I have a RAID 10 array on my DP55WG. I am running Rapid Storage Technology in Windows 7 Ultimate 64-bit. I've gotten into the habit of doing a Verify every few days because of some problems I was having. I always come up with 0 blocks with media errors, and I know that’s good, but I also always come up with a fistful (20 to 60) of verification errors. Almost all (or perhaps all) of them are in the first 10% of the verification process. Is this something I should worry about?
Neither the user’s interface nor the help file tells us much about what is going on. It may be normal to get this number of errors. Perhaps they are tabulated, but always corrected by ECC. I’m hoping someone more informed than me will weigh in. If it is something to worry about, what can I do?
I had posted a similar question earlier about RAID 1 after noting verification errors, but there was no clear resolution:
Another thread attacks the same question, with various details and speculation:
Down in the thread, the original poster speculates that maybe...
I believe that these errors I'm getting after the array rebuild pop up in the pagefile area. So doesn't doesn't seem really that critical. Maybe IMSM just catches the errors in the pagefile location taken by itself or something.
Since my system has been running perfectly for two years as far as I can tell, I'm assuming I'm not exeriencing true data errors, and there's a benign explanation. I can imagine some reasons why a software verification of an active disk might run into conflicts that would fail certain confirmations. And I would expect the Intel monitor would clearly announce any "real" errors. But I can't be entirely comfortable without knowing why the verification errors are displayed. Intel Rapid Storage Technology ought to document this display more clearly.
And while I'm griping about the lack of IRST clarity, I'm also interested in the S.M.A.R.T. reports. When a disk in part of an array, the individual S.M.A.R.T. analyzers can't read the data. The Intel web page claims IRST monitors S.M.A.R.T. and will issue a warning when appropriate, and that's reassuring. But it would be nice to have it issue an actual S.M.A.R.T. report.
Thanks for your response! I lean toward benign causes too, but it would be nice to know for sure.
I agree that there should be a lot more clarity on these issues. As an engineer, I have always documented my work very thoroughly. I don't see much value in a good design if it isn't well understood by the users and I could never understand why some designers don't seem to care. It really doesn't take that long to do a good job of documentation.
It seems to me we're asking a pretty basic question here: You run a verification test and you get a result, and we don't know what that result means in practical terms, and Intel seems unable or unwilling to tell us. This doesn't seem like the sort of thing where I'd need to chase after an Intel technician and get a personal answer, assuming that were even possible. Am I expecting too much of Intel, that a question like this should be answered in the doc or FAQs? Is there another source of information that I'm missing, or a better place to post queries? Or maybe very few people care, which is why there's a dearth of hard information.
We absolutely need a reply to this thread as I'm getting the same thing. I would have to agree that I think the problem is benign, but I would like some clarification from whoever helped design IRST.
The help section of IRST says the following:
Verification Errors Found: Reports the number of inconsistencies found during the last volume data verification.
So it found them LAST time I ran a verification check, but does that also mean it corrected them automatically?
Hello Intel... is anyone listening?