Re: aac0: COMMAND 0xffffffffxxxxxxxx TIMEOUT AFTER xx SECONDS

On Sun, 11 Jun 2006, Chad Leigh -- Shire.Net LLC wrote:
> On Jun 11, 2006, at 5:41 AM, Chris Hedley wrote:
>> I'm starting to get the impression that the 2410SA's low end design is even 
>> lower than its fairly low-end price would suggest.  Any suggestions for 
>> similarly priced cards with better performance?  The best I can manage for 
>> slot type is 66x64 PCI unless I change my motherboard, which I can't quite 
>> afford to do at the moment...
>
> I have not done any performance testing but I have some LSI MegaRAID SATA-150 
> 4 cards.  Maybe you can find some benchmark comparison reviews.  The Areca, 
> which is a bit more expensive, also gets good reviews.  I have a couple of 
> them but for my Solaris 10 machines and they are just now being set up.  Both 
> the LSI and Areca have FreeBSD drivers.

Must admit I'm tempted by an Areca, even with the high price tag.

I've been having a look at some reviews, but unfortunately few of them 
make it clear whether or not the hard drives' cache is set to write back 
or write through.  Needless to say, I'm not desperately enthusiastic about 
combining a RAID controller with write back caching, but I suspect that a 
lot of controllers are heavily dependent on it being enabled to attain 
their performance: it seems that my 2410SA's rather dismal 3-6 & 30-40MB/s 
respective RAID5 write & read speeds would increase dramatically were I to 
use write-back, but I'm not going there...  I guess my point is that I 
really don't want to find myself with another dog if I buy something with 
apparently superior performance if it's completely reliant on on-disc 
write back caching being enabled.

Does anybody know of any good review sites featuring controllers such as 
those that Chad mentions where the reviewers definitely had on-disc 
write back disabled?  The reason I'm asking is after reading reviews on, 
for example, xbitlabs.com, where the comment "We also enabled lazy writing 
for the hard disk drives" is almost added as an afterthought and isn't 
immediately obvious.  Exactly why they chose to enable it is a matter for 
another debate, I guess...

I'm really desperate to ditch the 2410SA controller: the performance, as 
mentioned, is terrible, and it tends to lock up the entire system with 
various timeouts if anything more than trivial read or write accesses are 
attempted (not sure if this is the case with the latest -current as I've 
been having too many problems with assorted panics with recent kernels to 
test it fully).

Cheers,

Chris.