• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

another raid card question

J

Just_Visiting

Weaksauce
Joined
Feb 3, 2008
Messages
77
I picked up a new LSI 9260-4i card for my file server build. Then a friend upgraded his file server and sold me his 3ware AMCC 9650SE-8LPML for $80. I'm still within my return period on the unused LSI card. I already have 16 1TB WD's 1001FALS and 8 Hitachi 5400RPM 3TB's waiting to go into my Norco 4224.

1. Am I really missing out on the raid card not being 6Gb/s?
2. Anyone have any major problems with the 9650SE's (I'll probably update the firmware to the most current when I see him next week).
 
Just_Visiting said:
2. Anyone have any major problems with the 9650SE's (I'll probably update the firmware to the most current when I see him next week).
Click to expand...

I have seen them eat arrays. If there isnt extra redundancy (IE anything but raid6) I have seen arrays on this card which will never rebuild even with the ignoreecc option turned on. Performance is ass on these cards. I have seen some people show decent benchmarks but on linux I saw under 30 MB/sec write speed to these cards in raid6.
 
houkouonchi said:
I have seen them eat arrays. If there isnt extra redundancy (IE anything but raid6) I have seen arrays on this card which will never rebuild even with the ignoreecc option turned on. Performance is ass on these cards. I have seen some people show decent benchmarks but on linux I saw under 30 MB/sec write speed to these cards in raid6.
Click to expand...

30MB/sec is sucky. My old perc5 RAID5 array is faster than that. Still got my eye out on an 9260-8i.
 
Guess my questions now is......Any real benefit to me having a 9260-8i over a 9260-4i? I have a pair of HP 300GB 10k RPM SAS drives that I planned on using to boot from. Will the Supermicro X9SCM-F-O be able to recognize them plugged directly into the board. That's the only reason I see for having the 8i over the 4i.
 
From experience I can say the LSI is a great card.

It also works great with the Intel expander (see sig). I would sell the 9650SE and buy the Intel expander. Then you won't ever need to upgrade and raid elements again (as with the expander you'll be set for filling up your Norco).
 
Blackdragon said:
From experience I can say the LSI is a great card.

It also works great with the Intel expander (see sig). I would sell the 9650SE and buy the Intel expander. Then you won't ever need to upgrade and raid elements again (as with the expander you'll be set for filling up your Norco).
Click to expand...

I was actually looking at the new Chenbro CK23601 expander to maximize the 4224 and give me the option to link another enclosure to it if needed.

Will the onboard sata ports on the supermicro x9smc-f-o support SAS drives?
 
Just_Visiting said:
30MB/sec is sucky. My old perc5 RAID5 array is faster than that. Still got my eye out on an 9260-8i.
Click to expand...

From my testing yes a perc5 is faster.

We had to change all these machines to raid10 to get decent performance. Awesomely enough just now I ran into yet another machine that craps out during the rebuild:

# lspci | grep -i raid
09:00.0 RAID bus controller: 3ware Inc 9650SE SATA-II RAID (rev 01)


Code: 
# tw_cli /c0/u0 show ignoreECC
/c0/u0 Ignore ECC policy = on

# tw_cli /c0 show

Unit  UnitType  Status         %RCmpl  %V/I/M  Stripe  Size(GB)  Cache  AVrfy
------------------------------------------------------------------------------
u0    RAID-10   REBUILDING     77      -       64K     3725.25   OFF    OFF

Port   Status           Unit   Size        Blocks        Serial
---------------------------------------------------------------
p0     OK               u0     931.51 GB   1953525168    5QJ0YVDG
p1     OK               u0     931.51 GB   1953525168    9VP11RPF
p2     OK               u0     931.51 GB   1953525168    5QJ0YBTY
p3     OK               u0     931.51 GB   1953525168    6TE04XGE
p4     OK               u0     931.51 GB   1953525168    5QJ0WYWY
p5     OK               u0     931.51 GB   1953525168    5QJ0ZVBQ
p6     DEGRADED         u0     931.51 GB   1953525168    9VP8EQ0Q
p7     ECC-ERROR        u0     931.51 GB   1953525168    5QJ0ZV3P

The p7 drive its reading from only had one sector pending re-allocation yet while rebuilding it gets tons of I/O errors:

Code: 
3w-9xxx: scsi0: ERROR: (0x03:0x0202): Drive ECC error:port=7.
sd 0:0:0:1: [sdb] Unhandled sense code
sd 0:0:0:1: [sdb] Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE
sd 0:0:0:1: [sdb] Sense Key : Medium Error [current]
sd 0:0:0:1: [sdb] Add. Sense: Unrecovered read error
sd 0:0:0:1: [sdb] CDB: Read(10): 28 00 00 24 63 9a 00 00 08 00
end_request: I/O error, dev sdb, sector 2384794
3w-9xxx: scsi0: ERROR: (0x03:0x0202): Drive ECC error:port=7.
sd 0:0:0:0: [sda] Unhandled sense code
sd 0:0:0:0: [sda] Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE
sd 0:0:0:0: [sda] Sense Key : Medium Error [current]
sd 0:0:0:0: [sda] Add. Sense: Unrecovered read error
sd 0:0:0:0: [sda] CDB: Write(10): 2a 00 00 0d c7 df 00 00 10 00
end_request: I/O error, dev sda, sector 903135
Buffer I/O error on device sda1, logical block 112884
lost page write due to I/O error on sda1
Buffer I/O error on device sda1, logical block 112885
lost page write due to I/O error on sda1
3w-9xxx: scsi0: ERROR: (0x03:0x0202): Drive ECC error:port=7.
sd 0:0:0:0: [sda] Unhandled sense code
sd 0:0:0:0: [sda] Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE
sd 0:0:0:0: [sda] Sense Key : Medium Error [current]
sd 0:0:0:0: [sda] Add. Sense: Unrecovered read error
sd 0:0:0:0: [sda] CDB: Write(10): 2a 00 00 88 b3 af 00 00 08 00
end_request: I/O error, dev sda, sector 8958895
Buffer I/O error on device sda1, logical block 1119854
lost page write due to I/O error on sda1
3w-9xxx: scsi0: ERROR: (0x03:0x0202): Drive ECC error:port=7.
sd 0:0:0:1: [sdb] Unhandled sense code
sd 0:0:0:1: [sdb] Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE
sd 0:0:0:1: [sdb] Sense Key : Medium Error [current]
sd 0:0:0:1: [sdb] Add. Sense: Unrecovered read error
sd 0:0:0:1: [sdb] CDB: Read(10): 28 00 9d 2b 0f a2 00 00 08 00
end_request: I/O error, dev sdb, sector 2636844962
sd 0:0:0:0: [sda] Unhandled sense code
sd 0:0:0:0: [sda] Result: hostbyte=DID_OK driverbyte=DRIVER_SENSE
sd 0:0:0:0: [sda] Sense Key : Medium Error [current]
sd 0:0:0:0: [sda] Add. Sense: Unrecovered read error
sd 0:0:0:0: [sda] CDB: Read(10): 28 00 00 5f 8b ff 00 00 08 00
end_request: I/O error, dev sda, sector 6261759

It has a 80 GB boot slice which is why its sda and sdb. Oh and so i pull slot 6 so its no longer rebuilding and reboot the machine and no more I/O errors. WTF 3ware!

I hate 3ware and its crap.
 
i have 2 x 3ware 9650se 24ml cards in my i7 system running w7 64bit.. firmware and drivers are upto date.

Running with no problems transfer speeds range from 250-350mb/sec transferring from raid5 to another raid 5

system is running 5 x raid5 (8 drives per array).

90% of drives are seagate, 10% are hitachi 2tb hard drives...

This year i have had to repalce 3 drives under rma due to sector repairs or just dieing.
Other than that no problems at all....

Regs John
 
You must log in or register to reply here.
Back
Top