Overclock.net › Forums › Components › Hard Drives & Storage › RAID Controllers and Software › PERC 5/i RAID Card: Tips and Benchmarks
New Posts  All Forums:Forum Nav:

PERC 5/i RAID Card: Tips and Benchmarks - Page 136

post #1351 of 7150
Thread Starter 
Link to latest Dell firmware then?

I believe the last one was a few months ago?
Once again...
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7 920 [4.28GHz, HT] Asus P6T + Broadcom NetXtreme II VisionTek HD5850 [900/1200] + Galaxy GT240 2x4GB G.Skill Ripjaw X [1632 MHz] 
Hard DriveOSMonitorKeyboard
Intel X25-M 160GB + 3xRAID0 500GB 7200.12 Window 7 Pro 64 Acer H243H + Samsung 226BW XARMOR-U9BL  
PowerCaseMouseMouse Pad
Antec Truepower New 750W Li Lian PC-V2100 [10x120mm fans] Logitech G9 X-Trac Pro 
  hide details  
Reply
Once again...
(13 items)
 
  
CPUMotherboardGraphicsRAM
i7 920 [4.28GHz, HT] Asus P6T + Broadcom NetXtreme II VisionTek HD5850 [900/1200] + Galaxy GT240 2x4GB G.Skill Ripjaw X [1632 MHz] 
Hard DriveOSMonitorKeyboard
Intel X25-M 160GB + 3xRAID0 500GB 7200.12 Window 7 Pro 64 Acer H243H + Samsung 226BW XARMOR-U9BL  
PowerCaseMouseMouse Pad
Antec Truepower New 750W Li Lian PC-V2100 [10x120mm fans] Logitech G9 X-Trac Pro 
  hide details  
Reply
post #1352 of 7150
http://support.dell.com/support/down...id=-1&impid=-1

Dell recommended this, and I flashed it last night...
post #1353 of 7150
Quote:
Originally Posted by bbarnett View Post
DuckieHo,

The only way to find other bugs, is to lose data.

Of course, some of those bugs could cause gradual corruption of data as well.

Do people really want to be a guinea pig with their data? Why take a rock solid card, and then turn around and run it in an unstable configuration?

Why risk it? Do you *need* LSI Logic tools?!

Weigh the risk.
You do realize that dell is going to say that the LSI stuff is unsupported for several reasons.

1) They do not make the LSI firmware, and can't attest to it's integrity
2) They want as little liability as possible
3) They will only be liable to things they have complete control over...ie. their card, their firmware

This is more of a political game than anything else. We tell people all the time that stuff isn't supported on our network. Even though we know for a fact that it works just fine.
Example: Game consoles are not supported on our network. We are required to tell people this when they ask. Does it work? Yes, it works just fine! But we still don't support consoles when they can't connect to the internet. We only support computers who can't connect to the internet.

The fact that so many of us have been using the LSI firmware for this long attests to the fact that is solid firmware.
Edited by bruestle2 - 3/3/09 at 12:07pm
My Pwny!
(13 items)
 
  
CPUMotherboardGraphicsRAM
Phenom II 920 Asus M3A79-T Deluxe 6950 2GB 8GB @ 800Mhz 
Hard DriveOSPowerCase
Perc 5/i Raid 0 w/ 2x 320GB + Raid 5 w/ 6x 750GB Windows 7 x64 Pro Antec 850W Antec 1200 
  hide details  
Reply
My Pwny!
(13 items)
 
  
CPUMotherboardGraphicsRAM
Phenom II 920 Asus M3A79-T Deluxe 6950 2GB 8GB @ 800Mhz 
Hard DriveOSPowerCase
Perc 5/i Raid 0 w/ 2x 320GB + Raid 5 w/ 6x 750GB Windows 7 x64 Pro Antec 850W Antec 1200 
  hide details  
Reply
post #1354 of 7150
Quote:
Originally Posted by bruestle2 View Post
You do realize that dell is going to say that the LSI stuff is unsupported for several reasons.

1) They do not make the LSI firmware, and can't attest to it's integrity
2) They want as little liability as possible
3) They will only be liable to things they have complete control over...ie. their card, their firmware

This is more of a political game than anything else. We tell people all the time that stuff isn't supported on our network. Even though we know for a fact that it works just fine.
Example: Game consoles are not supported on our network. We are required to tell people this when they ask. Does it work? Yes, it works just fine! But we still don't support consoles when they can't connect to the internet. We only support computers who can't connect to the internet.

The fact that so many of us have been using the LSI firmware for this long attests to the fact that is solid firmware.
I'm going to be rude here, because.. you're playing with people's data. You're willing to take that risk for yourself, and that's fine. However, when you council other people to take that risk, when talking out of your ass, that is most certainly not fine.

This firmware is NOT solid. I've run into one of the bugs it has. Boom, instantly that turns your statement of 'solid' into utter and complete crap.

I've experienced that bug, and spent time debugging it. In the course of debugging it, both I and the Dell engineer learned things. One of those things learned, is that this bug is most likely an issue with a pointer overflow, because 2*32 is 4TB. This code is most simply not sanitized.

Yes, Dell is going to tell you not to use an unapproved firmware. However, I was having a frank discussion with a Dell engineer, and he thought of many sanity checks that are not being done. The Dell firmware and the LSI logic firmware are not the same, after all. The Dell firmware is designed to work with the Dell bootloader, and other code on that card that isn't be replaced or flashed.

Lastly, you don't know if everyone else is doing fine. There are how many people using this firmware on this card? 1000? 500? 100? How many hours of this code running, does it take for a bug to be exposed? Files could be slowly corrupted on your disk, and you'd never know it for YEARS in some cases.

How many people here have been checksumming their files constantly, to look for slow corruption? Let's take movie files. Let's say you get a corrupt block. You likely won't even notice when watching it.

Worse, you could be the person that hits another bug with this combo, one that ends up with a wiped disk.

Honestly, you can't really believe that a few hundred people running a firmware means it is stable, can you?
post #1355 of 7150
Quote:
Originally Posted by bbarnett View Post
I'm going to be rude here, because.. you're playing with people's data. You're willing to take that risk for yourself, and that's fine. However, when you council other people to take that risk, when talking out of your ass, that is most certainly not fine.

This firmware is NOT solid. I've run into one of the bugs it has. Boom, instantly that turns your statement of 'solid' into utter and complete crap.

I've experienced that bug, and spent time debugging it. In the course of debugging it, both I and the Dell engineer learned things. One of those things learned, is that this bug is most likely an issue with a pointer overflow, because 2*32 is 4TB. This code is most simply not sanitized.

Yes, Dell is going to tell you not to use an unapproved firmware. However, I was having a frank discussion with a Dell engineer, and he thought of many sanity checks that are not being done. The Dell firmware and the LSI logic firmware are not the same, after all. The Dell firmware is designed to work with the Dell bootloader, and other code on that card that isn't be replaced or flashed.

Lastly, you don't know if everyone else is doing fine. There are how many people using this firmware on this card? 1000? 500? 100? How many hours of this code running, does it take for a bug to be exposed? Files could be slowly corrupted on your disk, and you'd never know it for YEARS in some cases.

How many people here have been checksumming their files constantly, to look for slow corruption? Let's take movie files. Let's say you get a corrupt block. You likely won't even notice when watching it.

Worse, you could be the person that hits another bug with this combo, one that ends up with a wiped disk.

Honestly, you can't really believe that a few hundred people running a firmware means it is stable, can you?
If they are running systems that hold extremely important data, they no they shouldn't use the LSI firmware, but for the people here who just want a fast raid card, using that firmware is fine.

btw i never updated my firmware, just because i saw no reason.
post #1356 of 7150
Quote:
Originally Posted by bbarnett View Post
I've experienced that bug, and spent time debugging it. In the course of debugging it
Does this happen when the array is empty or full?
I have 8x640GB in RAID5 ...
post #1357 of 7150
Which version of the LSI firmware were you using when you encountered the bug?
post #1358 of 7150
Quote:
Originally Posted by xpeh78 View Post
Does this happen when the array is empty or full?
I have 8x640GB in RAID5 ...
This bug has not been fully debugged!

Certainly Dell or LSI are not going to waste their time on such a silly bug. A bug caused by using the wrong firmware with the wrong card.

Most of what we know are possibilities, not certainties. We do know though, 100%, that this bug did not happen until my raid5 had 6 1TB drives. When I had 5 1TB drives, it did not occur.

This could mean it is a 32bit/64bit issue, having to do with problems and the different software on the card conflicting with the LSI firmware. It could have to do with filesystem having grown to the point, that the bug was discovered.. as it was overwriting part of my ext3 superblock (which got larger after a filesystem grow).

Frankly, what we do know is that it was not OS related, that it was the card causing the issue, and that the issue is completely reproducible with LSI's firmware, and not with Dell's.

Are you somehow benefiting immensely from using that LSI firmware? That is, do you now drop gold nuggets, where as before you only shat silver?

Why do you want that firmware? Why is the card 'crappy' with the Dell firmware, and 'incredibly awesome dude!' with the LSI logic? Are you actually using any of the improvements that firmware provides, in real life?

Do you really have useless data on that card?

This card is a GREAT deal. We should all thank DuckieHo for brining this incredible deal at ebay to our attention. This card ROCKS. However, it is still an INCREDIBLE deal as it sits, raw and virgin!
post #1359 of 7150
Quote:
Originally Posted by bbarnett View Post
This bug has not been fully debugged!

Certainly Dell or LSI are not going to waste their time on such a silly bug. A bug caused by using the wrong firmware with the wrong card.

Most of what we know are possibilities, not certainties. We do know though, 100%, that this bug did not happen until my raid5 had 6 1TB drives. When I had 5 1TB drives, it did not occur.

This could mean it is a 32bit/64bit issue, having to do with problems and the different software on the card conflicting with the LSI firmware. It could have to do with filesystem having grown to the point, that the bug was discovered.. as it was overwriting part of my ext3 superblock (which got larger after a filesystem grow).

Frankly, what we do know is that it was not OS related, that it was the card causing the issue, and that the issue is completely reproducible with LSI's firmware, and not with Dell's.

Are you somehow benefiting immensely from using that LSI firmware? That is, do you now drop gold nuggets, where as before you only shat silver?

Why do you want that firmware? Why is the card 'crappy' with the Dell firmware, and 'incredibly awesome dude!' with the LSI logic? Are you actually using any of the improvements that firmware provides, in real life?

Do you really have useless data on that card?

This card is a GREAT deal. We should all thank DuckieHo for brining this incredible deal at ebay to our attention. This card ROCKS. However, it is still an INCREDIBLE deal as it sits, raw and virgin!
Well, It sounds like you should probably talk to LSI about the issue. They would have more insight on the problem and might even be able to fix it assuming it is an issue with their firmware.
My Pwny!
(13 items)
 
  
CPUMotherboardGraphicsRAM
Phenom II 920 Asus M3A79-T Deluxe 6950 2GB 8GB @ 800Mhz 
Hard DriveOSPowerCase
Perc 5/i Raid 0 w/ 2x 320GB + Raid 5 w/ 6x 750GB Windows 7 x64 Pro Antec 850W Antec 1200 
  hide details  
Reply
My Pwny!
(13 items)
 
  
CPUMotherboardGraphicsRAM
Phenom II 920 Asus M3A79-T Deluxe 6950 2GB 8GB @ 800Mhz 
Hard DriveOSPowerCase
Perc 5/i Raid 0 w/ 2x 320GB + Raid 5 w/ 6x 750GB Windows 7 x64 Pro Antec 850W Antec 1200 
  hide details  
Reply
post #1360 of 7150
Quote:
Originally Posted by bruestle2 View Post
Well, It sounds like you should probably talk to LSI about the issue. They would have more insight on the problem and might even be able to fix it assuming it is an issue with their firmware.
Why would LSI waste one second of time on this problem? Their firmware works perfectly fine with their bootloader and other associated software on their card.

It does not work perfectly with the Dell boot loader, and software on the Dell branded card. This isn't a bug at all!

Are you actually suggesting that LSI should fix this config, for the paltry few people on this board.. just 'cause? Then, after fixing that bug, they need to suddenly perform quality assurance testing on all Dell branded cards, with their Dell bootloaders.. every time they update their own firmware?

Do you think LSI should suddenly start to support Dell's bootloader too?? Guess what, they won't, because if they were willing to do that in the first place, they'd have done it for Dell, and Dell would not have to release their own firmware!

Come on man, give it a rest, heh!
Edited by bbarnett - 3/3/09 at 2:26pm
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: RAID Controllers and Software
Overclock.net › Forums › Components › Hard Drives & Storage › RAID Controllers and Software › PERC 5/i RAID Card: Tips and Benchmarks