

	    Firmware V3.31_ Release Note for EonRAID 2510FS Series

Apply to the following Infortrend RAID controllers:
	EonRAID 2510FS-4S 
	EonRAID 2510FS-6S
	EonRAID 2510FS-4D
	EonRAID 2510FS-6D
	EonRAID 2510FS-4RH
	EonRAID 2510FS-6RH.

Apply to Infortrend RAIDWatch 1.51X9.1

===================================================================================
3.2x ==> 3.31M11
By Infortrend Technical Support 12/31/03
===================================================================================

New Functionality in 3.31_:
-----------------------------
.  On-Line Create logical drive
	LD is immediate ready for Host I/O,
	after LD Status changed from CREATE to INIT.
	it has a background task to initialize until entired LD is init done.
	except LD is a INCOMPLETE array or BAD array.

.  On-Line Expand logical drive capacity
	LD capacity Expand immediate ready for Host I/O
	after LD status changed from EXPAND TO INIT
	it has a background task to initialize until en
	tired LD is init done.
	except LD is a INCOMPLETE array or BAD array.

.  Multiple Rebuild / Expand
	Per LD can do a Rebuild/CheckP and/or a Expand/Init/AddScsi operation at same time.
	but CheckParity/Rebuild are only one of can be take at a time.
	    Create/Expand/AddDrv are only one of can be take at a time.
		
.  Support for variable striping sizes
	selectabe striping size at LD create time for individual LD
	4K,16K,32K,64K,128K,256K
	Notes:
	    128K is optimize for Sequential I/O in old version.
	     32K is optimize for Random I/O     in old version.

.  Write Through / Write Back Variable
	Per-LD Write Policy can be in different setting.
	and can be on-line modified

.  AddScsi are executable on degrade mode.

.  Media Scan
	scan media if found bad blk then immediate recover it from
	other redundant drive. (only RAID1/3/5)
	it look like Rebuild, but only the Bad Blk is Rebuild.
        when reconstruct write is taken, it will do 'Write Verify'.

.  To protect controller changed (controller damanged) when AddDrv paused

.  Degraded-mode bad sector handling in Rebuild/AddScsi/Host Write/CoverP
        
.  Bad sector Handling in RegenerateParity
	(Rebuild/AddScsi/Host Write will auto recover)

.  In degraded mode, pass error info through to host on reads 

.  Support for more than 31 physical drives in a LD

.  V3.31 Logical Volume, Structure is differs from before version.
   but V3.31 still support it.

. Maximum LD size 64TB when Optimize for Sequential 
                  16TB when Optimize for Random

. Support More than 8 LD/LV 
 
. support write back / through dynamic switch

. set default trigger event for write back/through dynamic switch

. not support rolling upgrade firmware from 3.27 to 3.31

. support Synchronize FTP File System to Formatted drives.

. Speed up FTP File writes.

. LD Regenerate Parity (RAID 1/3/5):
    add 2 optional parameters: 
    1. Events Generated when inconsistency encountered.
        enable to prompt Regeneration Event Message.
        there are 4 types event.
            a "Inconsistent Parity Encountered Block: xxxx", 
		where xxxx is base sector No.of corrsponding stripe.
            b."Inconsistent Parity Block xxxx Regenerated", 
		where xxxx is base sector No. of corrsponding stripe.
            c."Media Error Encountered During Check Parity ssss", 
		where ssss is the sector No. of bad sector of  the LD.
            d."Media Error During Check Parity ssss Recovered" , 
		where ssss is the sector No. of bad sector of the LD.

        notice:  if 2 bad blk are encountered on same stripe, 
		 then it will only have "MediaError" event
                 but no 'Media Error recovered" 

    2.  Parity Regeneration not performed when inconsistency encountered.
         "Overwrite Inconsistent Parity - Enabled"
              regenerate Inconsistent parity when encountered, also
              recovere the media error block when encountered.

         "Overwrite Inconsistent Parity - Disabled"
             will not do regenerate Inconsisten parity or 
	     recover the media error block when encountered.

   Note: when "Overwrite Inconsistent Parity" and "Prompt Event Message" both disabled.
        Regenerate Parity will stop immediately,  when Inconsistency or media error encountered.

. RS232 modification:
-> Drive side parameters:
   'Reset at power-up' item is removed.

-> Added support for http redirect to ssh 2.0 telnet. 
   I. java runtime: 1.4.2 or above
   II. usage - http://ip address/ssh.html		

-> add  fibre manual LIP function for EonStor F16F, in 'View and Edit SCSI Channel' 

-> change 'Event Trigger Operation', 
   'Temperature exceeds threshold'  not only detect CPU temperature,
    and Board temperature also detect need.

-> no more support Reserved space for "Backward Compatible(64K)"

-> Changed default telnet timeout from 25 minutes to no timeout, but can
  setting from NVRAM.

-> Do not check 'SMART' feature enable or not when adding 'Clone Fail Drive' item.

-> Add an one-column (*) indicator to "Partition/Size/Offset" list to indicate whether a partition has been mapped.

-> Add "Periodic Cache Sync Flush" option via RS-232 UI.

-> Remove bracketed codes from event messages.

-> Add "Page-Up/Down(Ctrl+P/Ctrl+N)" keys to various lists in RS-232 Terminal Emulation.

-> Add "Logical Drive Shutdown" option in RS232 UI.
