Re: Healthy (At Risk) Drives

[Neal Kemsley <neal.kemsley(at)gmail.com>]

Dear Phil,

 

The re-generate will not affect the existing data on the drives.

 

Yes the missing drive is highly likely to be the cause of the problem. Is there any predictability about which drive will need re-seating or do you just re-seat the whole lot regardless? If you can prove that it is a particular drive I would persue getting that drive replaced. (One of the motors in the drive may have a sticky spot and require a reseat to jog it free.)

 

In my experience with the MediaDaock LVD it needs a good long spin up time in order for all drives to have sufficient time to spin up. The drives are spun up based on the their slot position and the lower slots will take longer than the upper. Since the chassis makes so much racket it is sometimes hard to tell whether all the drives have actually spun up fully. Try to be patient with it after powering up (go away and make a cup of tea/coffee if you do not have clients breathing down your neck!) and wait a little more time than usual before firing up the CPU.

 

Another cause is that one of the drives is simply taking longer than its colleagues to spin up because it is beginning to fail. This is a really hard one to diagnose since this may be the only symptom right now. However this kind of problem has a really nasty habit of getting worse very quickly. ASM might show this drive as having errors although you are probably going to have to invest the time and aggrevation in breaking the stripe group concerned and doing a thorough destructive (Read/write) test on it in order to determine which drive is the culprit if it is not clear from other fault finding. It might be better to be safe than sorry here.

 

One last thing is that these chassis are prone to getting dust bunnies built up on their mid-plane over time. The fans pull quite a high volume of air over the drives and through the chassis. If this dust is getting on one of the SCSI connectors it might be causing an intermittant connection. Next time you have all the drives out have a good look in there with a lighting instrument of choice (Over here (I'm in Jover land!) we would say torch but a lot of US colleagues tell me that this conjures up the image of a wooden stick dipped in pitch and carrying an open flame!) have a good look for dusty connectors and blow out / vacuum any offenders.

 

Sorry for the rambing mail - I hope you are able to get to the bottom of this one!

 

With kind regards,

 

Neal

 

On 10/31/05, Phil Mastman <phil(at)sandlotpictures.com> wrote:

Hi Neal--

Thank you very much for your reply.  I'll give ASM and the re-generate
options a try.  Just to make sure, does the "regenerate" option endanger any
of the media that I have on the array?

There is one other factor that I forgot to mention... We have two sets of
drives that we regularly swap in the mediadock.  About 1 out of 3 times
, when I put in the set of drives that is generating the error, only seven
of the eight drives come up in the manage disks window.  After shutting
everything off and reseating the drives, all eight will show up.  Think this
is related?

--Phil




> Dear Phil,
>
> It means that for whatever reason Windows has found a fault with the
> stripe set. Often this is just a momentary thing (ie the drives
> were'nt quite spun up fully when you started up Windows) but it can
> mean a degraded SCSI system, drive or chassis. I agree with Scott in
> that you should right click and select "re-generate" and then keep an
> eye on the set for re-occurances for a while. As an extra precaution I
> would recomend installing Avid Storage Manager on your system and
> performing an overnight random read only storage test. This will give
> a report in the morning of any errors found and may restore some
> confidence in your drive set. If you are in a position to do a clean
> format on the drive set at any time you may wish to take the
> opertunity to perform a random read/write test on the drives. This
> will thoroughly excercise the set and weed out any weak ones if any.
> (Note that the utility will only run on Avid drives unfortunately.)
>
> Sylvain has kindly made ASM available form the support site. Make sure
> that you follow his directions to install the correct version of Java
> first per the site instructions.
>
> http://www3.softimage.com/DS2/right/download/cstools.htm
>
> With kind regards,
>
> Neal
>
> Neal Kemsley
>
> ACSR DS|NT|Unity
>
> On 10/28/05, Phil Mastman <phil(at)sandlotpictures.com > wrote:
>> Happy Friday, everyone--
>>
>> From the Windows "My Computer>Manage>Disk Management"  screen, my video
>> storage D: drive displays under each of the 8 disk numbers in the left
>> column "Online (errors)"  On the right side, where each of the drives of the
>> striped array is listed, it says "Healthy (at risk)".
>>
>> What does this mean, exactly?  Is there something I should do to remove the
>> errors and the risk?  The drive array has been working perfectly... No
>> issues whatsoever.
>>
>> Compaq, Equinox, 7.6 QFE 2
>>
>>
>> Thanks,
>>
>> --Phil Mastman
>> Sandlot Pictures, Inc.
>>
>> ---
>> Subscribe? E-mail Majordomo(at)Softimage.COM with the following text in body:
>> subscribe ds
>> Unsubscribe? E-Mail Majordomo(at)Softimage.COM with the following text in body:
>> unsubscribe ds
>>
>
>
> --
> _________________
>
> ---
> Subscribe? E-mail Majordomo(at)Softimage.COM with the following text in body:
> subscribe ds
> Unsubscribe? E-Mail Majordomo(at)Softimage.COM with the following text in body:
> unsubscribe ds

---
Subscribe? E-mail Majordomo(at)Softimage.COM with the following text in body: subscribe ds
Unsubscribe? E-Mail Majordomo(at)Softimage.COM with the following text in body: unsubscribe ds