S.M.A.R.T. --Failing or Not?

Why don't you talk to Micromat about it. It may be a problem with TT Pro if DU reports the status as "Verified." After all it's their program, not Apple's.

Why not post it to Micromat directly? Don't know who you talked with on MacFixit, but unless he/she is a Micromat programmer I wouldn't rely on advice to contact the drive manufacturer. The drive manufacturer doesn't know TTPro from a hole in the ground nor really care about it. WD will most assuredly tell you their drives comply with the SMART standards.

Even though the SMART status on a drive may say "Verified" or "Failing", that doesn't necessarily mean the drive is either OK or bad. At best SMART's reliability is only about 60 percent. See the following:

http://www.seagate.com/docs/pdf/whitepaper/enhanced_smart.pdf

I just had a brand new drive fail that I discovered by accident when running Disk Utility to repair permissions. And being gun-shy of the replacement going I posted here and was directed to: SMARTReporter that puts a little icon in the menu bar and allows you to set how often it polls the drive(s) displaying their condition. It has been a comfort to have an indicator other than wondering. This could also be a way for you to verify if the drive is in fact failing.

Western Digital drives are really good and the one time I had to call their tech support it was good also. As far as TechTool Pro, I have had nothing but problems and no longer use, even after being a beta tester for Micromat. Their tech support the last time I had to call was horrible. But then this was a while ago and they may have cleaned up their act.

Hope this helps.

G4 FW800 Mac OS X (10.4.4) Totally Stuffed/Max'd Out with lots of great toys.

Yeh, I love SMARTReporter. Glancing up at the little green hard drive. Such a pretty little thing. I don't think I'll care for the red one much though.

I remember reading some posts of TechTool reporting false positives for damaged files and suggesting or conducting unnecessary repairs that in turn did actual damage. This just might be related to that. But when it comes to a drive, one can never be too sure and all avenues of input are valuable.

But since you are getting two good reports from trusted applications and one not . . . I would doubt the TTPro report, but be very attentive and watch the drive, even archive it just in case and forward the error to WD and see what they say.

Like I said I just went through the heart stopping shock of loosing a drive and it's no picnic. So I feel for you.

Hi, Larry, Kappy, and "The Guy" —

Interesting thread. I hope I can add something you'll consider worthwhile...

I'd just returned from reading through the thread you initiated on the MacFixIt MicroMat-sponsored forum yesterday to notice that you posted an excerpt here. Thanks.

Imho, Micromat_Tech3 did you a favor by providing the link to Almico's S.M.A.R.T. attribute "190" thread. Btw, in case you haven't had a chance to search the Micromat forum hosted by MacFixIt, there have been some problems reported with TTP's S.M.A.R.T.-reporting component — but not of this nature.

I'm inclined to think it is a temporal (i.e., it changes during operation) finding from the WD2500JS drive's temperature sensor, as asserrted by wibble in the Almico thread.

I've done a little looking around in Western Digital's (WD's) KnowledgeBase. Beyond the article referenced by wibble (just linked in my previous ¶), you may be interested in these articles from WD:

• The technical specifications for your WD2500JS drive.
• What S.M.A.R.T. attributes are supported by WD EIDE and Serial ATA hard drives? (Article 1087). You may note that the temperature attribute given in this table is 194 rather than 190 — but I believe that's splitting hairs...
• What is S.M.A.R.T. and S.M.A.R.T. failure? (Article 251).
• Macintosh-related FAQs.
You may find other worthwhile info by searching from the Western Digital Service & Support home page.
Fwiw, I agree with Kappy's implicit assertion regarding predictive failure analysis vs. unpredictable failure potential. (Great link to the Seagate white paper, Kappy.) As the Seagate paper says, "Mechanical failures, which are mainly predictable failures, account for 60 percent of drive failure." As conjecture only, I'd say that the variable temperature finding you've observed probably is an ~accurate reflection of the drive's temperature, and oscillates appropriately. I'd consider it a strong indicator of predictive failure if it became a steadily reported out-of-spec. finding. But please keep in mind that's only my opinion.
I think that "The Guy" has a good idea in suggesting an "obvious" S.M.A.R.T. polling/reporting function that doesn't require remembering. Several disk utilities offer such a component, but I don't think they're as "immediate" as what The Guy's described. Does SMARTReporter only provide an overall "summary status" of the S.M.A.R.T. status — like Disk Utility's "Verified" — or is detailed reporting of all the attributes available? The latter is a plus for how TTP does it.
Btw, you may benefit from following this topic in the Using your iMac (Early 2006) forum. A quick search for < temperature sensor S.M.A.R.T. attribute > suggests to me this is an issue that has and will interest others there.
I'll be interested to hear what WD tells you.
HTH! — and
Warm regards to all,
Dean
To "The Guy" —
Do you know whether the drive failure you experienced was something that S.M.A.R.T. coulda/woulda detected and reported? Fwiw, please keep in mind that ~40% of potential failures "aren't predictable" — power surges, etc. The best I know of that we can do with stuff like that is to "repair disk" (directory) after any known interruption of writing to the drive, keep a warranty active as long as possible — and keep our fingers crossed...

[2,050 ⊥ 3,450]

Hi, Larry.

You're welcome. I try to take opportunities as they arise to "keep up" with S.M.A.R.T. as best I can.

You've seen my hypothesis in an effort to answer the question — i.e., that your drive isn't dying: it intermittently gets hot enough to cause the 190 "failed" datum to be reported, then cools again. (Real? a sensor problem? determining which would be a real diagnostic challenge.) In my own lingo, a "predictive" data point of a potential (but not imminent) unpredictable failure. Fwiw.

I'll be interested to hear whether WD ever proves willing to venture an opinion on this without having the drive in hand ( they'll think about issues like OEM, liabilitiy, etc...)

As to why TTP differs, I was implying that, AFAIK, the reason is that TTP provides detailed reporting rather than a single "S.M.A.R.T. status: Verified" datum as others do. Micromat_Tech3 explained the attributes available in TTP's "S.M.A.R.T. Self-Checks" in this thread from last March. I tend to doubt this is a TTP issue — all the software is doing is polling data that the drive sends. I haven't seen documentation of what "decision tree" Disk Utility, SMARTReporter, etc. may use to coalesce all the attributes into a single "failing" or "verified" statement. I'd be ~surprised if they showed "failing" for an "unknown" or temperature value that varies. Imo, that's your answer...

Btw — oops! Sorry — the link toward the end of my post should've been to the Using your iMac G5 forum. I guess your "two months ago" statement overwhelmed "(PPC)" in my aging brain when I read it. Please forgive...

I was checking whether Marcel Bresink's Temperature Monitor is Universal Binary/Intel-compliant when I realized my error. (It is — not that it matters to you.) Anyway, you might want to check out this shareware utility. It also offers "immediate" monitoring. Not knowing the specifics of your sensors, I don't know what it may enlighten — but I think you can "test drive" on a trial basis...

—Dean

Edit: The WD person's reply isn't a big surprise — for Macs in general, but particularly for drives that are marketed as OEM. (i.e., when support is typically binding only on Apple and not on the OEM.) But persist — my bet is that WD will be the source of an answer if you can get to the right tech. person. You could always try contacting AppleCare, and trying to reach the "product specialist" level of expertise there — but they didn't make the drive...
Btw... not to be critical, Larry — but if you already knew all this, why didn't you include/reference it yourself in the beginning of the thread? Perhaps I'm over-reacting to "the closest bit of relevant info that I've found" — but I consider [≥most of] the links provided by myself and others closely relevant, too. Trying to help here, you realize...

Hey, Larry —

Thanks for clarifying "where you're coming from."

Given that context, the closest thing to "a simple answer" is the one AppleCare gave you. S.M.A.R.T. is good at predicting relatively imminent mechanical failure. The intent is to provide an opportunity to save data before significant drive failure.

The "Attribute 190 (Unknown)... out of limits" is very probably either totally in error or an intermittent temperature departure that's not worth worrying about.

If you want simple answers, forget about monitoring stuff like this so closely: don't worry, be happy...

Best of luck,
Dean