RE: Adding a disk to a RAID array

"Stewart M. Ives" <[email protected]>
Newsgroups gmane.linux.redhat.release.enigma
Message-ID <[email protected]>
Liz or Elizabeth, what ever is your fancy:

No problem.  You can get a fast IDE controller (around $100 or less) with
multiple channels, put the CHEAP BIG IDE drives on the controller card,
software or hardware RAID, and do the moving and renaming.  Use them for
data.  Keep the system on your hardware SCSI RAID.

THEN put another single CHEAP BIG IDE drive on the systems IDE controller,
format it, and use it as an on-line backup to suplement your tape backup.
Same system, you can even dismount it when you are not doing a dump or
backup to it.

Of course you are backing this stuff up aren't you??

Anyway, with the price and size of IDE drives today it's a wonder we still
use tape.  The point I was making is to take advantage of CHEAP BIG drives
to provide _on-line_ backup in ADDITION to tape for long term backup.

You can DUMP and compress many many days or weeks of backup to this "backup
drive."

stew


[email protected] wrote:
> Stewart,
>
> Unfortunately, this solution creates a whole different set of issues.
> Because of the sensitive nature, defined by other people than myself
> and higher up the food chain, the security issues that I have with
> the one box would then propagate to the new one thus doubling the
> headaches I already have with the first box.  Lets just say that in
> the near future I have to move the box from my server room to a
> special secure area in to the "Fort Knox" room two floors below me.
>
> Elizabeth Martin
>
> "Stewart M. Ives" wrote:
>
>> Elizabeth,
>>
>> I'm going to tell you the same thing I said before.
>>
>> Put 2 cheap BIG (100+GB) IDE drives in the box.  Let Linux do the
>> RAID for you, software RAID.  Once this is up, move all the data
>> from the old data partition over to the new dataNEW partition.
>> Rename the old partition to dataOLD and rename the new partition
>> from dataNEW to data.  There.  Problem solved.
>>
>> One last thing.  For backup, put a BIG IDE drive in another cheap
>> cheap cheap old old old system unit, load linux on it, put it on the
>> network, and do your backups first to this sytem and then to tape.
>> Then you will basically have on-line backup in the event of a data
>> drive failure or someone deleted that oh-so-important file (probably
>> their resume).
>>
>> Good luck.
>>
>> stew
>>
>> [email protected] wrote:
>>> Bill,
>>>
>>> Yeah, that is what I was hoping for.  I wish I was lucky enough to
>>> be able to have the funds to have another system that I could test
>>> the process out on.  However, being a state funded institution most
>>> funds go to medical research and education and I get what's left
>>> over and have to do the best I can with that.  Unfortunately the
>>> data partition is what I need raided the most.  Research data and
>>> faculty teaching information is contained there.  I have people who
>>> freak out big time when there is a network problem and can't access
>>> the box - they are banging at my door wanting to make sure that
>>> their data hasn't been lost.  Looks like I am going to have to take
>>> a whole new look at the process of what I need to do.  I may even
>>> plan upgrading to RH8 and upgrading the RAID controller card to one
>>> recommended by Adaptec to replace the one I have as well as
>>> replacing the drives with larger ones even though I have 4 matched
>>> drives already and hope I can get the funds.  Like I need all of
>>> this headache with all I am being required to do now in order to
>>> get all of my systems compliant with Federal and State regulations
>>> like HIPAA etc.  I am going to have to lock myself in my office to
>>> plan out the next couple of months just to keep current with the
>>> have to do's.
>>>
>>> Elizabeth Martin
>>>
>>> Bill Gradwohl wrote:
>>>
>>>> So you are trying to expand an existing array. Lots of older
>>>> controllers had no way to expand an existing array. Those that did
>>>> claim to do so failed under many conditions. Very recent
>>>> controllers are much improved, but that's not a description of DPT.
>>>>
>>>> You are more trusting than I am. Candidly, even if I had the manual
>>>> in hand I wouldn't try to expand a RAID array on a production box
>>>> unless I could test it out first on the EXACT same hardware in a
>>>> test environment.
>>>>
>>>> Have you considered, and does the controller BIOS support, just
>>>> adding a non RAID drive to the controller and using it all by
>>>> itself in a non RAID fashion? Even that is risky because you don't
>>>> know what the controllers BIOS will do with it. I had one
>>>> controller (can't remember which, but it was IDE) step on the
>>>> array at boot up. No chance to take evasive action.
>>>>
>>>> If you are running SCSI drives, does that also mean you have a
>>>> second controller for a tape drive? Not a second channel, but a
>>>> whole new controller that's already in the same box. How about
>>>> spinning that drive up on the separate controller.
>>>>
>>>> As you can see, I'm trying to do anything but touch the array
>>>> because what the books say will work many times don't.
>>>>
>>>> Bill Gradwohl
>>>> [email protected]
>>>> (817) 224-9400 x211
>>>> www.ycc.com
>>>>
>>>> _______________________________________________
>>>> enigma-list mailing list
>>>> [email protected]
>>>> https://listman.redhat.com/mailman/listinfo/enigma-list
>>
>> ---
>> Outgoing SofTEC USA mail is certified Virus Free.
>> Checked by AVG anti-virus system (http://www.grisoft.com).
>> Version: 6.0.461 / Virus Database: 260 - Release Date: 3/10/2003
>>
>> _______________________________________________
>> enigma-list mailing list
>> [email protected]
>> https://listman.redhat.com/mailman/listinfo/enigma-list


---
Outgoing SofTEC USA mail is certified Virus Free.
Checked by AVG anti-virus system (http://www.grisoft.com).
Version: 6.0.461 / Virus Database: 260 - Release Date: 3/10/2003
lmpx.com only provides a reader for public news (NNTP) servers. It is not affiliated with the servers or forums shown here and is not responsible for the content of articles, which is written by their respective authors.