Ok, yes im a dumarse in my haste to build my server i order 2 identicle wd20ears drives from the same place. Whats the likely hood of them actually dying at exactly the same time? Iv set one as my parity drive and the other for data i thought this way they are being used different amounts and may stagger their life spans?

 

Am i right in thinking I'm ok to order the same model drives from other websites because they will be a different batch?

Ok, yes im a dumarse in my haste to build my server i order 2 identicle wd20ears drives from the same place. Whats the likely hood of them actually dying at exactly the same time? Iv set one as my parity drive and the other for data i thought this way they are being used different amounts and may stagger their life spans?

 

Am i right in thinking I'm ok to order the same model drives from other websites because they will be a different batch?

Highly un-likely for them to fail at exactly the same time.  unRAID uses disks differently than most other RAID implementations.  They will not see the same amount of activity or hours spinning.

Chances are you will be OK.

 

If it's a bad batch, then it's a bad batch and will fail around the same time if they were used 24x7.

 

Where ordering multiple drives from at the same time becomes an issue is when you go to 3 or more drives or very large batches.

I buy drives in pairs pretty often.  In the past I've used them in raid1 scenarios (24x7usage) and they would fail within a couple months of one another.

 

When buying drives using newer technology, large batches may fail within a close range of one another.

This happened to us when we bought the Hitchi 1TB drives when they were just coming out.

We had a 4 drive raid 1+0 arrangement in a webhosting server. After one drive failed, another failed within a week.

 

Since the usage pattern of the 2 drives will vary, chances are you will be OK.

 

We had a 4 drive raid 1+0 arrangement in a webhosting server. After one drive failed, another failed within a week.

But they were also spinning identically.  The only "risk" is an electrical malfunction where a common part breaks down after exactly the same number of hours, but even then with spin-down, the risk is really-really small.

 

Joe L.

We had a 4 drive raid 1+0 arrangement in a webhosting server. After one drive failed, another failed within a week.

But they were also spinning identically.   The only "risk" is an electrical malfunction where a common part breaks down after exactly the same number of hours, but even then with spin-down, the risk is really-really small.

 

Joe L.

 

I made mention of the spinning identically and 24x7 usage.

Then again, sometimes it's something not related to age, such as firmware or sensitivity to heat. .

Another thing you might check is how close are the serial numbers?  That might also indicate if they might be from the same batch or not.

 

Like others have said, you should be fine in most cases. As long as when 1 does fail, you replace it in a timely manner.

Ok thanks for clearing that up guys put my mind at ease a bit.

Just preclear both drives (as you should do for all new drives).

Ok, yes im a dumarse in my haste to build my server i order 2 identicle wd20ears drives from the same place. Whats the likely hood of them actually dying at exactly the same time? Iv set one as my parity drive and the other for data i thought this way they are being used different amounts and may stagger their life spans?

 

Am i right in thinking I'm ok to order the same model drives from other websites because they will be a different batch?

 

There's no math behind this myth.  Weebo's experience with 2 drives failing close to the same time is more related to how early in the product cycle he bought, not to mention the company that made the drives.  They obviously sold those drives before they were stable.  The WD EARS seem to be pretty stable.

I have always purchased drives in multiples of 5 and never had any issues. I have had a few drives fail over the years but no pattern

1 WD Raptor fail 74gb

2 maxtor fail 500gb

2 Seagate    80gb & 500gb

2 Samsung    500gb

out of a total of 45 HDD's over the last 7 years

 

Ok, yes im a dumarse in my haste to build my server i order 2 identicle wd20ears drives from the same place. Whats the likely hood of them actually dying at exactly the same time? Iv set one as my parity drive and the other for data i thought this way they are being used different amounts and may stagger their life spans?

 

Am i right in thinking I'm ok to order the same model drives from other websites because they will be a different batch?

 

There's no math behind this myth.  Weebo's experience with 2 drives failing close to the same time is more related to how early in the product cycle he bought, not to mention the company that made the drives.  They obviously sold those drives before they were stable.  The WD EARS seem to be pretty stable.

 

Add to this that the drives were running 24x7.

 

The pattern of multiple drive failure is not just my own.

How many times have we read reports of people having a raid5 failure then having a failure when rebuilding the failed drive. It's not just a myth.

 

It's related to large bulk drives and usage patterns too. In raid5 all drives have to be spinning, so the usage patterns are similar.

 

In my experience, it's happened a couple times on RAID5 and RAID1+0 arrays where there were large bulk purchases.

When one drive failed, another would fail within days or weeks of one another. (I've not experienced hours).

 

Add to this that the drives were running 24x7.

 

The pattern of multiple drive failure is not just my own.

How many times have we read reports of people having a raid5 failure then having a failure when rebuilding the failed drive. It's not just a myth.

 

It's related to large bulk drives and usage patterns too. In raid5 all drives have to be spinning, so the usage patterns are similar.

 

In my experience, it's happened a couple times on RAID5 and RAID1+0 arrays where there were large bulk purchases.

When one drive failed, another would fail within days or weeks of one another. (I've not experienced hours).

 

Trying to derive some random report of a raid5 failure into don't buy multiple drives from the same vendor to avoid bad batches is such a stretch.  This non-science is a myth.

 

Add to this that the drives were running 24x7.

 

The pattern of multiple drive failure is not just my own.

How many times have we read reports of people having a raid5 failure then having a failure when rebuilding the failed drive. It's not just a myth.

 

It's related to large bulk drives and usage patterns too. In raid5 all drives have to be spinning, so the usage patterns are similar.

 

In my experience, it's happened a couple times on RAID5 and RAID1+0 arrays where there were large bulk purchases.

When one drive failed, another would fail within days or weeks of one another. (I've not experienced hours).

 

Trying to derive some random report of a raid5 failure into don't buy multiple drives from the same vendor to avoid bad batches is such a stretch.  This non-science is a myth.

 

Well, I am not sure what is being stated to mean "bad batch" when referring to hard drives.

 

I personally have never run across, or personally knew anyone that had a *bad* batch of HDD's per se. What I have experienced myself is HDD's with bad design spec's that need to be updated (rev .2 or whatever). I once had a situation where I had purchased a number of the dreaded IBM Deskstar's 80GB (after that I called the Deathstars, for how they blow-up). The design in that rev was pretty bad, low quality spindles and ball bearings if I recall right. The vendor had to redesign the spec's, and send out replacement (rev 2's) for customers with failed drives. In a situation like this, consecutive serial numbers would have made NO difference. They were all bad, until they fixed the design.

 

The way modern manufacture works, for drives or anything else is, they are either pretty much all good, or all bad.

 

When I order HDD's, for my own, or company use, I have never hesitated to order in bulk, and have never had any major issues. I have had individual drives arrive DOA, or fail after a short time of use, but never the "batch". That is why I appreciate the preclear.sh script SO much. It gives me a chance to weed out the potentially weak individual drives. I have had many good drives from pretty much all the different vendors, and some duds from all the different vendors.

 

For me, the lesson is "use the preclear script". It will save me heartache down the line.

 

Anyway, that is how I look at it, your mileage may vary.

 

Bruce

Well, I am not sure what is being stated to mean "bad batch" when referring to hard drives.

 

Generally a bad batch in this sense is when there is an entire manufacturing run with defective (or substandard) parts.  ie. Seagate's manufacturing facility receives a batch of IC's that were contaminated during one of the many lithography steps at the foundry and fail pre-maturely due to heat.  These IC's are used in the next 10,000 drives manufactured and Seagate now has a "bad batch" of drives that are prone to fail early in their life cycle.

 

It has nothing to do with the design of the drive, or the design of the IC that was used.  It fails based on a manufacturing error.  While not common, this kind of thing happens.  The idea is that there are quality checks along the way that should catch this, but they don't always work.

OK, let's say that were to happen. I would think that any manufacturer worth their salt would catch that. In the cut-throat world of drive manufacture, I would think something like this could ruin a company. I mean they do have quality control measures in place. I just cannot see 10,000 units making it out the door, into public consumption.

 

Has anyone here every had that happen to them?

 

Not rumors, or stories from a friend of a friend, I mean has anyone, personally had that happen to them? I know that I have not. I would be very interested to hear from someone who has had this happen to them.

It's funny, we give lots of advice about good practices to have a healthy array. Things like running monthly parity checks, backing up important data, keeping drives cooled, running preclears, and monitoring the smart reports. But NO recommendation has gotten such instant compliance as this one to not buy multiple drives at the same time. Here we have someone ready to jump off a bridge for committing this horrible breach of decency. (apologies to OP - this is just tongue in cheek)

 

In almost 3 years here I have only seen 1 verified case of two drives failng at near the same time, and those were not bought together. Compare that to the near monthly reports of someone detecting a bad disk through preclear.  The risk of buying a few disks from the same vendor at the same time are too low for me to worry about.

 

I do acknowledge that this experience might be different for RAID users, but for unRaid, it is the least of my considerations.

An interesting point, bjp999.  I believe it is because this practice is one of the easiest to follow.  At most, you end up wasting a few bucks on shipping from various vendors.  I'm coming around to your point of view that this probably isn't as big of a concern as I and others have made it out to be.  Now that I preclear all my drives before use (for a long time I didn't bother), I'm now more willing to buy drives in pairs.

 

Add to this that the drives were running 24x7.

 

The pattern of multiple drive failure is not just my own.

How many times have we read reports of people having a raid5 failure then having a failure when rebuilding the failed drive. It's not just a myth.

 

It's related to large bulk drives and usage patterns too. In raid5 all drives have to be spinning, so the usage patterns are similar.

 

In my experience, it's happened a couple times on RAID5 and RAID1+0 arrays where there were large bulk purchases.

When one drive failed, another would fail within days or weeks of one another. (I've not experienced hours).

 

Trying to derive some random report of a raid5 failure into don't buy multiple drives from the same vendor to avoid bad batches is such a stretch.  This non-science is a myth.

 

I can only speak from personal experience. I worked for two firms where when they bought drives in bulk, If there was a set of drives that were going to fail, they would do so in succession.

 

In one data center we would deploy multiple machines running 8 drives in raid 1+0 and see drives failing close enough to one another that made us change practice.

 

I've always recommended spreading the purchase of drives over time, places of purchase and vendors.

An interesting point, bjp999.  I believe it is because this practice is one of the easiest to follow.  At most, you end up wasting a few bucks on shipping from various vendors.  I'm coming around to your point of view that this probably isn't as big of a concern as I and others have made it out to be.  Now that I preclear all my drives before use (for a long time I didn't bother), I'm now more willing to buy drives in pairs.

 

I used to "always" buy drives in pairs. Many times today I still do. unRAID makes it easier in that I do not have to any more.

I buy as I need. preclear helps us insure we have a functioning drive.

 

I believe with unRAID's usage pattern, the multiple drive failure rate is pretty low.

 

However, I would not buy 20 drives of the same model, from one location and not expect to have 1 or 2 drives fail pretty close to one another.

 

I've seen it happen too many time in my years when doing system administration for large corporations.

OK, let's say that were to happen. I would think that any manufacturer worth their salt would catch that. In the cut-throat world of drive manufacture, I would think something like this could ruin a company. I mean they do have quality control measures in place. I just cannot see 10,000 units making it out the door, into public consumption.

 

Has anyone here every had that happen to them?

 

Not rumors, or stories from a friend of a friend, I mean has anyone, personally had that happen to them? I know that I have not. I would be very interested to hear from someone who has had this happen to them.

 

Seagate had a bad batch of 7200.11's or 7200.12's.  Don't remember the actual models, I would have to crack a case to find out.  Ended up going through a single manager that managed the incident from support and it took 5 drive shipments inside 2 weeks to get 2 drives that would not fail inside 5 hours of use.  This is testing the drives in diff facilities under diff operating conditions.  Neither system killed drives before or since.  The manager would not out right say the words "bad batch" but he danced very close to it.

 

As far as 2k drives, that was just a number pulled out of the air.  I have no idea how many facilities the drive manufacturers use, what their production runs are, or what their QC standards are; but sh*t happens and sometimes it doesn't get caught until it's too late.  Companies that deal with mass production understand the risks and what to do about it much better than we do.  That's one of the reasons there are warranties, dealing with the faults that do get through the system.  Yes it may hurt their rep for a short period of time if a company the size of Seagate/WD/Samsung sells a few thousand bad drives, but it would probably not affect their stock price by 1 full point, and that's what matters most to these companies; bottom line.

However, I would not buy 20 drives of the same model, from one location and not expect to have 1 or 2 drives fail pretty close to one another.

 

I also wouldn't want to buy 20 drives of the same model - even across different stores - at the same time.  I'd wouldn't want to buy that many all at once, even different models and different stores if I could help it.  But I usually buy 2, sometimes 3, at a time.  If the price were awesome I'd probably buy as many as 5 or 6.  I like to have diversity in my array, with drive purchases spread out over time and from a variety of drive models and manufacturers.  I worry less about where I buy them from - I think every disk in my array are from Newegg.

However, I would not buy 20 drives of the same model, from one location and not expect to have 1 or 2 drives fail pretty close to one another.

 

I also wouldn't want to buy 20 drives of the same model - even across different stores - at the same time.  I'd wouldn't want to buy that many all at once, even different models and different stores if I could help it.  But I usually buy 2, sometimes 3, at a time.  If the price were awesome I'd probably buy as many as 5 or 6.  I like to have diversity in my array, with drive purchases spread out over time and from a variety of drive models and manufacturers.  I worry less about where I buy them from - I think every disk in my array are from Newegg.

 

I don't worry about where I buy them from unless...

1. They do not pack multiple drives well (newegg is hit or miss here, but I still buy drives from them).

2. If I had to buy 10 drives for some project, I may diversify who I buy them from.

 

Now if I were to be buying drives from Sun with a Sun Support contract, then this changes things.

We insist on on site spares and an fast turn around for support.

 

OK, let's say that were to happen. I would think that any manufacturer worth their salt would catch that. In the cut-throat world of drive manufacture, I would think something like this could ruin a company. I mean they do have quality control measures in place. I just cannot see 10,000 units making it out the door, into public consumption.

 

Has anyone here every had that happen to them?

 

Not rumors, or stories from a friend of a friend, I mean has anyone, personally had that happen to them? I know that I have not. I would be very interested to hear from someone who has had this happen to them.

 

I have, Past patterns is when many drives are purchased and put in service in the same server.  25x7 RAID5 or RAID1+0

Also on other machines, When drives push past a boundary in size and we step into a new size.

Since these drives are on the bleeding edge, you may see multiple drive failures within a short period of time.

 

In my own personal purchase patters, I never purchase drives on the bleeding edge any more (and I used to).

I wait until a drive has been out for a while after any issues have stabilized.

 

In the past in my own personal usage, I would by pair of drives on the bleeding edge.

After about a year or so of spinning 24x7, I would start to see drive failures. One drive of the mirror would fail and within weeks, the other drive would fail.  I never lost data from this as I always choose to do advanced RMA.

 

Now I always have an onsite spare.

 

I'm not trying to spread FUD. This has been my personal experience.

unRAID has different usage patterns, so I've not experienced this type of multiple drive failure again.

I have no issues of buying 2 drives at the same time any more. In fact I usually do if the price is right.

But I still wait for a drive's maturity before doing it.

 

My array does have, 15 WD EACS 1TB drives, purchased over the course of a year or so.

After a couple of failures and replacements, the array has been stable ever since.

In my own personal purchase patters, I never purchase drives on the bleeding edge any more (and I used to).

I wait until a drive has been out for a while after any issues have stabilized.

 

In the past in my own personal usage, I would by pair of drives on the bleeding edge.

After about a year or so of spinning 24x7, I would start to see drive failures. One drive of the mirror would fail and within weeks, the other drive would fail.  I never lost data from this as I always choose to do advanced RMA.

 

 

 

This is the best policy.  Buy products that have matured.  Read reviews.  Take precautions by preclearing drives and updating firmware. 

 

 

Where I work, they bought 200+ 750gb Seagate 7200.9 drives for a handful of backup-to-disk arrays.  I can attest to what WeeboTech says, they do fail in batches.  We were having 5-10 drives failing within a few days of each other, then having several months of no failure, then another 5-10 drives failing at a time.  Over 4 years about 25% have failed.  Of those 25% that were replaced, 50% of the reman drives they sent as replacements for the warranty failed.

 

On the other hand, they bought 45 750gb Seagate ES drives, not one failure in 4 years.

Where I work, they bought 200+ 750gb Seagate 7200.9 drives for a handful of backup-to-disk arrays.  I can attest to what WeeboTech says, they do fail in batches.  We were having 5-10 drives failing within a few days of each other, then having several months of no failure, then another 5-10 drives failing at a time.  Over 4 years about 25% have failed.  Of those 25% that were replaced, 50% of the reman drives they sent as replacements for the warranty failed.

 

On the other hand, they bought 45 750gb Seagate ES drives, not one failure in 4 years.

 

That's an interesting experience.

It's close to what I've experienced. A batch of drives having multiple failures. Then good for a while, then batches of drives start failing again (Usually after that we swap out the model completely).

 

I've not dealt with the ES drives, I'm curious what the variances between them are.