Lord Xeb wrote: Give me a few days and I can get all the data you need.
I can't tell what your expectations are in your post, so let me be clear - I've given suggestions above, about where I would start to look for any evidence that drive 2 has a problem, if I was in your situation, and if I wanted to try to find such evidence.
However your comment about getting "all the data you need" might mean you're expecting that I'll keep working on this until there is a definite conclusion - that isn't the case.  It is possible that no evidence of a problem will be found in the areas which I have suggested - which would then be inconclusive as to what problem (if any) did occur when your system was slow.
My suggestions also assume (as networks pointed out) that investigating "drive 2" is what you are asking - but I can't see a clear statement that this is your goal.
It also seems that you actually may not be able to supply some of the info that I suggested looking at, since you mention that you're re-installing that system. Specifically, have you still got the Windows event log from the time when you report that RAID volume was slow, and any other logs / evidence from that time, from the Intel RAID functionality (which you will need to investigate, since I don't have any experience with that type of RAID)? Or are you hoping that you'll be able to recreate the problem again, and gather the logs at that time?
Just to be accurate, in the title of this thread, you report that there is "1 dead" drive (your "drive 2"). No, there isn't a dead drive. The fact that you could extract its SMART data shows that it isn't dead.
In fact it is quite common for RAID controllers (of many different types that I've worked with over the years), to declare a drive as "dead" or "failed" or whatever term that specific controller uses, when the drive is not really dead or faulty, and the problem is actually elsewhere - or there was no actual fault, just a drive behaviour which the RAID controller didn't like (e.g. drive taking a long time performing internal read recovery retries, and the RAID controller reached its timeout limit, and gave up waiting - you can read about TLER & ERC if you're interested in this type of potential issue, but I have no idea if it applies to the Intel RAID on your motherboard).
Until you can find more info about exactly what problem the Intel RAID "thinks" it detected (using whatever error logging it does), then the only thing I can suggest (which is why I suggested it!) is to start to look for evidence of typical drive problems in the places that I mentioned - but it's quite a shot in the dark, without you having more info about specifically what problem you're looking for.
Good error logging is one of the features you pay for with business RAID systems, in order to be able to quickly identify the cause of problems, and it is one of the limitations that I see with smaller SOHO RAID systems. If the behaviour you observed didn't leave any error messages (or they have been erased), and the problem can't be reproduced, then there may not be evidence to reach a conclusion about what happened.
|
Hopefully he'll clarify what assistance he is requesting...