Format Seagate For Maceverye

• Format Seagate For Mac And Windows
• Format Seagate Drive For Mac
• Format Seagate For Mac And Pc
• How To Format Seagate For Mac

Yet again another graphics firmware update apparently my imac doesn't need even though the description says for iMacs with lion,which i am running on my imac. Checked with software update and it wasn't found,but you can't always trust software update bc i have found updates i do need even though they didn't show up in SO and yes they were not hidden! even Seagate and WD drives improves. Generally though folks with critical information will use a raid system. A raid will combine multiple separate drives into one virtual drive that increases your reliability. Also protecting your drive from sudden movement while in. How to re-format a Seagate external hard drive for both Mac and a PC. For some reason I lost half of my space on this hard drive once I formatted it. Seagate internal drives must be formatted before use. Drives that will be bootable are formatted during the process of installing the operating system (ie, Windows or macOS). If you have a drive that is formatted for a different type of computer or a drive that is not preformatted, you will need to format the drive before you can use it.

item.176325

David Charlap

Thanks much for the interesting discussion.

To James Cutler: You were right. My case didn't recognize a 3TB drive. It showed up as 800GB. After some Googling, it appears that Vantec has silently upgraded their bridge boards over the years. Version 1.02 or later is needed to support 3TB drives. I picked up a new case (same model, but version 1.03), and it worked correctly.

And since I now have a leftover case that can't take a 3TB drive, I bought a new 1TB Seagate Constellation CS drive (for $100) to go into that case. It is running as the Time Machine volume for my wife's laptop (which previously had no real backup strategy.)

Sadly, the Vantec NST-360UFS is a $70 case - mostly because of the FireWire port. USB-only cases are much less expensive, but I wanted FW.

Several people have mentioned various brands of external drives. I don't have personal experience with these, but given various posts over the years, I wouldn't bother with any. Many models (especially LaCie) have serious problems with power supplies. And others (like some WD MyBooks) seem to require proprietary software drivers. A lot also seem to have locked firmware, so the internal drive can't be used with any other case, and the case can't be used with a replacement drive. For all these reasons, I have always built my own external drives - buying an internal drive and matching it up with a case. It usually doesn't cost much more (sometimes costs less), requires no special software, and it's easy to swap out the drive or case, should one or the other fail.

To Tracy Valleau: Yes, I've heard people say the same thing. I hope nobody expects a 1M-hour MTBF drive to last 100 years - that's clearly nonsense. But it is reasonable to use the MTBF numbers for comparison purposes - a drive with a higher MTBF will (probably) last longer than one with a lower MTBF. This will be especially true for drives from a single manufacturer and nearby model-years, but it should also be somewhat accurate across different manufacturers.

In the not-too-distant past, every manufacturer published MTBF numbers. Enterprise drives had higher numbers than desktop drives, but all were pretty good. 5 years for a drive was expected and many drives (especially those running continuously, in my experience) would last longer. The MTBF numbers then were usually in the 500,000-800,000 hour range. Today, those kinds of MTBFs are only published for enterprise drives, and they don't publish the figures at all for desktop drives. Given the anecdotal evidence I've heard from many people, I think that's because they don't have very much quality control for desktop drives anymore - they prefer to ship everything and let the warranty replacement process handle defects.

Interestingly, Western Digital claims to not publish MTBF at all (even though the ad copy for their 'Re' drives claims 1.4M hours. See How drive reliability is measured and the MTBF of WD drives, where they say they prefer to publish 'component design life' and 'annualized failure rate' numbers, which, while possibly more applicable to real-world experiences, make it impossible to compare two different drives from different manufacturers. Saying that a drive is intended to last 5 years and has an 0.8% annualized failure rate (according to the above-linked FAQ) doesn't really sound that good to me. It tells me that they consider it OK for 4% of all drives to fail within 5 years and for all drives to fail in the 6th year.

Just another case of 'figures don't lie but liars can figure'. MTBF may not be the most reliable estimate, but it's at least a standard figure that does a reasonable job of comparing reliability of similar products, and is a whole lot better than marketing promises and total silence.

item.176296

MacInTouch Reader

Re:

'Seagate handled the exchange painlessly, except I now have 2 refurbished drives'

When a new item arrives defective or fails immediately, it is much better to return it to the seller and get a new replacement, rather than returning it to the manufacturer for service or replacement with a refurb.

item.176308

Emory Lundberg

Colleen Thompson writes:

'Let's see, we send them new, and they send us reconditioned?'

I have returned failed drives to every manufacturer, and from what I can remember, they all replace with a refurb.

Of course they do, if you send them in for RMA. I wouldn't RMA a brand-new disk that arrived as a failure, I'd return it to my merchant and get another brand-new one. I'm not sure it would even occur to me to RMA a brand-new anything.

(As a datapoint, I had an RMA swap-a-thon with Seagate and 1TB SATA disks about two years ago; I ended up at the end of the whole experience having two refurb'ed disks and one brand new 1TB disk from a different product line when I was done. This was during Thailand flooding, so may have been an extreme edge case, and I certainly didn't expect it.)

item.176317

Skot Nelson

Re:

I stay clear of desktop Seagates. I have a 150% failure rate with Seagates. How do you get above 100%? 50% of the replacements die, as well.

This isn't very helpful without having some information about the timeframe involved...

item.176319

MacInTouch Reader

Sadly, it's not uncommon for new 'in the box' HDs to fail soon after installation. So, whenever I build a system or replace an HD, I always burn in and test it before use. Here are the steps I take:

1. Attach the new HD directly to a SATA port on a Linux box I use just for this purpose, and boot up. I've found over the years that Linux is simply the best OS for this purpose, because it does not attempt any automatic operations on the new HD, nor will it power down the drive during burn in.

2. Let the HD idle with power on for 24 hours.

3. Run

dd if=/dev/zero of=/dev/sdx bs=4096

This writes zeroes to the entire HD surface. Although new HDs appear to be zeroed, nothing has ever actually been written on the HD, so undetected surface defects may exist. Better to find them now.

4. Run

smartctl --test=long /dev/sdx

This will cause the drive to self test, including an attempt to read every block on the HD.

5. Run

smartctl --all /dev/sdx | less

Should the report show self test failure of any sort, the HD is bad and should be rejected. Pay attention to reallocated block counts. At this point in its life, any reallocated blocks indicate defects that will only get worse -- the HD should be rejected.

I always RMA defective HDs as DOA. The vendor I use has always replaced DOA HDs from new stock. Of course, this requires that you test HDs within your vendor's return window; otherwise, you'll be stuck with having to RMA to the manufacturer, a process that takes longer and usually results in a 're-certified' HD.

item.176321

MacInTouch Reader

I just wanted to add that, in my experience, HDs from all manufacturers are suspect until thoroughly tested. I no longer have a favorite manufacturer; they all exhibit a certain amount of 'suck'.

item.176361

MacInTouch Reader

Re: instructions for testing new hard drives from Linux command line:

Would these steps actually be less useful if executed from the Mac command line? In that case, would the Mac's higher-level utility functions still be more intrusive into the process than Linux would?

Alternatively, what about executing these steps from a Linux 'appliance' VM in VMware Fusion? I'm thinking of how you can tell the VM to 'own' the drive -- presumably so it doesn't even show up to the host. (I've never actually tried that.) Would that work as well as running the testing steps from a 'real' Linux box?

item.176364

Marc Marshall

I find it a little odd that nearly everyone keeps talking about 'hard drive manufacturers' as if there are a lot of them to choose from.

Assuming we're talking about rotating-platter hard drives here, there are only 3 companies remaining that manufacture them, and if we're talking about desktop-grade 3.5' drives, one of the three (Toshiba) just got into the form factor, and its products are uncommon at best.

There's Western Digital and Seagate -- that's pretty much it. HGST (formerly Hitachi, IBM before that) is a wholly-owned subsidiary of WD. Samsung sold their hard drive division to Seagate. Maxtor sold to Seagate years ago. Fujitsu sold to Toshiba, and they only made enterprise and mobile drives anyway.

The third remaining player, Toshiba, makes mostly 2.5' hard drives. They recently bought a factory (I believe from WD) and started making a line of budget-oriented 3.5' drives, but they're not something you're going to see on the shelf unless you really go out of your way to buy one.

It's a different matter if we're talking about SSDs, of course, although even there there are a limited number of NAND chip and controller manufacturers, so most of the brand names are really just assemblers (though some may have custom firmware).

item.176351

Riley Werts

Thanks to 'MacInTouch Reader' for posting his procedure for verifying new hard drives within Linux. As I am in charge of equipment maintenance as part of my work, I would very much like to find, and cannot help but suspect that there is available, a Mac-based testing routine that can accomplish the same on our favorite platform. An actual application that can be run, and can then outline its findings, would be ideal. I'm not afraid of Terminal, though, and a compendium of Terminal commands that do the same would be acceptable. Maybe something along the lines of 'Applejack'?

Anyone have any ideas? In the meantime, thanks to all for this conversation!

item.176338

David Grant

Fom an earlier part of this thread:

(That is an over-simplification, but basically correct.)

You can almost work it backwards: if a drive is warrantied by the manufacturer for 5 years, and the MTBF is 1 million hours, then there were likely (114/5) or about 22 drives used in the test. This gibes nicely with the Google drive life report.

MTBF serves as an indicator of reliability vs other similar products.

It seems to me that knowing MTBF means essentially nothing unless you also know how many drives were in the test and how long the test actually ran. From the above 2 drives, each run for 100 hours when 1 fails, gives a MTBF of 200 hours. 100 drives run for 2 hours each when 1 fails gives a MTBF of 200 hours. Without knowing the details of the test I'm not sure how one would use the MTBFs to compare different products.

item.176353

A Kaleberg

Format Seagate For Mac And Windows

According to the book Turing's Cathedral, when von Neumann's group was building the ENIAC, they were very concerned about vacuum tube reliability. There were all sorts of tubes with all sorts of ratings. Since they were going to need thousands, they decided to do some testing. It turned out that the most reliable tubes were the most common ones, a basic dual triode (I believe) that was manufactured by the millions over the years. It beat out just about everything else, including a variety of military spec and pre-burned in tubes. It was also cheaper, for the obvious reasons.

I'm not sure how applicable this is to disk drives, especially since the technology is still evolving rapidly. By the time ENIAC was being built, vacuum tube technology was quite mature and soon to be replaced by the as yet to be invented transistor.

I'll also throw in a link to Google's analysis of disk drive reliability:

There main conclusions were that the manufacturer mattered, heat didn't seem to matter and, yes, some SMART parameters could be used to predict disk failues.

item.176352

Jonathan Duke

A MacInTouch reader suggested using a Linux box to verify new drives.

A simpler solution is to use SoftRAID 4's Disk Certification tools to test each disk before using it.

I have an enclosure with a fan that I mount each drive in and test it for 3-6 passes (depending upon what the final use might be).

SoftRAID is connection-agnostic - works with drives via USB, FireWire & Thunderbolt just fine.

In fact, I was able to certify 8 drives simultaneously in a Thunderbolt case in less time than 2 drives in a FireWire case.

Highly recommended, and I have yet to have a certified drive fail prematurely.

Cheers

item.176381

MacInTouch Reader

The reason you can run one item to failure and count the hours, or run many items until one failure and count the hours, is from the assumption of constant failure rate. Many electronic items have a constant failure rate after some initial burn-in period.
It's important to note that MTBF or MTTF is just a number, that may not be the number you're interested in. They are means [i.e. averages], thus 50% chance of failure.
The time to 1% or .1% chance of failure (insert your risk tollerance for losing your precious data here) is going to be a much smaller number.

item.176383

MacInTouch Reader

I'm the person who wrote the earlier post giving Linux command line instructions for testing HDs.

If you want to run smartctl on Mac OS X, you need to install Fink or MacPorts, then install smartmontools. Go here:

for further information.

HDs in external enclosures present additional challenges for smartmontools because Apple's drivers do not pass the necessary commands/responses over the interface. This project:

promises to replace Apple's drivers with ones that do. Building and installing it is a complex task and leaves your Mac very hacked.

I have not tried any of this, so YMMV. Make no mistake, I'm an Apple partisan, and use a Mac for almost everything I do. Sometimes, though, Linux is the right tool for the job.

item.176387

Henry Norr

Thanks to A. Kaleberg for the link to Google's analysis of hard drive reliability. The authors note that 'Failure rates are known to be highly correlated with drive models, manufacturers and vintages,' but don't off any specifics 'due to the proprietary nature of these data.' They do footnote a published study on this subject, but it's a decade old (published January 2004). Does anyone know of any non-proprietary and relatively up-to-date large-scale analysis of drive reliability by manufacturer and model?

[Though quite limited, there is some reliability data posted at storagereview.com. -Ric Ford]

item.176390

MacInTouch Reader

Re:

The third remaining player, Toshiba, makes mostly 2.5' hard drives. They recently bought a factory (I believe from WD) and started making a line of budget-oriented 3.5' drives, but they're not something you're going to see on the shelf unless you really go out of your way to buy one.

As a condition of their purchase of HGST, WD sold the 3.5' portion of HGST's business to Toshiba.

The 3.5' Toshiba drives are essentially the next generation of what were to be HGST drives, and are readily available at outlets like Newegg and Frys.

item.176391

Davide Guarisco

Re: testing hard drives with a Linux box.

Anonymous 'MacInTouch Reader' gave some good suggestions.

But there's no need for a Linux box. OS X is perfectly capable of doing the same!

The dd command is available, and I have compiled the current version of smartmontools, the suite to which the smartctl command belongs, on my Mac every time I bought a new hard drive.

It does require one to be confident with Terminal, but no particular other skills.

item.176406

Greg Chesney

And the HGST drives are worth it, in my experience. Never had one fail, and they haven't cheated us down to a two-year warranty like Seagate and others. Their SAS are 5 years; SATA 3 years.

Other than these, I will only buy WD Black, since prices are back down to tolerable levels.

item.176407

Greg Chesney

As a long time user of SoftRAID, they have some wonderful reliability tools that I use to stress test every drive I buy. If there is a flaw, their tools are likely to find it. If a drive doesn't pass, I immediately RMA it.

If you don't use SoftRAID, then I agree with others, the Unix dd tool from the Terminal can do wonderful things for testing a drive.

item.176421

David Kilbridge

For those not so confident with Terminal, there's a GUI application SMART Utility, which intelligently interprets the output of smartmontools so you don't have to.

(Thanks, Colleen, for your gentle but persistent evangelism, which finally got me to join the choir as a satisfied user.)

item.176416

Colleen Thompson

Re:

HDs in external enclosures present additional challenges for smartmontools because Apple's drivers do not pass the necessary commands/responses over the interface. This project: https://github.com/RJVB/OS-X-SAT-SMART-Driver
promises to replace Apple's drivers with ones that do. Building and installing it is a complex task and leaves your Mac very hacked.

Hmm, SAT SMART Driver is an optional install with SMART Utility. There is a standalone installer. It has enabled me to use SMART Utility on about one third of the external enclosures I've tried it on.

If this leaves my Mac 'very hacked', I would sure like to know about it.

item.176431

MacInTouch Reader

Colleen asks,

'If this leaves my Mac 'very hacked', I would sure like to know about it.'

I need to edit myself better and avoid loaded language. What I really meant to say was that OS X SAT SMART Driver adds kexts to your system that may or may not work when you apply a system update or a version upgrade. This is because the kexts are based on Darwin sources and depend upon internal Apple interfaces that can and will change as Apple updates its code, possibly rendering your machine inoperable.

So, OS X SAT SMART Driver is very useful in the lab, but it may not be the best option for deployment on user desktops.

Hopefully Apple will update its I/O code to properly handle SMART for HDs in external enclosures. Everyone would benefit.

item.176435

Harold Zeh

Concerning Smart Utility, it is practically useful for Macs that are off AppleCare or you want to know the status just for the sake of knowing. Apple [may ignore] the results, should you believe your drive is going south based on a parameter that has failed in Smart Utility's results.

Before my 2010 iMac was replaced, I checked the drive status with Smart Utility - Disk Utility said it passed, of course. The more detailed S.U. reading of the drive parameters indicated the drive was failing. I brought it to the Genius Bar at the Apple Store, as it was still covered for several more months, thinking they will confirm this obvious reading and replace the drive.

Wrong. They checked it on the counter and said it was totally fine. I persisted, and showed them it failed according to Smart Utility, so they said they would keep it overnight and beat up on it much more severely in the back. Next day, they pronounced it completely healthy and also told me in order to replace the drive it has to fail their tests and Smart Utility is not on their list of relevant analyzers.

Obviously, Smart Utility does not make up the results and is just the messenger of the same drive parameters Apple should be checking with whatever utility they use. Seems not to be so.

I left thinking that either buying the extended AppleCare was a waste of money or Smart Utility helpfully pronouncing the drive as failing was a waste of information. I went with the former and replaced the drive myself. My rationale was that the new owner would be better served with a new drive in the iMac - no matter what (and I would feel better about selling it).

Apple has no real motivation to replace a drive they can claim is dandy based on their tests that specifically ignore whatever details Smart Utility reports.

item.176440

Nathaniel Paust

For monitoring the SMART information on my hard drives, I really like SMARTReporter. Nice little app that runs in the background and shoots me an email if something goes wrong.

Also, regarding SMART and external drives, I recently upgraded one of my Seagate GoFlex desktop drives with the Thunderbolt adapter (it came from Costco with USB3), and suddenly I'm getting SMART data from it. That seems like a huge advantage of Thunderbolt over USB3!

item.176444

David Charlap

An anonymous MacInTouch Reader described zeroing out a new hard drive using the Linux command

dd if=/dev/zero of=/dev/sdx bs=4096

This can definitely work, but I'm pretty certain that a Mac user can do the same thing with Disk Utility. Select the drive itself (not any volumes), go to the 'Erase' tab and choose one of the security options (1-pass, 7-pass, etc.) to tell it to overwrite the entire drive and not just recreate the control structures.

That should do the same thing as the 'dd' command and should be much friendlier for someone unfamiliar with low-level UNIX/Linux commands.

item.176454

MacInTouch Reader

Responding to:

This project:
https://github.com/RJVB/OS-X-SAT-SMART-Driver
promises to replace Apple's drivers with ones that do. Building and installing it is a complex task and leaves your Mac very hacked.

That link was for a temporary fork of the actual project page:

The changes in the fork have been merged back into the main repository, so it no longer serves a purpose. It's also out of date (the latest version is 0.6, not 0.5), but, unfortunately, there are still sites and docs linking to it and misidentifying the developer. Jarkko Sonninen (kasbert) is the actual developer of the driver.

There is no need to compile it from scratch, as a pre-compiled installer can be downloaded here:

Lastly, this is a very stable driver that I've personally tested on my machines, and is the farthest thing from a hack. This is a solution, for something Apple failed to provide. More feedback can be found here.

item.176449

Stephen Hart

Greg Chesney wrote:

'And the HGST drives are worth it, in my experience. Never had one fail, and they haven't cheated us down to a two-year warranty like Seagate and others. Their SAS are 5 years; SATA 3 years.'

Judging a drive's likely lifetime by its warranty might be sensible, unless the company sees the warranty length as just more advertising.

But replacing a failed drive with a refurb of the same size and manufacture date (maybe even the same batch) after even one year? I wouldn't. This is even a problem with in-warranty Macs. Apple could well have a stock of same-batch refurbs for just such replacements.

After a year or more, I'd opt to replace the drive with a larger, newer-generation drive.

item.176457

Stephen Hart

Harold Zeh wrote:

'[Apple Geniuses] checked it on the counter and said it was totally fine. I persisted, and showed them it failed according to Smart Utility, so they said they would keep it overnight and beat up on it much more severely in the back. Next day, they pronounced it completely healthy and also told me in order to replace the drive it has to fail their tests and Smart Utility is not on their list of relevant analyzers.'

I had the a better experience with my 2010 iMac. The genius wrote my report of Smart Utility (on his iPad); they kept the iMac overnight, and replaced the drive.
I assume they ran Hardware Test overnight.

item.176460

David Krafchick

Harold,

Your experience leaves me with a couple of questions. First and most important is how old is the hard drive? If it is less than 2 years, I might question whether the report is accurate. My other thought is that what Apple did behind closed doors could have been more extensive than what Smart Utility offered.

I have drives that have failed at less than 3 years. I also have LaCie Porsche drives that have lasted 10 years and counting.

If Smart Utility shows that the drive is failing, that might be true. It also could be reporting something that is possible, but not definitive.

In long days past, I had a utility that checked the drive and marked bad blocks. Those bad blocks were locked off and never written to. This was a way to extend the life of your drive. As far as I can tell, this is no longer available.

I would still like to have something that would alert me that something is failing when it shows signs of failing and allow me to save what I need. I have not found anything that can do that where I can afford it.

item.176463

Barry Levine

I can't comment on the Smart Utility app but I have had drives pass Apple's SMART in Disk Utility yet still behave badly in operation. I will boot from a known good drive with TechTool Pro installed (an external or via Target Disk mode). Often, TTP will find an issue (usually bad blocks) that Apple's Disk Utility will not find. When I call AppleCare and tell them the results of my tests (and sometimes eMail them the TTP results screen), I've never had them refuse to replace the drive. I do not ever take a Mac to the Apple Store; instead, I ask for a visit by their Field Service tech. So we've already diagnosed the issue and the visiting tech simply swaps out the drive.

I realize this won't cover all issues but I find that performing as many diagnostics as I can before calling AppleCare actually helps them get the cases resolved quickly.

item.176459

MacInTouch Reader

The elephant in the room that hasn't been discussed is damage due to shipping and handling. HDs are delicate electronic and mechanical devices that don't tolerate excessive G forces. I wonder how many of the HD failures people are experiencing are due to careless handling?

Perhaps if HD manufacturers added shock stickers to their packaging it might be possible to sort out the HDs that have been badly handled.

item.176495

Harald Striepe

Re. 32MB, huge for a first drive... Try 5 MB, same price range, and an awesome experience with Concurrent CP/M.

I guess I date myself...

item.176471

MacInTouch Reader

Smart Utility parses information from the hard drive and tries to determine if it is good, failing or bad. It usually does a great job of determining the pass/fail status of a drive.
Apple tries to do the same thing with Disk Utility but they set the bar very, very high. So Disk Utility will not report SMART failures until it is too late. The diags that Apple uses in-house are a little better, but they will frequently report that a failing drive is good. It will take a lot of complaints from customers for Apple to change their diags so that they are actually reporting correctly.

item.176473

Samuel Herschbein

Re: Harold Zeh and SMART Utility:

SMART Utility's algorithm for determining when a drive with errors is 'failing' isn't foolproof. There's room for interpretation; you need to examine the actual errors and how many of each occurred. Here are some of my experiences:

A replacement Hitachi Enterprise drive came with a few reallocated Bad Blocks. SMART Utility says it is failing. I use the drive for rotating backups - in a couple of years, it hasn't had any more errors.

SMART Utility says another drive is failing; it has several DMA errors. These errors occurred years ago when the drive was in a Mac that had a bad motherboard. The drive has been working since without error.

In my experience, when SMART Utility reports recent, multiple bad blocks, that's a good indication of impending drive failure.

item.176484

MacInTouch Reader

Samuel Herschbein is correct on the interpretation of S.U. results. A common issue that we're seeing is failing hard drive cables in MacBook Pro 13' machines manufactured from 2009 to 2013.
When some of these cables start failing, the HD will log DMA errors. As long as you don't get any bad/reallocated/pending sectors, then the drive is probably OK and it was just a bad cable. S.U. will always need to be interpreted for that drive because of this issue.
For those in a corporate environment, it would be a huge win if they install a LaunchDaemon that reports SMART status back to one of their servers and can replace it before major data loss occurs. SMART Reporter is the only publicly available utility that I have seen that does this.

[I can confirm the failing hard drive cables on 13-inch MacBook Pros. It's really a nightmare, if you don't know the cable could be a problem and you're trying to find a problem elsewhere. -Ric Ford]

item.176487

Michael Fussell

I really do not see the downside of a SMART Utility type program falsely predicting the failure of a hard disk. The downside, if any is $100 to $150. Typically, these false failure predictions occur after some years of hard disk prediction in which much of the value of the hard disk has been extracted. I would rather have false predictions than failure to predict. The value of data is so much more than the value of the hard disk. And hard disks are basically cheap. (My first hard disk, 32 megabytes, cost $1200.)

Format Seagate Drive For Mac

item.176488

Stephen Hart

Samuel Herschbein wrote:

'In my experience, when SMART Utility reports recent, multiple bad blocks, that's a good indication of impending drive failure.'

Another obviously bad sign would be if Smart Utility reports increasing numbers of reallocated blocks.

item.176490

Colleen Thompson

David Krafchick said several things...

[Harold Zeh's] experience leaves me with a couple of questions. First and most important is how old is the hard drive? If it is less than 2 years, I might question whether the report is accurate.

In years and years of using SMART Utility, I can only remember two cases of it not catching a failure and one or two false positives. To judge the amount of my experience, in my garage are three plastic milk crates filled with failed hard drives. This, in addition to the many that were taken apart for their magnets.

My other thought is that what Apple did behind closed doors could have been more extensive than what Smart Utility offered.

Apple's techs are constrained in what they're allowed to use. My local Apple-certified tech (not at an Apple Store) must do a surface scan. Usually he is happy to run a full scan on a drive looking for bad blocks on the basis of SMART Utility's output.

If I'm going to try to get Apple to replace a 'failing' drive, I call armed with results from multiple utilities (Smart Utility, TechTool Pro surface scan, etc.) and can usually convince them. The one time I couldn't - a new refurb iMac that had 1 bad block after a week's use - the local tech ran a stress loop on the iMac for a couple of days, whereupon more blocks went bad, and the drive was replaced.

I've also had luck with manufacturer warranty RMAs by providing SMART Utility's confirmation of bad blocks, despite some online 'requirements' that you run their Windows-only utility.

If Smart Utility shows that the drive is failing, that might be true. It also could be reporting something that is possible, but not definitive

This comes down to the definition of 'failing', which I will address at the end.

In long days past, I had a utility that checked the drive and marked bad blocks. Those bad blocks were locked off and never written to. This was a way to extend the life of your drive. As far as I can tell, this is no longer available.

Disk Utility's Erase function that will 'write all zeroes' does this very thing. It writes zeros all over the disk, and when it runs across a bad block, it will lock it away. Here is the problem of using this to 'extend the life of your drive': a bad blocks failure (the most common failure I see, by far) is almost always a progressive type of failure (confirmed by every other tech I've talked to). My understanding is that it is the actual magnetic coating flaking off the platter. Think of your exterior house paint. Once it gets old and starts flaking, it just keeps getting worse. In most cases, trying to keep using a drive that is undergoing this type of failure is a false economy, especially with the low cost of today's drives.

I would still like to have something that would alert me that something is failing when it shows signs of failing and allow me to save what I need. I have not found anything that can do that where I can afford it.

If you have a current backup this would not be a concern. Some types of failure are catastrophic and instantaneous. Surely you're not working without a net?

I know I constantly sound like an advertisement for SMART Utility, but here is why I prefer it. The most common type of failure I see is bad blocks; the official SMART threshold for 'failure' is such a high number of blocks that Disk Utility almost never will admit a drive is failing, ditto many other utilities that use the official threshold. You can do a surface scan with TechTool Pro and others to find bad blocks, which can take hours and hours (and is often useful to confirm what SMART Utility is telling you.) SU will tell you the exact number of blocks that are bad (so will smartmontools and others). SU has its own algorithm for judging failures and will red-flag a drive with even just one bad block. Since it's a progressive condition, this is very useful information.

You can get a SU license for a modest fee, and it lives in your dock or menubar and will turn yellow or red to indicate impending or actual failure.

Other apps can confirm bad blocks or other failure. TechTool Pro has the surface scan and other SMART info. Disk Warrior will complain of 'disk malfunction'; Carbon Copy Cloner's log will contain references to 'I/O error' (and additionally will tell you what files it couldn't copy due to damage); Console also logs I/O errors indicating a disk problem. I believe SMARTReporter now contains some more detailed SMART information than before. The Finder will exhibit random and frequent 15-second SPODs* as it tries to access the inaccessible sectors.

The main point is that the official SMART threshold for a bad blocks failure is much, much too high.

Usual disclaimer, I have no relationship with Volitans Software except as a very satisfied customer.

*SPOD = Spinning Pizza of Death

item.176491

Colleen Thompson

Re:

For those in a corporate environment, it would be a huge win if they install a LaunchDaemon that reports SMART status back to one of their servers and can replace it before major data loss occurs. SMART Reporter is the only publicly available utility that I have seen that does this.

FWIW, SMART Utility recently added email notifications. I finally got it working, using my ISP's SMTP server. I would love to get it to work with my iCloud account, since this is a laptop that's not always at my house.

item.176498

David Charlap

SMART defines a broad set of parameters but does not require drive makers to support any particular subset. For each parameter, the drive reports the current value and a manufacturer-specified threshold value. The manufacturer assumes that actual values should never cross the thresholds and that, if they are crossed, the drive should be considered 'failed'.

As I understand it, Apple's Disk Utility reads the values and reports 'failed' if any threshold has been crossed. It reports 'pass' if none have been crossed. It makes no attempt to interpret what the actual values mean, nor does it track trends (e.g. an increasing count of reallocated sectors that hasn't yet crossed the manufacturer's threshold.)

SMART Utility takes these steps that Apple does not. It tries to interpret the raw data, reporting potential problems before the manufacturer threshold has been crossed and looks for trends that may indicate problems, despite the value remaining in safe ranges. This is a double-edged sword. It can spot impending failures sooner than Apple's method, but it also means that it will often declare a drive failed when it is still operating within manufacturer specifications (and is therefore ineligible for warranty replacement.)

item.176518

Martin Pilhall

My tip for checking the health of your hard drives? Drive Genius 3's Drive Pulse. This checks your drives regularly and also monitors the SMART status. This has discovered several failing disks in an early stage for me, a very satisfied customer :-)

item.176573

Format Seagate For Mac And Pc

Victor Drummond

For you MacInTouch gurus out there --

I have a HDD that apparently has a mind of its own. It almost continually sounds like it is accessing data, even though Activity Monitor doesn't seem to be indicating any actual I/O. There are periods where it stops this activity for 5-10 seconds, then resumes. There is no problem accessing files on the HDD.

Hardware: I have a mid-2011 27' iMac, running OS X 10.8.4, in which I installed a 256GB SSD and replaced the 1-TB HDD with a Hitachi Deskstar 7K3000 3TB HDD.

How To Format Seagate For Mac

I had never encrypted drives but decided to use FileVault to encrypt the SSD boot disk. All went well, so I thought I'd encrypt the four partitions on the HDD. I used the Finder pop-up menu to 'Encrypt..' the partitions. Encryption ran in the background with no user interface. In my case, each partition took between 4-8 hours to encrypt.

The resulting 'structure' shown in Disk Utility for the HDD looked very different with 'volume families' and 'volume groups.'

After several days I decided I wanted to non-destructively remove the encryption from the HDD partitions, so I used the Finder pop-up 'Decrypt...' to do that for each partition. Again this took hours without a user interface to show the progress. The 'structure' of the HDD in Disk Utility continued to look strange to me. I decided since I had a full backup of each partition I would just erase and repartition the HDD to remove all the encryption structure.

The problem was that Disk Utility disabled the 'Partition' window entries for the HDD. I couldn't use it to repartition the drive. Internet searching led me to Terminal commands using 'diskutil' to 'erase' the logical volume groups and families created by 'core storage.'

After mucking around in Terminal, I got the HDD in a state where I *could* use Disk Utility to repartition and restore the HDD to a non-encrypted state. That all seemed to go well.

But somehow I have apparently left the HDD in a state with this apparent continually accessing of the HDD.

I repartitioned the HDD a second time with no change to the continual HDD accessing.

I thought perhaps it was a glitch in OS X 10.8.4, so I started up using a developer version of Mavericks, with the same apparent accessing of the HDD.

A very long story, but does anyone have any suggestions to stop this apparent accessing? Why doesn't a reformat restore the HDD to a 'normal' state? I can still access data on the HDD, but the sound of this apparent accessing is annoying. Is there any way to 'fix' this problem?

My very last option would be to have to open the iMac again and replace the HDD.

Any suggestions would be greatly appreciated.

item.177685

Eric Z

I'm interested in the type of media people are using for booting/imaging their newer iMacs and MacBook Pros/Airs.
I have some USB3 bus-powered drives that work great, but when using a flash drive, it is painfully slow (hours to boot and install OS), almost as if the drive is being indexed.

I had one person suggest it is the read/write speed of the flash drive (but haven't found any vendors that share what their specs are.
Thanks

[Flash drives (USB sticks) are typically far, far slower than hard drives, even the faster USB3 variants. For speed, an SSD drive would be ideal - far *faster* than a hard drive. Just stick one into an external USB3 case, and it'll fly on a Mac with USB3 ports. For broader compatibility, I'd actually want FireWire 800 support, as wel, and OWC has a product line with that and SSD options, their On-The-Go Pro series. -Ric Ford]

item.179978

MacInTouch Reader

We've always been able to replace Apple's hard drives with drives of other manufacturers without problems and have replaced some with various brands of SSDs as well. However in a recent email exchange with a drive supplier, I was asking about whether I could still put 2 drives in the latest model of iMac since Apple only seems to sell 1 drive per iMac now. In the course of the reply, they mentioned that it was not presently possible because of new hardware limitations that Apple had put on the hard drive bay in the iMacs. They indicated that drives must now have Apple firmware on them or else the fans will go to high speed and stay at high speed.

Can anyone confirm this is true? If so it's hard to believe that Apple has done something like this which limits the end users even further in being able to use the hardware as we see fit, or to upgrade to minimize the shortcomings of the configurations they offer. I'd also love to know for sure if there is still room for a 2nd HD/SSD like my Mid-2011 iMac has (1 HD, 1 SSD).

item.179985

Greg Weston

A MacInTouch Reader says:

'I was asking about whether I could still put 2 drives in the latest model of iMac since Apple only seems to sell 1 drive per iMac now. In the course of the reply, they mentioned that it was not presently possible because of new hardware limitations that Apple had put on the hard drive bay in the iMacs. They indicated that drives must now have Apple firmware on them or else the fans will go to high speed and stay at high speed.
Can anyone confirm this is true?'

There's some coverage of it on the Other World Computing blog.

item.179987

MacInTouch Reader

Another MacInTouch reader writes,

'In the course of the reply, they mentioned that it was not presently possible because of new hardware limitations that Apple had put on the hard drive bay in the iMacs. They indicated that drives must now have Apple firmware on them or else the fans will go to high speed and stay at high speed.'

Not true.

Apple's drives have an additional temperature sensor (not custom firmware), and without that, the fans run at high speed.

There are third party cables with the sensor, and by using such a cable, everything works as expected, with any aftermarket drive.

item.179993

Chuck K

Re:

'Can anyone confirm this is true? If so it's hard to believe that Apple has done something like this which limits the end users even further in being able to use the hardware as we see fit, or to upgrade to minimize the shortcomings of the configurations they offer. I'd also love to know for sure if there is still room for a 2nd HD/SSD like my Mid-2011 iMac has (1 HD, 1 SSD).'

Essentially, yes. Instead of using an external temperature sensor, or polling SMART data for temperature every x seconds, Apple has had the drive manufacturers repurpose the drive activity light pin with firmware to pulse based on the drive temperature.

For at least the 2011 iMacs, there is an external sensor kit available from OWC that you can stick to a drive and gives the correct signals to the motherboard. I haven't done any research on anything newer than the 2011.

item.179995

MacInTouch Reader

Re:

'Clearly Mavericks caused some heretofore unknown issue [with SSDs]...'

I never realized that drives (either HDDs or SSDs) were so closely tied to a particular OS version. I assumed they operated on a lower level than that. After all, it's not that hard to use the same drive with both a Mac and Windows... yet you have to worry about which versions of Mac OS it will work with?

My specific concern is this: I'm not going to replace Snow Leopard with Mavericks, but I might stick another drive (HDD or SSD) in my Mac Pro, install Mavericks on it, and then switch-boot between them. If I do that, will I have to worry that the drive will work with one of the OSX flavors but not the other? (For example, say I have an SSD meant for booting Mavericks... I might still want to access some files that are on it while booted from the other drive into Snow Leopard, or vice versa.)

Do the firmware-tied-to-OS-version issues still matter if you run one of the OS flavors via VMware Fusion rather than switch-booting?

item.180029

Colleen Thompson

Re:

We've always been able to replace Apple's hard drives with drives of other manufacturers without problems and have replaced some with various brands of SSDs as well. However in a recent email exchange with a drive supplier, I was asking about whether I could still put 2 drives in the latest model of iMac since Apple only seems to sell 1 drive per iMac now. In the course of the reply, they mentioned that it was not presently possible because of new hardware limitations that Apple had put on the hard drive bay in the iMacs. They indicated that drives must now have Apple firmware on them or else the fans will go to high speed and stay at high speed.

Yep, just happened to me with a client's mid-2011 iMac. Prior to this, when a stock hard drive failed, I would have the local Apple-authorized service shop do the labor, but I would always supply my own drive (preferring Western Digital Caviar Black -- 5-year warranty and one-third the price). On some of the pre-2011 models, my tech, also a friend of mine, had to perform some kind of magic with the thermal sensor in order to make the new hard drive play nice with the iMac.

Not so in this case. No matter what the tech did, and he tried several things, the fans would run full blast with the non-Apple drive installed. So we ended up springing over $300 for an Apple Drive, where the Western Digital I supplied would have been under $100.

I have to say I am not fond of this kind of proprietary behavior.

This, of course, does not directly address your question about putting two drives in a modern iMac, but everybody's hard drive is going to fail at some point, and they should be aware of the limited choices for replacement.

item.180014

Ted Vaida

Re:

'I never realized that drives (either HDDs or SSDs) were so closely tied to a particular OS version.'

Without going too deeply into the mechanics of the SATA and SCSI specifications, it is difficult to explain this; however, a first-order approximation would be that the various versions of operating systems change the algorithms used to select the order and frequency of disk I/O (in blocks), and which of the multitudes of optional commands from the standards to use. As previously unused or underused features of the standards appear in new OS releases, operations or sequences of operations that were never seen by the SSD vendor during testing can occur. These are 'corner-cases', as they are known in engineering speak, and they are the curse of any hardware vendor.

With regard to compatibility, multi-boot and even virtual machine use, this is generally a 'forward' issue, where the vendors are reasonably guaranteed to ensure continued operation with older releases. Since the drive vendor cannot count on the software vendors to patch their 'obsolete' releases to maintain compatibility with the drives (cough... cough... 10.6), and drive lifetimes are often pushed to extremes (there are drives over 20 years old in use today...), retaining backwards compatibility is the expected norm, and I would not lose any sleep over it. You can confirm this by looking at currently shipping drive models and note that most SATA-III disks with 2Tbyte or more capacities will usually respond correctly and without errors even to an i386 motherboard using PIO mode instead of UDMA - that capability takes up space in the firmware, time for the engineers to develop and test, and adds little incremental value to the disk, but is considered a 'must have', generally.

item.180047

Brian S

Colleen wrote:

'Yep, just happened to me with a client's mid-2011 iMac... So we ended up springing over $300 for an Apple Drive, where the Western Digital I supplied would have been under $100.'

FWIW: OWC sells hard drive cables specifically to overcome fan speed issues associated with using a non-Apple drive:

IIRC (and see their installation videos for details), their thermal sensor connects to the motherboard (with the sensor attached to the top of the hard drive).

item.180049

Colleen Thompson

Re:

FWIW: OWC sells hard drive cables specifically to overcome fan speed issues associated with using a non-Apple drive:

Thanks for the tip. It's too late for this project, but you can bet I will remember that going forward.

item.180067

MacInTouch Reader

Does anyone have any recommendations on software so that my Mac can write to NTFS-formatted disks (by default, Macs can only read NTFS volumes)? I tried the one software package, but it caused a Finder memory leak so I had to uninstall it.

item.180074

Kevin Lepard

In response to MacInTouch Reader, I've used Paragon NTFS for years without any problems.

If you watch, you can sometimes find it discounted or in a bundle.

HTH.

item.180076

Ross Bogue

I've had good luck with Crystalidea's freeware Macs Fan Control handling the fan by polling the drive's smart data. There's no need for Apple's custom firmware, nor custom cables.

item.180077

MacInTouch Reader

Go to MacUpdate.com and do a search for 'NTFS' and you'll get a list of potential software candidates for reading and writing to NTFS devices.

item.180084

MacInTouch Reader

I tried Paragon NTFS but it caused a memory leak and also prevented my computer from going to sleep on schedule -- these have been periodic issues for some users. I read about other commercial/freeware solutions, but Paragon seems to be the best one. For now, I am using Parallels (WinXP) to mount my external NTFS-formatted drive to my Mac desktop -- I can read/write NTFS volumes this way without any additional software, but it's annoying to boot up Parallels/WinXP when I want to write to an NTFS-formatted drive.

item.180089

Samuel Herschbein

Re: MacInTouch Reader wanting to write to NTFS-formatted disks:

I used to have Paragon. I switched to Tuxera a couple of years ago when Paragon was not working properly for me.

Tuxera works OK. Writing can be very slow, under 10 KB/second to a USB flash drive that copies Mac files radically faster. I've also noticed that when writing/reading NTFS drives, the Finder in Mountain Lion has issues multi-tasking. On occasion the Finder will hang. My solution has been to start a copy to an NTFS drive and let it complete un-interrupted. Not very efficient, but it works...

item.180316

Colleen Thompson

Re:

I've had good luck with Crystalidea's freeware Macs Fan Control handling the fan by polling the drive's smart data. There's no need for Apple's custom firmware, nor custom cables.

That's an unsatisfactory solution in many cases, since it could be broken by system updates or reinstalls.

item.180144

MacInTouch Reader

I'm curious why some readers are jumping through so many hoops to get their Macs working with NTFS drives. I just format all my cross-platform flash drives with FAT32, and they work fine with no special software; and VMware Fusion/Windows 7 doesn't care that my main drives are HFS+. Is there any technical advantage for a mostly Mac user to mess with NTFS? Or is there just some external (organizational) reason why you have to deal with those drives?

item.180146

Mike Jackson

I have had Paragon NTFS drivers since 10.5 days up through my current 10.8.5 and have never had a problem with it allowing either my Mac Pro (2009 Nehalem) or my 2012 MacBook Pro or my 2008 Mac Book Pro go to sleep. I run Boot Camp partitions on all of those Macs and used Paragon to format the drive (and sometimes even repair the NTFS partition) for Windows 7 x64 Pro. So chalk up my experience as a 'good one' for Paragon Systems.

item.180179

Bill Ehrich

I use NTFS on USB drives to carry TV shows to a friend's BlueRay player, which can't read HFS or exFAT. Paragon works well for this except that I've had better luck when I first format the drive on a Windows PC.

item.180185

John Varsik

One reason for needing NTFS is that FAT32 does not support individual files larger than 4GB.

item.180181

Ted M

Thanks for the responses. To answer some of the replies: my organization is changing policy that any external hard drives used with work computers have to be formatted NTFS, and I use an external hard drive to transfer work files between home and work (unfortunately, cloud options like Dropbox are banned); I did review the options in macupdate.com and only the Paragon driver seemed to be a viable choice, due to speed and stability. I am suspecting that using the Paragon driver introduced a compatibility issue with existing third-party software on my computer, but since I can read/write NTFS volumes using a Parallels 8 virtual machine, I have not spent any time tracking down the incompatibility.