300-002-136_A01_elccnt_0

1

EMC CX300, CX500, CX500i, CX700

FLARE Software Version 02.16.yyy

Release Notes P/N 300-002-136

REV A01 January 12, 2005

These release notes contain supplemental information about EMC® CX300, CX500, CX500i and CX700 FLARE™ software. The software versions are

02.16.300.5.004 for the CX300 02.16.500.5.005 for the CX500, CX500i 02.16.700.5.004 for the CX700

Topics include

Product Description................................................................................... 2 New Features and Changes...................................................................... 2 Environment and System Requirements ................................................ 5 Fixed Problems........................................................................................... 6 Known Problems and Limitations......................................................... 16 Technical Notes ........................................................................................ 23 Documentation......................................................................................... 27 Installation ................................................................................................ 27 Troubleshooting and Getting Help........................................................ 30 Copyright and License Information ...................................................... 31

2

Product Description

EMC CX300, CX500, CX500i, CX700 FLARE Software Release Notes

Product Description FLARE™ software is the component of the FLARE Operating Environment that controls the operation of the CX300, CX500, CX500i and CX700 disk-array storage systems. Each storage system ships with a complete copy of FLARE software installed on the first four disks on back-end loop 0. Disks 0_0 and 0_2 store mirrored copies of the software for SP A, and disks 0_1 and 0_3 store mirrored copies of the software for SP B. When you power up the storage system, each SP boots and executes FLARE software. FLARE software manages all functions of the storage system and supports very basic configurations. Customers desiring larger configurations, especially multiple host configurations, will benefit from an available Access Logix™ option enabler (see Access Logix Release Notes, P/N 300-002-138). Access Logix is preinstalled at the factory on all CX300, CX500, CX500i and CX700 storage systems. You can enable storage groups with Navisphere® software.

FLARE software without Access Logix enabled supports only the following configurations:

a single host dual-host split-bus two clustered hosts

New Features and Changes

iSCSI

Overview

The iSCSI protocol provides a mechanism for mapping SCSI commands, data and status into IP packets and transporting them over the Ethernet via the TCP/IP internet. Because iSCSI moves information over the standard IP network, it uses existing network services and protocols to support management, discovery, authentication, and encryption.

Benefits include:

• The potential for using a single network for both storage and non-storage traffic (although in many cases it may be important to use separate networks for storage and general IP traffic).

• More cost-effective solutions for data replication and distribution over wide-area distances.

• Protocols that include SNMP, Ping, Traceroute and Chap.

3



SCSI Authentication

In previous CLARiiON storage products storage systems were either directly attached to hosts or connected via a Storage Area Network. This configuration led to storage being distinctly separate from communication networks. With the advent of SCSI-over-IP network, storage implementations must provide a means to protect against active and passive attacks via this network.

The iSCSI Specification allows for 2 separate security mechanisms: in-band authentication between the initiator and target at the iSCSI login level and packet protection by IPsec at the IP level. This release supports only in-band authentication.

During an iSCSI login the target must authenticate the initiator and the initiator may authenticate the target. The authentication method that is performed is negotiated between initiator and target.

This release will support only the Challenge Handshake Authentication Protocol (CHAP). This method is required for an initiator and target to be compliant with the iSCSI Specification.

iSCSI Connection State Log and Display

The recording of any iSCSI login and logout events including login failures will be stored in the Navisphere event log. This information can be viewed with the Navisphere UI. User options are available to narrow the search for these iSCSI login/logout events.

This logging functionality gives the user the ability to view the following information:

• Initiators that are currently logged in. • Initiators that were logged in but have since logged out. • Initiators that failed to log in due to some type of login problem. • Initiators that were rejected from logging in.

Requirements

iSCSI functionality is supported only on the CX500i platform.

LUN Migration

LUN migration is a feature that will, without disruption of host applications, allow a user to change performance and other characteristics of an existing LUN by migrating data from a source LUN to a destination LUN with the same or larger capacity and with desired

4



characteristics within a single storage system. After LUN migration the original source LUN will be destroyed, the destination LUN will take on the source LUN's identity, and host I/O requests will continue to the destination LUN. This process occurs with no disruption to host I/O.

Migration can be performed from a source LUN to a destination LUN of different characteristics, including RAID geometry, drive types (FC vs. ATA), disk utilizations, and disk locations.

Migration rate can be specified at initial configuration time and changed during migration to control performance impact.

Migration has no data availability disruption to host or layered driver applications to the source LUN. It copies the content of the source LUN to the destination LUN while mirroring writes to both source and destination. It services host reads from the source LUN.

Migration will hide the destination LUN from host or other layered driver.

Upon completion of a migration the destination LUN atomically takes on source LUN's identity, including LUN nice name, WWN, and LUN ID while keeping its other own parameters. The destination LUN is then exposed to the host or other layered driver and the source LUN is destroyed.

A customer can cancel a migration. In this case the source LUN is left intact and unchanged while destination LUN is destroyed for security reasons.

Migration handles failure cases including recoverable and non-recoverable errors.

Two Drive R1/0

With this release a 2-drive R1/0 LU can be configured. This LU will support all features that exist on a 4+ drive R1/0 LU. The main advantage of a 2 drive R1/0 LU over a R1 LU is that it will support RAID Group expansion.

5

Environment and System Requirements


Environment and System Requirements This section identifies the minimum requirements necessary for the proper operation of FLARE with the CX300, CX500, CX500i and CX700 storage systems.

Storage-System Type

Minimum Hardware Requirements

Minimum Software Requirements

Supported 2-Gbyte Fibre Channel Disks

Supported ATA Disks

CX300, CX500, CX500i, CX700

For write caching

• 5 disks minimum, installed in slots 0 through 4 of enclosure 0

• 2 SPs

• 2 power supplies

• 1 standby power supply (SPS)

Navisphere SP Agent version 6.16 and Navisphere Management Server version 6.16.

Please see ESM. Please see ESM.

6

Fixed Problems


Fixed Problems This section identifies issues fixed with this release or recent previous releases.

In addition to the following fixes, this release contains all fixes documented in prior releases of FLARE 02.xx software for the CX200, CX400 and CX600 platforms. See the FLARE Release Notes (P/N 085090686) for details.

Problem Detail

CX300, CX500, CX500i, CX700 Benign 0x873 errors entered into the event log at powerup. (105149)

Fixed in Revision:

02.16.xxx.5.004

02.16.xxx.5.005 CX500, CX500i only

Solution: Do not log the 0x873 errors unless the condition still persists at the end of system initialization.

CX300, CX500, CX500i, CX700 SP left in a non pingable and unresponsive state after a re-imaging attempt. (111471)

Fixed in Revision:

02.16.xxx.5.004

02.16.xxx.5.005 CX500, CX500i only

Solution: Added data verification test.

CX300, CX500, CX500i, CX700 Degradation seen in large write performance during backup to ATA disks. (115796)

Fixed in Revision:

02.16.xxx.5.004

02.16.xxx.5.005 CX500, CX500i only

Solution: Add performance fixes for large element host writes and R3 cache writes.

CX300, CX500, CX500i, CX700 After power failure, cache dumped followed immediately by SP B panic. (113981)

Fixed in Revision:

02.16.xxx.5.004

02.16.xxx.5.005 CX500, CX500i only

Solution: Remove recursion in cache code that caused stack overflow.

CX300, CX500, CX500i, CX700 Default metaLUN stripe element multiplier defaults to 40 resulting in a stripe depth too large. (105374)

Fixed in Revision:

02.16.xxx.5.004

02.16.xxx.5.005 CX500, CX500i only

Solution: Changed default stripe element multiplier to 4.

CX300, CX500, CX500i, CX700 NDU issues seen during Customer NDU testing. (107442, 107891, 109695, 110515, 112989, 113786, 114034, 115210, 116115, 117903)

Fixed in Revision:

02.16.xxx.5.004

02.16.xxx.5.005 CX500, CX500i only

Solution: Added fixes and enhancements to the code to make NDU more robust.

7

Fixed Problems


Problem Detail

CX300, CX500, CX500i, CX700 Miscabled backend caused the SP to reboot and trip the reboot counter. (107216)

Fixed in Revision:

02.16.xxx.5.004 CX300 only

02.16.xxx.5.005 CX500, CX500i only

Solution: Keep port disabled when miscabling from expansion port to expansion port exists.

CX300, CX500, CX700 Storage system checks when Cisco name server switch logs into the storage system during name server query. New switch behavior. (108332)

Fixed in Revision:

02.16.xxx.5.004

02.16.xxx.5.005 CX500 only

Solution: Reject unexpected/illegal login from name server.

CX300, CX500, CX500i, CX700 Occasionally cannot bind new LU’s to an RG after multiple trespasses occur while the RG is expanding. (109828).

Fixed in Revision:

02.16.xxx.5.004

02.16.xxx.5.005 CX500, CX500i only

Solution: Correctly handle concurrent RG expansion and LU trespasses.

CX300, CX500, CX500i, CX700 If a user starts a RAID Group expansion and then unbinds all of the LUNs in the RAID Group prior to the expansion completing, the RAID Group will become inaccessible. (107829)

Fixed in Revision:

02.16.xxx.5.004

02.16.xxx.5.005 CX500, CX500i only

Solution: Check for a valid RG table before modifying it.

CX300, CX500, CX500i, CX700 Rebuild panic “RBLD_PEER_OUT_OF_SYNC” (0x0070001C). This panic has been seen in the following dimms – 94610, 94991, 95144, 97938, 104094, 107089, 108786, 115237, 115461, 116497, 117670.

Fixed in Revision:

02.16.xxx.5.004

02.16.xxx.5.005 CX500, CX500i only

Solution: Added new and improved rebuild processing support in code base. This enhancement addressed the RBLD_PEER_OUT_OF_SYNC panic seen in multiple dimms. It also made the rebuild code in general more robust and better able to handle abnormal conditions.

CX300, CX500, CX700 On rare occasions, due to heavy trespass activity for mirrored LUNs, a storage processor may encounter a panic (0xE111805F) (117334)

Fixed in Revision:

02.07.016

Solution: Fixed code to handle buffers in an optimized fashion.

CX300, CX500, CX700 If a drive fails in an ATA enclosure and is faulted by FLARE, sometimes the replacement drive never comes ready for use by FLARE. It stays in the powering up state until the system is reset. (109727)

Fixed in Revision:

02.07.015

Solution: Upgrade to the latest FLARE code that contains Yukon’s FRUMON firmware code 1.53 that runs on the Bridge Control Card (BCC)

8

Fixed Problems


Problem Detail

CX300, CX500, CX700 If a drive fails in an ATA enclosure and is faulted by FLARE, sometimes the replacement drive is not recognized as being physically inserted in the enclosure. It is unusable until the system is reset. (114461)

Fixed in Revision:

02.07.015


CX300, CX500, CX700 If one of the storage processors panics or reboots for an unrelated reason while a trespass is in progress, the LUN will become inaccessible until the surviving storage processor is rebooted (117852)

Fixed in Revision:

02.07.015

Solution: Fixed bug in Trespass error handling code to handle unexpected peer storage processor reboots

CX300, CX500, CX700 In some cases when the storage system was previously configured as a NAS backend and then converted to SAN prior to the upgrade, Storage processor may fail to reboot as a part of the Non-Disruptive Upgrade. (115324)

Fixed in Revision:

02.07.015

Solution: Enchanced NAS to SAN conversion code to automatically remove NAS-specific configuration information.

CX700 On very rare occasions, storage processor may encounter a panic (0x01982013) when one of the chip components on the storage processor returns erroneous result (115559)

Fixed in Revision:

02.07.015

Solution: Added code to handle the specific error condition.

CX300, CX500, CX700 Online DAE2-ATA enclosure additions are prohibited if the enclosure being added has down-rev firmware from what is on the storage system. (1t12512)

Fixed in Revision:

02.07.015

Solution: Upgrade to the latest FLARE code that contains Yukon’s FRUMON firmware code 1.53 that runs on the Bridge Control Card (BCC) to remove this restriction for future upgrades.

CX300, CX500, CX700 Online BCC replacements are prohibited if the BCC has down-rev firmware from what is on the storage system. (112512)

Fixed in Revision:

02.07.015

Solution: Upgrade to the latest FLARE code that contains Yukon’s FRUMON firmware code 1.53 that runs on the Bridge Control Card (BCC) to remove this restriction for future upgrades.

CX300, CX500, CX700 After an NDU while the BCC firmware is being updated, the possibility exists that all drives could fault in the target enclosure, or the target enclosure could become unresponsive, or the SP could panic. (116162)

Fixed in Revision:

02.07.015


9

Fixed Problems


Problem Detail

CX300, CX500, CX700 Occasionally during periods of heavy I/O, all fault lights on an enclosure are lit and storage processor on the B-side looses access to all drives in the enclosure. (117160)

Fixed in Revision:

02.07.015


CX300, CX500, CX700 Storage processor may panic (0x00800094 or 0xE115800C) when a drive with faulty hardware causes power glitch (116323)

Fixed in Revision:

02.07.013

Solution: Added special code to handle power glitch events caused by faulty drives on the backend.

CX300, CX500, CX700 Sub-optimal performance for large write requests to LUNs bound using ATA drives (115777)

Fixed in Revision:

02.07.013

Solution: Improved buffer handling code for better performance.

CX300, CX500, CX700 Storage processor may panic (0x0000001e) due to certain unexpected disk drive underrun events. (114439)

Fixed in Revision:

02.07.011

Solution: Storage-system software will correctly handle the error.

CX300, CX500, CX700 Storage processor constantly rebooting due to drive that reported unexpected status. (112362)

Fixed in Revision:

02.07.011

Solution: Retry the failing operation 3 times and shut down the failing drive.

CX300, CX500, CX700 One or more drives in an ATA enclosure are stuck in Transitioning state. This can happen when one of the Bridge Control Cards (BCC) in an ATA enclosure crashes and reboots, and while it is rebooting the peer BCC card in this enclosure also crashes and reboots. This causes the drive(s) to get stuck in powerup mode on either one of the two back-end (BE) buses connecting to this enclosure. (114120, 108858)

Fixed in Revision:

02.07.011

Solution: This problem is very rare and is manifested by the way BCCs respond to certain faults on the backend. FLARE is reacting to the error messages sent by the BCCs. The fix is a two-part fix:

1. Improved error handling by FLARE

2. Uprade to the latest firmware code (Yukon’s FRUMON code 1.53) that handles these situations in a more graceful manner.

CX300, CX500, CX700 FRU stuck in Transitioning state after a bad drive is inserted because the lock information is erroneously dropped. (114117)

Fixed in Revision:

02.07.011

Solution: The fix involves checking if the peer storage processor is waiting for us to release the lock before clearing the lock information. If it is waiting, then send an appropriate response to the peer before clearing the lock information.

10

Fixed Problems


Problem Detail

CX500, CX700 A chip errata for one of the components on the Storage processor could result in a processor hang. The failure has not been observed. (113250)

Fixed in Revision:

02.07.011

Solution: New BIOS microcode to address this HW chip errata

CX300, CX500, CX700 If two of drives 0_0_0 through 0_0_2 are faulted and then one is replaced, under rare conditions the storage processor may panic with code 0xE115800C. (114446)

Fixed in Revision:

02.07.011

Solution: Fixed race condition in the LUN assignment code.

CX300, CX500, CX700 LUNs belonging to a double-faulted RAID Group (RG) are not marked as unowned resulting in poor failover performance. The problem was observed when a storage system served as a backend for a Celerra/NAS environment. (113488)

Fixed in Revision:

02.07.011

Solution: Correctly mark LUN(s) as unowned when double-faulted.

CX300, CX500, CX700 Storage processor may encounter a panic (0x0000000) if the attached host sends login-frames containing invalid World Wide Node (WWN) name and Class of Service information. (114801)

Fixed in Revision:

02.07.011

Solution: Modified login validation algorithm to reject illegal login frames with appropriate error message.

CX300, CX500, CX700 LUNs may fail to assign (go unowned) after a drive replacement, a hot spare equalize where the hot spare is a different size than the drive it replaced, and then a trespass to the peer SP (112438)

Fixed in Revision:

02.07.010

Solution: Correct defect in code that determines the rebuild checkpoint following drive replacement.

CX300, CX500, CX700 Installation of some software packages required for conversion operations will fail if running the 02.07.006 (CX300) or the 02.07.008 (CX500, CX700) version of the FLARE Operating Environment. (111885)

Fixed in Revision:

02.07.007 (CX300); 02.07.009(CX500, CX700)

Solution: Fix NULL pointer in code that installs conversion and recovery software images.

CX300, CX500, CX700 Storage processor may encounter a panic (0x0070001d) if one of the FLARE database drives (0, 1 or 2) fails while the peer storage processor is in a failed or powered-down state and the storage system is in the process of rebuilding the peer’s error log database. (109403)

Fixed in Revision:

02.07.006 (CX300); 02.07.008 (CX500, CX700)

Solution: Correctly handle the case where a drive fails while an error log rebuild is in progress.

CX300, CX500, CX700 Enhance disk drive error reporting so that errors that were previously reported as 801 Soft SCSI Bus Error with sense key of 09 will now be reported as 803 Recommend Drive Replacement with sense key of 09 to enable filtering the benign 801 errors from the more serious drive replacement cases. (109675)

Fixed in Revision:

02.07.006 (CX300); 02.07.008 (CX500, CX700)

Solution: Disk drive check condition 04 (hardware error) is now reported as Recommend Drive Replacement instead of Soft SCSI Bus Error.

11

Fixed Problems


Problem Detail

CX300, CX500, CX700 Storage processor may encounter a panic (0x7115800C) if committing new software while drive 0 is removed from the storage system. (Note that it is recommended that all system errors be resolved prior to upgrading and committing new software.) (108867)

Fixed in Revision:

02.07.006 (CX300); 02.07.008 (CX500, CX700)

Solution: Query all drives for revision information at commit time instead of just drive 0, in case drive 0 is removed.

CX300, CX500, CX700 Trespass of a metaLUN failed to trespass one of the component LUNs due to a prior error connecting to the LUN. (110510)

Fixed in Revision:

02.07.006 (CX300); 02.07.008 (CX500, CX700)

Solution: Correctly handle error path when a host fails to connect to a LUN.

CX300, CX500, CX700 LUNs fail to assign (unowned) after drive replacement, a hot spare equalize where the hot spare is a different size than the drive it replaced, and then a trespass to the peer SP. This problem only occurs if LUN partition 0 of the RAID Group (RG) has been unbound. All LUNs on the RG will become unowned if any in the RG attempts a trespass. (110438)

Fixed in Revision:

02.07.006 (CX300); 02.07.008 (CX500, CX700)

Solution: Fix defect in the address calculation code in the hot spare swap-out routine.

CX300, CX500, CX700 RAID Group expansion operation may hang following multiple trespass operations. (108856)

Fixed in Revision:

02.07.006 (CX300); 02.07.008 (CX500, CX700)

Solution: Fixed case where expansion operation is not correctly restarted following trespass.

CX300, CX500, CX700 Storage processor may encounter a panic (0x019n1001) when it receives a switch name server login while a name server query is in progress. This is illegal switch behavior, but has been observed on one occasion. (108849)

Fixed in Revision:

02.07.006 (CX300); 02.07.008 (CX500, CX700)

Solution: Reject the illegal login instead of panicking the storage system.

CX300, CX500, CX700 The Navisphere Manager Enabler was inadvertently excluded from the manufacturing code images. (110022)

Fixed in Revision:

02.07.006 (CX300); 02.07.008 (CX500, CX700)

Solution: Include the Navisphere Manager Enabler.

CX300, CX500, CX700 Under certain conditions when attempting to establish a new MirrorView connection, or re-establish an existing connection following a storage processor reboot, the login to establish the connection fails due to a collision of operations between the peer storage processors. Recovery required disabling and re-enabling the connection. (109682)

Fixed in Revision:

02.07.006 (CX300); 02.07.008 (CX500, CX700)

Solution: Correct defect in the flag management in the Test Unit Ready code path.

CX300, CX500, CX700 Following an NDU that includes updated LCC FRUMON code, if an enclosure is powered down while the FRUMON code is still upgrading, the SP may panic with code (0x0080010e). (97487)

Fixed in Revision:

02.07.001 (CX300, CX500, CX700) Solution: Added check for all enclosures on a bus going away after the FRUMON upgrade

begins.

12

Fixed Problems


Problem Detail

CX300, CX500, CX700 Panic (0x82974da8) following software upgrade under heavy I/O load. The FLARE upgrade completes, but the storage processor times out while trying to determine whether the LCC FRUMON code requires upgrading.

(97791)

Fixed in Revision:

02.07.001 (CX300, CX500, CX700) Solution: Improve timing of check for required FRUMON upgrade.

CX300, CX500, CX700 Fibre driver encounters a panic following drive firmware download. Storage processor would have rebooted anyway as a result of the drive firmware upgrade. (97865)

Fixed in Revision:

02.07.001 (CX300, CX500, CX700)

Solution: Fixed contention issue during disk firmware download.

CX700 When loading disk drive firmware to an individual drive with a FRU number greater than 255 (will occur on the CX700 on back-end busses 2 and 3) the following error appears:

"APM00032501973: Installation failed: Data wrong size or did not checksum (0x79508004)".

(97988)

Fixed in Revision:

02.07.001 (CX300, CX500, CX700) Solution: Allow drive firmware download to FRUs greater than 255.

CX300, CX500, CX700 When FA MIB is enabled on the storage system, 02.06 FLARE has an SNMP vulnerability (CAN-2004-0123). See Microsoft security alert MS-04-011 or EMC Security Advisory EMC-2004-11 for more details.

(107987)

Fixed in Revision:

02.06.013 (CX300, CX700); 02.06.014 (CX500)

Solution: Addressed security vulnerability in patch release.

CX300, CX500, CX700 When powering up pre-zeroed disk drives for the first time (typically a new storage system or new enclosure) the drives may hang in the powering up state. This occurs only on the very first powerup, so if the storage system was preconfigured and powered up at the factory this will not occur at a customer site. For storage systems that are not pre-configured at the factory, for add-on enclosures or for add-on disk drives, please see the recommended work-around. (97503)

Fixed in Revision:

02.06.012 (CX300, CX700); 02.06.013 (CX500)

Solution: Eliminate discarded write operation when both SPs try to concurrently perform a FRU signature write.

CX300, CX500, CX700 On rare occasions, disk drives exhibit a failure behavior where they repeatedly go offline and return online on one back-end Fibre Channel bus, by asserting and releasing one of their two port bypass bits. The frequency of the state changes and the fact that only one SP has visibility to the behavior can result in the two CLARiiON storage processors getting out of sync with respect to their view of the state of this drive. This can lead to the drive getting stuck in the powerup state. (106301)

Fixed in Revision:

02.06.012 (CX300, CX700); 02.06.013 (CX500)

Solution: Improve the peer-to-peer synchronization during the drive powerup process.

CX300, CX500, CX700 Disk drive diagnostic tool does not operate on drives on back-end busses 2 or 3. (106797)

Fixed in Revision:

02.06.012 (CX300, CX700); 02.06.013 (CX500)

Solution: Use physical bus count constant to determine the number of back-end busses.

13

Fixed Problems


Problem Detail

CX300, CX500, CX700 MetaLUN configuration operations (changing Aggregate Volume parameters or LUN nice name expansion rate) while expanding a metaLUN or performing I/O that spans multiple components of the metaLUN can result in SP failure. (106329)

Fixed in Revision:

02.06.012 (CX300, CX700); 02.06.013 (CX500)

Solution: MetaLUN I/O manager needs to check for quiesce for a configuration change operation before generating I/O instead of after.

CX300, CX500, CX700 MetaLUN support fails to initialize a stack correctly, which can result in errors during non-I/O operations such as trespass. This problem has been observed only while testing a future release of the software, but has the potential to fail in the current software. (104718)

Fixed in Revision:

02.06.010 (CX300, CX700)

02.06.011 (CX500)

Solution: Correctly initialize the IRP stack.

CX300, CX500, CX700 Replacing drive 0 on a running storage system can result in an SP panic (0x00800131 or 0x0070001c) due to referencing invalid database information on the replacement drive. (102037)

Fixed in Revision:

02.06.010 (CX300, CX700)

02.06.011 (CX500)

Solution: Compare revision information on all database drives to determine whether drive 0 is valid.

CX300, CX500, CX700 If a metaLUN is assigned a “Nice Name” that includes any of the following characters (>, <, &, “,’) then on a subsequent reboot the metaLUN will become inaccessible and the storage system may become unmanaged. (104274)

Fixed in Revision:

02.06.010(CX300, CX700)

02.06.011 (CX500)

Solution: Store special characters as escape sequences for the name parser.

CX300, CX500, CX700 When a storage system is powered up with drive 0 removed, FLARE may fail to swap in a hot spare if it has not completed its powerup cycle. (104389)

Fixed in Revision:

02.06.010(CX300, CX700)

02.06.011 (CX500)

Solution: FLARE will wait for the hot spare to be in the Ready state prior to attempting to swap it in.

CX300, CX500, CX700 If a metaLUN is bound with certain offset values, then the user cannot perform an Expansion operation on that metaLUN. (103787)

Fixed in Revision:

02.06.010(CX300, CX700)

02.06.011 (CX500)

Solution: Correct the metaLUN Expansion code to correctly account for bind offsets.

CX300, CX500, CX700 Under certain stressful I/O load conditions, the storage system may not respond to a front-end fibre login command (PLOGI or PRLI) before FLARE times out. This could inhibit the ability of some hosts to log in and communicate with the storage system. (102698)

Fixed in Revision:

02.06.010(CX300, CX700)

02.06.011 (CX500)

Solution: Increase the login timeout value.

14

Fixed Problems


Problem Detail

CX300, CX500, CX700 Race condition in metaLUN ownership code can result in an SP panic during a metaLUN trespass if one of the component LUNs is double faulted. (104402)

Fixed in Revision:

02.06.010(CX300, CX700)

02.06.011 (CX500)

Solution: Add additional locking in the metaLUN trespass code path.

CX500 On rare occasions, the storage system can fail to successfully power up because the BIOS does not properly initialize a hardware register. (103580)

Fixed in Revision:

BIOS 3.04 included in 02.06.011 (CX500)

Solution: Correctly initialize the machine check register at powerup.

CX300, CX500, CX700 The last disk drive on the back-end fibre loop may fail after software upgrade. (103093)

Fixed in Revision:

02.06.007 (CX300, CX700)

02.06.009 (CX500)

Solution: Correctly initialize a pointer used by the fibre driver.

CX300, CX500, CX700 Incorrect handling of an unexpected variation of disk drive under-run errors can result in returning old data on a read operation. The data on the disk is correct. This was encountered during disaster recovery simulation on a test lab storage system that included a number of old drives experiencing these under-run errors. (103273)

Fixed in Revision:

02.06.007 (CX300, CX700)

02.06.009 (CX500)

Solution: Correctly detect and handle this particular error condition.

CX300, CX500, CX700 Occasionally the storage processor will panic with code (0x 008000de) when a storage system is upgraded or an SP is rebooted while a LUN rebuild is in progress. (100366)

Fixed in Revision:

02.06.007 (CX300, CX700)

02.06.009 (CX500)

Solution: Correct a defect in the initialization of the unit number in the rebuild code.

CX300, CX500, CX700 Storage processor may encounter a panic (0x01964036) when it receives a front-end target reset or abort. (101288)

Fixed in Revision:

02.06.007 (CX300, CX700)

02.06.009 (CX500)

Solution: Correct abort handling to avoid corruption of a pointer.

CX300, CX500, CX700 Storage processor may encounter a panic (0x80000003) when doing misaligned I/O to metaLUNs with SAN Copy™ and Read Cache enabled. (101802)

Fixed in Revision:

02.06.007 (CX300, CX700)

02.06.009 (CX500)

Solution: Fixed race condition in metaLUN handling code for cases where data is in the Read Cache and responds immediately.

15

Fixed Problems


Problem Detail

CX300, CX500, CX700 Hot spare gets stuck in the powerup state when the system experiences back-end loop failures. (103184)

Fixed in Revision:

02.06.007 (CX300, CX700)

02.06.009 (CX500)

Solution: Improved peer synchronization and error recovery in hot spare handling when one storage processor detects back-end loop failure.

CX300, CX500, CX700 Storage processor may encounter a panic (0x00300027) when a failed drive is replaced on a FLARE 02.06 storage system with a drive previously installed in a storage system running 02.04 or earlier, or that has 02.05 installed but not committed. A hot spare must be running in place of the original failed drive at the time of the replacement for this failure to occur. (103336)

Fixed in Revision:

02.06.007 (CX300, CX700)

02.06.009 (CX500)

Solution: Fixed an error in the FRU number masking when a hot spare is swapped in.

CX300, CX500, CX700 Storage processor may encounter a panic (0x00f2001b) if performing intensive write I/O to more than 47 mirrored RAID Groups (R1 or R1/0) with write cache enabled and the write cache initiates intensive page flushing due to a fault or idle LUNs. (99583, 100865)

Fixed in Revision:

02.06.006 (CX300, CX700)

02.06.007 (CX500)

Solution: Increased the number of Nonvol write records to track outstanding I/Os and improved the error handling in the RAID driver.

CX300, CX500, CX700 Due to a resource queuing issue in the Clean/Dirty processing code, if synchronous bursts of I/O are done to very large numbers of LUNs, I/O response time can be degraded due to conflicts in marking LUNs clean or dirty. This performance degradation is likely to be seen only if this pattern of I/O is performed simultaneously to 500 or more LUNs. (100271)

Fixed in Revision:

02.06.006 (CX300, CX700)

02.06.007 (CX500)

Solution: Fixed a resource queuing issue in the Clean/Dirty processing code.

CX300, CX500, CX700 Enclosure 0 Fault LED remains lit after powerup if an SPS is not installed in the storage system. All customer installations require an SPS, so this is just a manufacturing test issue. (100216)

Fixed in Revision:

02.06.006 (CX300, CX700)

02.06.007 (CX500)

Solution: Correctly detect peer powerup complete and turn off the fault LED. Do not wait for SPS test to complete if one isn’t installed.

16

Known Problems and Limitations


Known Problems and Limitations This section identifies outstanding issues with this release, and important warnings associated with the connectivity and compatibility of this product. It also provides useful recommendations.

Known Problems Frequency

The frequencies for the known problems listed in this section have the following meanings:

Frequency Definition

Every Time Occurs whenever the specific condition is met.

Very Frequent Occurs between 30 and 50 percent of the time when the specific cause is repeated.

Frequent Occurs between 10 and 30 percent of the time when the specific cause is repeated.

Intermittent Occurs between 1 and 10 percent of the time when the specific cause is repeated.

Very Intermittent Occurs less than 1 percent of the time when the specific cause is repeated.

Extremely Intermittent Failure mechanism may never be seen except under the most stressful QA test environments.

Description Detail

CX300, CX500, CX500i, CX700 Under certain rare circumstances, the firmware update process that follows the NDU procedure will exhibit one of the following failure modes. These fault modes do not result in loss of access; however, manual corrective action is necessary by your EMC Service Representative.

• An individual drive will be faulted.

• An enclosure fault will be indicated, and the Navisphere log will show a RESUME prom error.

• One of the BCC controllers will fault, causing loss of access to a DAE2-ATA enclosure on one side only.

Frequency Extremely Intermittent after an NDU if an enclosure/storage system exhibits one of the failure modes listed here.

Effect Fault light on an ATA enclosure lights up and/or error message appears in the Navisphere logs.

Prevention/Resolution Please contact your EMC Service Representative.

17



Description Detail

CX300, CX500, CX500i, CX700 Do not physically remove an ATA drive from the DAE2-ATA enclosure if that drive is actively running I/O.

Frequency Intermittent when an ATA drive that is actively running I/O is removed.

Effect This drive or any replacement drive that is inserted in that slot may not successfully power up.

Prevention/Resolution Suspend all I/O to the drive before removing it. If the storage system experiences this failure or you must replace the drive under active I/O load, then please contact your EMC Service Representative.

CX300, CX500, CX500i, CX700 During an NDU operation a message similar to the following appears in the Event Log:

Checking file system on C:

The type of the file system is NTFS.

Volume label is FLARE Partition.

A disk check has been scheduled.

Windows will now check the disk.

Cleaning up minor inconsistencies on the drive.

…

Windows has finished checking your disk.

Please wait while your computer restarts.

This is due to an intentionally scheduled run of a utility to verify that all files were transferred and installed on the storage system correctly.

Frequency Every time an upgrade or revert occurs

Effect Error message appears in the Event Log.

Prevention/Resolution The message is expected and should be ignored.

CX300, CX500, CX500i, CX700 Trespass timeouts can occur if you attempt to perform an NDU under I/O load when metaLUNs are configured to include more than 30 LUNs from a single RAID Group. (96958)

Frequency Intermittent during NDU under load.

Effect MetaLUN trespass times out and an SP panic (0xe117b264) occurs.

Prevention/Resolution Do not configure metaLUNs with over 30 component LUNs from a single RAID Group. MetaLUNs are intended to aggregate LUNs from different RAID Groups.

18



Description Detail

CX300, CX500, CX700 - When performing one or more LUN migrations in the AIX operating environment with PowerPath 4.2 installed, errors may be logged to the host operating system indicating reservation conflict or device busy on certain commands. These are benign error messages and have no adverse effect on the operation of the AIX host or CLARiiON storage system. A later version of PowerPath 4.x for AIX is expected to eliminate the error messages. (109068)

Frequency Intermittent under conditions described.

Effect No effect on I/O or other storage-system operation.

Prevention/Resolution Run a version of PowerPath for AIX that is later than release 4.2.

CX300, CX500, CX700 - The storage system will fail to log out initiators when all initiators are removed from the zone the storage system is in, leaving just the one storage-system port. The removal of initiators is accomplished by either physically removing all initiators at the same time or by rezoning. (118871).

Frequency Every time

Effect Initiator resources for the removed initiators stay comsumed until a rezoning is performed. If the removed initiators are returned they will use the existing resources.

Prevention / Resolution Ensure at least one other host or storage-system port is in the same zone.

CX500i – If the host iSCSI name is being changed from the default, the new name must be a minimum of 16 valid characters in length. (118537).

Frequency Every time

Effect If the iSCSI name is shorter than 16 characters the host will not configure its LUNs at powerup.

Prevention / Resolution Create an iSCSI name with a minimum of 16 valid characters.

Limitations

This section lists issues that may affect the configuration or repair of the storage system.

CX300, CX500, CX500i and CX700 Storage Systems

Background Verify (BV) When using Navisphere CLI to start a background verify on an SP that does not own a LUN, the command is accepted, but the log shows a BV started/BV aborted. This happens every time a background verify is started on an SP that does not own a LUN.

Non-Disruptive Software Installation (NDU) You should not perform a non-disruptive software installation (NDU) when heavy I/O activity is occurring.

Restrictions during Non-Disruptive Software Installation (NDU) You cannot initiate any bind or expansion operations while an NDU is in progress. The software will reject any attempts to do so.

19



Non-Disruptive Software Installation (NDU) and Tru64 UNIX Servers If a Tru64® UNIX® server is connected to more than 180 LUNs on a single SP, you should not perform a non-disruptive software installation (NDU) while significant I/O activity is occurring. N/A for CX500i.

Software Commits After performing a non-disruptive software update (NDU), you cannot bind or unbind LUNs, or create or destroy RAID Groups until you have committed the new software. Once software is committed, you cannot revert to the previous revision of software. You can revert only before the software is committed. Additionally, you must commit the currently active software before attempting an NDU of a new package.

When performing an NDU commit, do not execute other configuration administration commands until the commit is complete. This includes unbinding LUNs, creating/destroying mirrors, adding/removing secondary mirrors, allocating/deallocating the remote mirror write intent log, creating/deleting a snapshot, adding/removing a LUN to/from the snap cache, adding/deleting snap clones, or allocating/deallocating snap clones private LUNs.

Additionally, if the commit operation fails to complete for any reason, (such as a power failure), retry the commit operation. Once the commit has been issued, it is not possible to revert the installation.

Unbinding LUNs If you try to unbind a LUN while the LUN is performing I/O, the operation fails and the LUN becomes inaccessible.

To avoid this situation, make sure all I/O to the LUN has stopped before unbinding it.

To correct this situation, stop all I/O and then unbind the LUN.

20



Disk Address Conflicts FLARE software cannot manage disk or enclosure address conflicts on the back-end buses. If you insert disks that do not follow standard FC-AL addressing in the storage system, FLARE software may unintentionally shut down disks currently installed in the storage system or exhibit some other type of failure, such as panics or system hangs. This also happens if you add a new DAE2 to a CX-Series storage system that has its enclosure address in conflict with another DAE2 on the same back-end loop. You can recover by removing the disks you just installed and then powering the storage system down and up. To avoid problems, you should double-check all cabling and enclosure address settings when you add enclosures or make any changes to their addresses or cabling.

Always set the enclosure address on the DPE to enclosure address 0 or the storage system will not power up properly.

DAE2 Replacement Restrictions You can install only Fibre Channel drives in the DAE2 enclosures with link control cards (LCCs) and only ATA drives in the DAE2-ATA enclosures with FC-to-ATA link control cards. If you install the wrong drive type, the system will not recognize it. If you replace an LCC card with the incorrect type, it will not successfully power up and operate. Although you can mix FC and ATA enclosures within a storage system, once you have installed an enclosure and created RAID Groups, you cannot replace that enclosure with an enclosure set to the same enclosure number but of the other type unless you first destroy all of the RAID Groups.

DAE2-ATA Application Limitations The standard DAE2 enclosure (Fibre Channel disks) has better performance than the DAE2-ATA (ATA disks). The DAE2-ATA is targeted for applications where speed is not critical, such as backup storage. Do not use a LUN in the DAE2-ATA as a host OS boot device.

NDU on a Busy Windows Server If an NDU operation is performed on an array connected to a Windows hHost with PowerPath® installed and that host is too busy, PowerPath may be unable to trespass all LUNs from one SP to the other during the NDU operation. This would cause application I/O to those LUNs to fail. To avoid this issue, before you start an NDU, from the Windows Task Manager Performance screen, verify that the CPU usage on the host is less than 50%. Reduce host activity if necessary. If CPU usage exceeds

21



75%, PowerPath will frequently fail to trespass some LUNs. By reducing host CPU usage to 50% or below, sufficient margin exists to ensure that LUNs will be trespassed correctly and the array upgrade will be non-disruptive.

LUN Migration

The migration of a component LU which is part of an aggregate LU is not supported in this release.

The destination LUN must be a 'public' LUN when it is chosen for migration (this is the normal case).

The destination LUN may not be in use by any other layered feature. The destination LUN may not be in a user defined Storage Group. Migration and LUN expansion are mutually exclusive operations.

An expanding LUN may not be migrated, and a migrating LUN may not be expanded.

The source LUN cannot be a private LUN. If the source LUN is being used by a layered application, it may only

be migrated to a LUN of the same size. After a migration completes the data reported in the inquiry

command will still reflect the LUN's original RAID type. The new RAID type is visible through the Navisphere interface. This behavior is as designed and required to maintain compatibility with some operating systems.

The maximum number of migrations that can be executing at one time are as follows:

• CX700 12 • CX500, CX500i 8 • CX300 4

New Maximum Settings for Read/Write Cache Size With this release 02.16.xxx.5.00x the maximum memory available for cache has decreased on the CX300, CX500, CX500i and CX700 platforms. This change is the result of reallocation of some memory in support of new features for this release.

Before an NDU if the total size of read and write cache per SP is greater than the value listed below then after the NDU the original settings will not be restored. This will result in read and write cache being disabled. The cache will stay disabled until the values are adjusted to be within the new limits.

22



Note: On the CX700 the maximum write cache value has always been less than the maximum read cache value. This has not changed with this release.

CX300 561 Mbytes

CX500, CX500i 1497 Mbytes

CX700 3194 Mbytes

This condition will be noted in the Navisphere Event Log with the following entry:

Failed to restore cache settings.

Original Cache settings cannot be restored, Memory available for cache is less than the memory required by original cache settings.

Technical Notes

23 EMC CX300, CX500, CX500i, CX700 FLARE Software Release Notes

Technical Notes This section includes important information that does not appear anywhere else.

CX300, CX500, CX500i and CX700 Storage Systems

NDU This upgrade should not be installed on a system that is in any way faulted. This upgrade will not fix a system that is already exhibiting problems.

SP Startup EMC recommends that you reboot an SP only during periods of light I/O activity.

Disk Firmware Downloads

When downloading new disk firmware (a Customer Service procedure only), you must do the following before starting the download operation:

1. Disable the storage-system write cache.

2. Pause all I/O requests to the storage system.

You must download disk firmware to the disks in the following order:

1. Download disk firmware to disk 0_2 and 0_3.

2. Download disk firmware to disks 0_0 and 0_1

3. Download disk firmware to all the other disks in the storage system.

The storage system reboots after each download.

Note 1: Current Seagate drive firmware does not come back online following the download of firmware to other drives and the subsequent reboot of drive 0. If this happens, remove drive 0 and reinsert it into the storage system.

CAUTION If only one SP is running when you download disk firmware, do not start the second SP until the download completes.

24

Technical Notes


DAE2 Firmware Download FLARE software packages include DAE2 LCC controller FRUMON code. Following an NDU, new FRUMON code will automatically install on each enclosure in the storage system. This process takes approximately 10 minutes per enclosure. DAE2-ATA enclosures that do not have any drives inserted will not upgrade. If drives are later inserted, the enclosure must be power-cycled, or the enclosure LCC card disconnected and reconnected in order for the enclosure FRUMON to complete its upgrade.

Note that FRUMON upgrade on diskless enclosures is supported with this revision of software, so upgrades from this release to a future release of software will not have the restriction detailed above.

Soft Logical Sector Data Errors A soft logical sector data error indicates that the data redundancy information for a sector stored on a redundant LUN is not valid. These errors occur because a power failure or FRU failure interrupted a write in progress. FLARE software fully recovers from these errors, and reports them to the Event Log only for informational purposes. When the storage system returns to a fully functional state, the sniff-verify utility finds all redundant data inconsistencies and takes the necessary steps to correct them. A write to a sector with inconsistent redundant information also corrects the inconsistency. These errors are normal in the event of a power failure.

Event Log Entries Appear Out of Order FLARE software may not always enter items into the Event Log in the order that they occurred. This happens only when several events occur at approximately the same time. Time stamps are correct.

Power Failing with LUNs Bound Across DPE/DAE2s LUNs bound across enclosures in a storage system may be marked for rebuild on a power failure because the power to a DPE connected to an SPS stays on until the cache dump is complete. EMC recommends that you do not bind LUNs across enclosures that include a DPE connected to an SPS.

Hot Spare Ownership Once a hot spare is assigned to a degraded LUN, that LUN owns it until either the LUN is repaired and the equalize completes or the LUN is unbound. If the LUN is accidentally double-faulted because you removed the wrong disk during repair, then re-inserting the incorrectly

25

Technical Notes


removed disk should cause the LUN to return to its original state (including continued operation with the hot spare engaged).

DAE2-ATA Drive Ownership and Activity LED Behavior In the DAE2-ATA enclosure, the two FC-to-ATA link controller cards (LCCs) share ownership of the drives based on I/O load. If one FC-to-ATA LCC is removed from the enclosure, the remaining LCC will take and hold ownership of all of the enclosure drives. When this occurs, the drive activity lights will steadily blink at a rapid rate. The activity light behavior will return to normal when the missing LCC card is replaced.

Disk Insertion and Removal When inserting or removing more than one disk from a storage system, you must pause between each removal or insertion to allow the storage system to recover. Before inserting or removing the next disk module you should make sure that

A newly inserted disk's amber light is off and its green light has returned to a normal state: on or blinking due to I/O in progress.

On a disk removal, the green lights for the other disks return to a normal state: on or blinking due to I/O in progress.

The storage system appears to have resumed normal operation.

Database Disk Replacement You should replace a disk in slots 0_0, 0_1, or 0_2 of DPE enclosure 0 only when the storage system is fully operational. The disks in these slots are FLARE software database disks. Each database disk contains an image of the storage-system configuration and current operating status. FLARE software ensures that the newly inserted disk has the correct database information copied to it when it is installed while the storage system is running. If you insert a disk into one of these slots when the storage system is powered off, the correct FLARE software database information is not written to the replaced database disk. This can result in storage-system configuration problems when the storage system is powered up or if another database disk fails. If you install a FLARE software database disk and you find that existing LUNs are not enabled or caching cannot be enabled or disabled, contact your EMC customer service representative.

Software Image Disk Replacement The FLARE software image is stored on disks 0_0, 0_1, 0_2 and 0_3. Disks 0_0 and 0_2 form a mirror with SP A's software, and disks 0_1 and 0_3 form a mirror with SP B's software. You should never remove and/or replace both disks in one of the mirrors at the same time. If you

26

Technical Notes


do, no software will exist to support that SP, and it will become non-functional.

Background Verify Started After NT to XP Conversion After a conversion from a CX200, 400, 600 system to a CX300, 500, 700 system a background verify will be executed on all user LU’s.

Extended Inquiry Info Extended Inquiry information is not available when running with DMP failover mode in a Fibre Open initiator type.

CX300, CX500 and CX500i Storage Systems

SP Will Not Power Up When DPE Power Supply Module Is Unplugged If a DPE power supply module is unplugged when you power up the storage system, it does not come up until you plug the power supply module back in.

CX700 Storage Systems

SPE LCC Failure Management FLARE software is designed to panic with a 0x00800094 code if it is not able to access any disks on its back-end loop. This ensures the integrity of user data when an SP loses access to its vault and FLARE software database disks and initiates a trespass of this SP's LUNs to the peer SP. This panic is a very unlikely event and typically only an SPE LCC hardware failure causes it. Following the panic, the SP attempts to reboot several times, and fails to completely initialize as long as this hardware failure is present. In this situation, the SP remains in a shutdown state. After the LCC is repaired, you need to power-cycle the storage system or remove the SP that panicked and re-insert it (effectively power-cycling the SP) to allow the SP to reboot and return to an operational state.

SP Will Not Power Up When SPE Fan Module Is Unplugged If the SPE fans are unplugged when you power up the storage system, it does not come up until you plug the fans back in.

27

Documentation


Documentation The following table lists the applicable documents for this product release. These documents ship with Navisphere Manager.

Part Number Publication

069001125 EMC Navisphere Manager Administrator's Guide

Installation FLARE software is installed on the storage system at the factory. The 02.16 release supports a new packaging model where the entire FLARE Operating Environment is pre-installed in the factory or available in the upgrade bundle. Optional software components are enabled by purchasing and installing the appropriate enabler package. Please refer to the FLARE Operating Environment Release Notes (P/N 300-001-793) for additional details.

This section describes the following:

Installation restrictions Installation procedures

Please refer back to the Software Commits section for additional information relevant to software upgrades.

Installation Restrictions

This section describes installation restrictions for the various storage-system types.

Note: Hosts running the Microsoft Windows 2000 operating system could be susceptible to data loss during a LUN expansion operation unless Windows 2000 Service Pack 4 is installed. For this reason, it is strongly recommended that all Windows 2000 host servers be upgraded with Windows 2000 Service Pack 4 before this FLARE software is installed. For additional details on the Windows 2000 issue, see EMC Solution ID emc73538 and Microsoft Knowledge Base Article 327020.

Installation Restrictions CX300 CX500 CX500i CX700

Access Logix™ software is pre-installed on your array and will remain on your array after subsequent upgrades of FLARE software. Use Navisphere to enable use of storage groups.

• • • •

28

Installation


Installation Restrictions CX300 CX500 CX500i CX700

When installing a new version of FLARE software, you must not remove an SP or reboot the storage system while the software installation is in progress. Doing so may leave the storage system in a transient state from which it cannot be restarted. Please wait for the entire installation process to complete and automatically reboot.

• • • •

You cannot enable MirrorView™ or SAN Copy™ software on a CX300 and CX500i storage system.

• •

DAE2-ATA enclosures must have at least one disk drive inserted in order to upgrade the enclosure FRUMON code.

This restriction is removed if you are running 02.07.016 FLARE or higher.

• • • •

Installation Procedure

For storage systems where EMC does not support using the Wizard, only a service provider should perform the installation procedure. Refer to the Navisphere Manager Release Notes, P/N 300-001-808, for a list of supported storage systems.

EMC recommends limiting software installation (non-disruptive upgrade) to periods of light I/O activity. To install FLARE software, follow these steps:

1. Verify there are no faults on the array before you begin. If there are faults, repair them before continuing with the upgrade or commit.

2. Locate the packages you will load.

3. Pick a time of lighter I/O load.

4. Pre-bind any new LUNs you may want later if a commit will be delayed for any reason.

5. Consult the appropriate documentation for other recommendations.

6. Use the Wizard where appropriate.

7. Monitor progress.

8. Evaluate the results of the installation.

9. Commit the software when ready. (Note: If drive 0 has been removed or faulted since the software has been installed, do not proceed with the commit. Contact your service provider to repair drive 0 prior to software commit.)

29

Installation


Installation Procedures CX300 CX500 CX500i CX700

Use the Software Installation Wizard found in Navisphere Manager. For more information on the Wizard, refer to Manager’s online help or Chapter 5 in the EMC Navisphere Manager Administrator’s Guide (P/N 069001125). For further information on software installations and upgrades, refer to the EMC CLARiiON Software Installation Guide (P/N 300-001-504). This manual may be found on the EMC Powerlink™ website.

•

Only a service provider should perform the installation procedure.

• • •

30

Troubleshooting and Getting Help


Troubleshooting and Getting Help This section provides specific instructions for contacting EMC Customer Support and obtaining additional information about EMC products.

Where to Get Help

For questions about technical support and service, contact your service provider.

If you have a valid EMC service contract, contact EMC Customer Service at:

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

Canada: (800) 543-4782 (543-4SVC)

Worldwide: (508) 497-7901

Follow the voice menu prompts to open a service call, then select CLARiiON® Product Support.

Sales and Customer Service Contacts

For the list of EMC sales locations, please access the EMC home page at:

http://www.emc.com/contact/

For additional information on the EMC products and services available to customers and partners, refer to the EMC Powerlink website at:

http://powerlink.emc.com

For CLARiiON documentation, after you log in to Powerlink, select Support > Document Library.

http://www.emc.com/contact/

http://powerlink.emc.com

31

Copyright and License Information


Copyright and License Information Copyright ©2005 EMC Corporation. All Rights Reserved.

EMC believes the information in this publication is accurate as of its publication date. However, the information is subject to change without notice.

THE INFORMATION IN THIS PUBLICATION IS PROVIDED "AS IS." EMC CORPORATION 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 EMC software described in this publication requires an applicable software license.

EMC2, EMC, CLARiiON, Navisphere and PowerPath are registered trademarks and Access Logix, FLARE, MirrorView, Powerlink, SAN Copy, and SnapView are trademarks of EMC Corporation. All other trademarks used herein are the property of their respective owners.

Info-ZIP Copyright This is version 2002-Feb-16 of the Info-ZIP copyright and license. The definitive version of this document should be available at ftp://ftp.info-zip.org/pub/infozip/license.html indefinitely.

Copyright (c) 1990-2002 Info-ZIP. All rights reserved.

For the purposes of this copyright and license, "Info-ZIP" is defined as the following set of individuals:

Mark Adler, John Bush, Karl Davis, Harald Denker, Jean-Michel Dubois, Jean-loup Gailly, Hunter Goatley, Ian Gorman, Chris Herborth, Dirk Haase, Greg Hartwig, Robert Heath, Jonathan Hudson, Paul Kienitz, David Kirschbaum, Johnny Lee, Onno van der Linden, Igor Mandrichenko, Steve P. Miller, Sergio Monesi, Keith Owens, George Petrov, Greg Roelofs, Kai Uwe Rommel, Steve Salisbury, Dave Smith, Christian Spieler, Antoine Verheijen, Paul von Behren, Rich Wales, Mike White

This software is provided "as is," without warranty of any kind, express or implied. In no event shall Info-ZIP or its contributors be held liable for any direct, indirect, incidental, special or consequential damages arising out of the use of or inability to use this software.

Permission is granted to anyone to use this software for any purpose, including commercial applications, and to alter it and redistribute it freely, subject to the following restrictions:

Redistributions of source code must retain the above copyright notice, definition, disclaimer, and this list ocnditions.

Redistributions in binary form (compiled executables) must reproduce the above copyright notice, definition, disclaimer, and this list of conditions in documentation and/or other materials provided with the distribution. The sole exception to this condition is redistribution of a standard UnZipSFX binary as part of a self-extracting archive; that is permitted without inclusion of this license, as long as the normal UnZipSFX banner has not been removed from the binary or disabled.

Altered versions--including, but not limited to, ports to new operating systems, existing ports with new graphical interfaces, and dynamic, shared, or static library versions--must be plainly marked as such and must not be misrepresented as being the original source. Such altered versions also must not be misrepresented as being Info-ZIP releases--including, but not limited to, labeling of the altered versions with the names "Info-ZIP" (or any variation thereof, including, but not limited to, different capitalizations), "Pocket UnZip," "WiZ" or "MacZip" without the explicit permission of Info-ZIP. Such altered versions are further prohibited from misrepresentative use of the Zip-Bugs or Info-ZIP e-mail addresses or of the Info-ZIP URL(s).

Info-ZIP retains the right to use the names "Info-ZIP," "Zip," "UnZip," "UnZipSFX," "WiZ," "Pocket UnZip," "Pocket Zip," and "MacZip" for its own source and binary releases.

ftp://ftp.info-zip.org/pub/infozip/license.html

32


Expat Copyright Copyright (c) 1998, 1999, 2000 Thai Open Source Software Center Ltd

Permission is hereby granted, free of charge, to any person obtaining a copy of this Expat software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

EMC is using a modified version of the above software in CX -Series FLARE versions 02.05.x.xx.x.xxx and higher. In accordance with the above license the EMC modified Expat source is available to users of this Expat software. The Expat source may be found here: http://www.emc.com/ClariionOpenSource/

http://www.emc.com/ClariionOpenSource/

300-002-136_A01_elccnt_0

Documents

bridge control

hot spare

navisphere

cx700 storage

cx700 storage

latest flare

drive firmware

peer storage