentially flag as a false positive. For more information
about this Common Vulnerability and Exposure (CVE), see DSA-2018-018 on the
Customer Support website.

Security 27
Resolved issues

202320
The version of perl installed on OneFS was updated to 5.26.1. For more
information about this Common Vulnerability and Exposure (CVE), see
DSA-2018-104 on the Customer Support website.

95517
OneFS was modified to update to Apache 2.4.

Upgrade
215330
If you tried to upgrade a OneFS cluster, the upgrade process got stuck due to the
isi_tardis core dump issue.

28 OneFS 8.1.2 Release Notes

CHAPTER 4
Known issues

This chapter describes issues that are known to affect this release of OneFS.
For information about issues known to affect other OneFS releases, log in to EMC
Online Support and view the available OneFS Release Notes.
l Backup, recovery, and snapshots.......................................................................30
l Events, alerts, and cluster monitoring known issues.......................................... 30
l File system.........................................................................................................30
l Hardware........................................................................................................... 30
l HDFS.................................................................................................................. 31
l Security.............................................................................................................. 31

Known issues 29
Known issues

Backup, recovery, and snapshots

219469
Due to a clock skew between the isi_ndmp_ndu and isi_ndmp_d processes, an
upgrade or rollback process might sometimes fail for large clusters.
Workaround: Retry the upgrade or rollback process to resolve the issue.

Events, alerts, and cluster monitoring known issues

211886
When monitoring the OneFS cluster using InsightIQ, File system analytics (FSA)
fails with Failed to write 100 entries beginning with
10cde3a2d0: Invalid argument error.

211881
Insight360 does not have CELOG queries in OneFS 8.1.x for the following:
l Event groups by set of severities
l Event groups by set of device ids
l Event list by set of eventgroup ids
l Event groups with the - Add option to list the causes for event group
definitions response (to differentiate between the n-1 and 1–1 event group
types)

210888
If the new entry does not match with the existing entry in
cl_merg_item_process, the File system analytics (FSA) fails.

File system
231425
If you try to reboot a cluster using the isi config reboot all command,
protocol services for nodes go down.
Workaround: Until this issue is resolved, you can call kill the lwsmd process.

227620
If a file on an SMB share in a SmartLock directory had an atime value of -1, the
file can not be accessed.

Hardware
211836
The isi_hw_status command displays misleading solid-state drive (SSD)
capacity on some Generation 6 nodes. The isi_hw_status command displays
maximum possible SSD drive cache size as opposed to what the node actually
contains.

30 OneFS 8.1.2 Release Notes

 Known issues

HDFS
232385
Added support for HTTP SPNEGO authentication to enable one-way SSL with
Apache Ranger. However, SPNEGO communication is not currently supported
with an Active Directory provider. This will be fixed in later versions.
234095
Files may be incorrectly renamed over the HDFS v7 protocol when using Apache
Hadoop versions 2.0.0--2.0.6.

Security
219660
After an upgrade from a Security Technology Implementation Guides (STIG)-
hardened cluster, if you try to run security hardening commands (such as isi
hardening revert) while logged on as a non-root or non-privileged user, a
Permission Denied error appears on the console.
Workaround:
Log on as a root user and change the permission of the /ifs/.ifsvar/CHE
directory to 755.

HDFS 31
Known issues

32 OneFS 8.1.2 Release Notes

CHAPTER 5
OneFS Release Resources

Sources for information about and help with the OneFS operating system.

l OneFS information and documentation.............................................................. 34

l Where to go for support.....................................................................................35
l Provide feedback about this document.............................................................. 35

OneFS Release Resources 33

OneFS Release Resources

OneFS information and documentation

You can access technical information about Dell EMC Isilon products, OneFS product
documentation, and other resources through the following channels.
Table 1 Additional OneFS resources

Channel Description
Isilon Info Hubs Isilon info hubs organize Isilon documentation, videos, blogs, and user-contributed
content into topic areas, making it easy to find content about subjects that interest you.

EMC Isilon Community Network The EMC Isilon Community connects you to a central hub of information and experts.
From this site, you can demo our products, ask questions, view technical videos, and get
the latest Isilon product documentation.

EMC Isilon OneFS Product Page The EMC Isilon OneFS product page provides access to Isilon product documentation and
current software releases. Access to this page requires that you log in to the EMC Online
Support site.

EMC Isilon Security Advisories An EMC security advisory (ESA) is published in the EMC knowledgebase when an EMC
product is impacted by a security vulnerability and a remedy that must be applied by
customers is available from EMC. For the most up to date list of Isilon ESAs, see EMC
Isilon Security Advisories (ESAs) on the EMC Isilon Community Network. You can also
subscribe to receive ESAs automatically. See article 334017 for steps to subscribe.

EMC Isilon Technical Advisories An EMC technical advisory (ETA) is published in the EMC knowledgebase when EMC
identifies an issue that may cause serious negative impact to a production environment.
For the most up to date list of Isilon ETAs, see EMC Isilon Technical Advisories (ETAs) on
the EMC Isilon Community Network. You can also subscribe to receive ETAs
automatically. See article 334017 for steps to subscribe.

EMC Isilon YouTube playlist The EMC Isilon YouTube playlist on the EMC Corporate YouTube channel contains Isilon
how-to videos, information about new features, information about Isilon hardware, and
technical overviews.

Help on This Page From any page in the OneFS web administration interface, select Help on this Page
from the Help menu to see specific information about the controls and settings on that
page. The Help on this Page option does not require internet connectivity.

Isi Knowledge The Isi Knowledge blog links to Isilon support content, announces the availability of
content, offers product tips, and provides information about new ID.TV videos.

Online Help Select Online Help from the Help menu in the OneFS web administration interface to
see information from the OneFS Web Administration Guide and the OneFS Event Reference.
The Online Help contains the latest available versions for these guides. The Online Help
option requires internet connectivity.

Isilon Advisor The Isilon Advisor is the same application that is used by EMC Isilon Technical Support
Engineers and Field Representatives to resolve service requests. You can use it to
diagnose and troubleshoot issues. You can also use it to analyze the current health of
your cluster and identify items that require attention. This can help you avoid issues that
might arise in the future.

34 OneFS 8.1.2 Release Notes

 OneFS Release Resources

Where to go for support

This topic contains resources for getting answers to questions about Isilon products.

Online support l Live Chat

l Create a Service Request
For questions about accessing online support, send an email to
[email protected].

Telephone l United States: 1-800-SVC-4EMC (1-800-782-4362)

support
l Canada: 1-800-543-4782
l Worldwide: 1-508-497-7901
l Local phone numbers for a specific country are available at
EMC Customer Support Centers.

Isilon The Isilon Community Network connects you to a central hub of

Community information and experts to help you maximize your current storage
Network solution. From this site, you can demo Isilon products, ask
questions, view technical videos, and get our latest Isilon product
documentation.
Isilon Info Hubs For the list of Isilon info hubs, see the Isilon Info Hubs page on the
Isilon Community Network. Use these info hubs to find product
documentation, troubleshooting guides, videos, blogs, and other
information resources about the Isilon products and features you're
interested in.

Support for IsilonSD Edge

If you are running a free version of IsilonSD Edge, support is available through the
Isilon Community Network. If you purchased one or more IsilonSD Edge licenses,
support is available through Isilon Technical Support, provided you have a valid
support contract for the product.

Provide feedback about this document

The links in this topic enable you to send feedback directly to the Isilon Information
Development team.
Your suggestions help us to improve the accuracy, organization, and overall quality of
the documentation. Send your feedback to https://www.research.net/s/isi-
docfeedback. If you cannot provide feedback through the URL, send an email message
to [email protected].

Where to go for support 35

OneFS Release Resources

36 OneFS 8.1.2 Release Notes