A computer components & hardware forum. HardwareBanter

If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below.

Go Back   Home » HardwareBanter forum » General Hardware & Peripherals » Storage & Hardrives
Site Map Home Register Authors List Search Today's Posts Mark Forums Read Web Partners

Estimating RAID 1 MTBF?



 
 
Thread Tools Display Modes
  #61  
Old July 25th 04, 07:43 PM
Jesper Monsted
external usenet poster
 
Posts: n/a
Default

wrote in
news:1090041101.742428@smirk:

[snip]
This is where you get
yourself a big HP server, run IBM DB2 on it, and put the data on four
Hitachi Lightnings (two of them at the local site, each internally
RAIDed, mirrored across the two with a LVM or a SAN virtualization
box, and each then synchronously remote mirrored via rented dark fiber
to two offsite facilities).


Does it have to be rented? Some of us just plow it down ourselves

Ah, the wonders of being a telco.

By the way, all brand names in the
preceding sentence are meant as humorous illustrations; the fact that
they might be my former, current or future employers is one of these
funny coincidences. This system will cost you several ten M$, but it
is unlikely to go down or lose data.


Or you could get fairly decent hosts, running oracle RAC or other sorts of
clusters off mirrored HDS9570's or Clariions and have an uptime that's just
a fraction lower than running on big iron (or even higher, if you're lucky
enough to get a (as said by a Sun techie) "Monday model" E10k.)

If it happens to lose data (for
example because the field service and support team of the vendor who
put it together and manages for you f***ed up, which does happen in
real life), the CEO of the vendor will call your CIO, and offer to
kiss any bodypart the CIO wants to have kissed ... not to mention some
major financial apologies. In this environment, quietly marking a
sector bad would be tantamount to treason, and might even start a
lawsuit.


Hmm, we got an "Oops, sorry" when a midrange storage system by one of the
three-letter-guys experienced a double disk fault in a RAID 5.

[snip]
I've been
thinking of getting a cheapo IDE RAID card (I could probably swipe a
used 3Ware 4-port card from the office, we used a few of them in a
test setup and they are now gathering dust)

[snip]

Send me one


--
/Jesper Monsted
  #63  
Old July 26th 04, 09:05 AM
_firstname_@lr_dot_los-gatos_dot_ca.us
external usenet poster
 
Posts: n/a
Default

In article 3,
Jesper Monsted wrote:
....

If it happens to lose data (for
example because the field service and support team of the vendor who
put it together and manages for you f***ed up, which does happen in
real life), the CEO of the vendor will call your CIO, and offer to
kiss any bodypart the CIO wants to have kissed ... not to mention some
major financial apologies. In this environment, quietly marking a
sector bad would be tantamount to treason, and might even start a
lawsuit.


Hmm, we got an "Oops, sorry" when a midrange storage system by one of the
three-letter-guys experienced a double disk fault in a RAID 5.


You bought *** midrange *** hardware. You got the midrange treatment.
That's like a Travelodge (not quite as bad as Motel 6).

If you wanted the Ritz-Carlton treatment, you would have to buy
high-end hardware (Lightning, XP, Symmetrix, Shark, high-end Sun,
whatever it is called). Probably even pay for the complete end-to-end
treatment (get both implementation aid and services from the same
vendor).

I happen to know several stories where the CEOs of the big computer
companies had to talk to customers directly, after data loss
incidents. Typically happens if the fault of the data loss is clearly
with the vendor (human error that is clearly identifiable, for example
screwup by the local field apps engineer, or shipping faulty firmware
on the disk array). If the fault can be pointed at other vendors (for
example Seagate or Hitachi or Maxtor or Quantum disks), the customer
is out of luck. In the real world, finding the root cause of a data
loss is often quite difficult, as there are often multiple problems
cooperating (for example, one disk failure, one backplane or FC
failure, plus a bug in the firmware that temporarily lost the fault
tolerance, and then the engineer trying an untested workaround).

Life is not black and white; there are lots of shades of gray.

--
The address in the header is invalid for obvious reasons. Please
reconstruct the address from the information below (look for _).
Ralph Becker-Szendy _firstname_@lr _dot_ los-gatos _dot_ ca.us
  #65  
Old July 26th 04, 09:39 PM
Jesper Monsted
external usenet poster
 
Posts: n/a
Default

wrote in
news:1090829125.9518@smirk:

Hmm, we got an "Oops, sorry" when a midrange storage system by one of
the three-letter-guys experienced a double disk fault in a RAID 5.


You bought *** midrange *** hardware. You got the midrange treatment.
That's like a Travelodge (not quite as bad as Motel 6).


I expect the same level of data integrity of all of my hardware, midrange
or high-end, but i expect a whole lot more reliability and uptime from
the high-end stuff we buy. Loss of data is never an option, no matter
what i've paid for the stuff (with single-drive systems or other such
crap as the only possible exception). This was a known bug that has gone
unfixed for a while apparently - the box fails a drive, then fails it's
neighbor under some specific circumstances.

If you wanted the Ritz-Carlton treatment, you would have to buy
high-end hardware (Lightning, XP, Symmetrix, Shark, high-end Sun,
whatever it is called). Probably even pay for the complete end-to-end
treatment (get both implementation aid and services from the same
vendor).


We did - these systems were a part of the overall package. The drives
cost the same, but the system itself is somewhere between 1/3 and 1/4 the
price of the high-end system. I haven't succeded in convincing my boss to
only buy the high-end things, even if the overall cost of them isn't much
worse than the midrange things.

I happen to know several stories where the CEOs of the big computer
companies had to talk to customers directly, after data loss
incidents. Typically happens if the fault of the data loss is clearly
with the vendor (human error that is clearly identifiable, for example
screwup by the local field apps engineer, or shipping faulty firmware
on the disk array). If the fault can be pointed at other vendors (for
example Seagate or Hitachi or Maxtor or Quantum disks), the customer
is out of luck. In the real world, finding the root cause of a data
loss is often quite difficult, as there are often multiple problems
cooperating (for example, one disk failure, one backplane or FC
failure, plus a bug in the firmware that temporarily lost the fault
tolerance, and then the engineer trying an untested workaround).


We've had several cases run at priority "-1" or similar at different
vendors, all with daily followups between the CEOs and the heads of CS.
We've seen things that had been "pickled" for weeks or months that were
suddenly fixed in a day or two

This double drive fault was one hardware fault combined with a software
glitch that mistakenly reported it's neighbor broken. It was test data so
we didn't do much about it, but i would've liked an official apology and
assurance that it won't happen again.

Life is not black and white; there are lots of shades of gray.


I've seen a strange thing called "color" too

--
/Jesper Monsted
  #66  
Old July 26th 04, 09:43 PM
Jesper Monsted
external usenet poster
 
Posts: n/a
Default

Malcolm Weir wrote in
:

If the fault can be pointed at other vendors (for
example Seagate or Hitachi or Maxtor or Quantum disks), the customer
is out of luck.


If you happen to run into an old-time EMC person, ask them about NEC
disk drives....


Ask an HDS representative how it works now, and they should take the blame
for any data loss caused by hardware or software delivered by them.

--
/Jesper Monsted
 




Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

vB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
IDE RAID Ted Dawson Asus Motherboards 29 September 21st 04 03:39 AM
Need help with SATA RAID 1 failure on A7N8X Delux Cameron Asus Motherboards 10 September 6th 04 11:50 PM
Asus P4C800 Deluxe ATA SATA and RAID Promise FastTrack 378 Drivers and more. Julian Asus Motherboards 2 August 11th 04 12:43 PM
What are the advantages of RAID setup? Rich General 5 February 23rd 04 09:34 PM
Gigabyte GA-8KNXP and Promise SX4000 RAID Controller Old Dude Gigabyte Motherboards 4 November 12th 03 08:26 PM


All times are GMT +1. The time now is 05:45 AM.


Powered by vBulletin® Version 3.6.4
Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
Copyright ©2004-2024 HardwareBanter.
The comments are property of their posters.