RE: RE: RE: RE: 24 x 7 on NT?

From: Christopher Spence <cspence_at_FuelSpot.com>
Date: Tue, 26 Jun 2001 09:49:36 -0700
Message-ID: <F001.00338789.20010626093033@fatcity.com>

Ross Mohan for president!

"Walking on water and developing software from a specification are easy if both are frozen."

Christopher R. Spence
Oracle DBA
Fuelspot

-----Original Message-----
Sent: Tuesday, June 26, 2001 12:27 PM
To: Multiple recipients of list ORACLE-L

I am hearing such amazing stories...."running for seven years".... "no failures in 4 years"....."never any failures except when the NT administrator brought down the power grid", etc.

I am not an old hand, nor am I a greenhorn, but in my experience, "real, live production systems" ( e.g. more than 100 users, round the clock availability, frequent software updates...hardware adds to account for growth, etc. ) just don't run for four years without any downtime. I have never seen this. New systems have bugs shaken out....old systems have legacy MTBF hiccups....all systems need occasional hw/sw tweaks to accomodate unplanned business needs.

Now, if you factor OUT *scheduled* maintenance, then, hell, ANY system can stay up for months...years...decades. And, guess what? If you're NOT upgrading application or system software, or patching firmware or doing OS upgrades, it's not what I'd call a live production system. Hell, my HP calculator has been running whenever I want it, nonstop, since 1987.

As for running Nuclear stuff, I would NEVER run Oracle or Unix or NT for ANYTHING to do with Nuclear stuff ( missiles or power ). Oh My God. Please don't tell me any more about that. Even Oracle Corp says "don't use our stuff in places where people's lives are directly at stake."

(But that's just me.)

Lastly, this business about "being down for one minute costs us 12 Million dollars" is bohunk is most every case. There just isn't the data to support that. Yea, sure, maybe the a site's average intake is 12 Million during a typical one hour outage (that one site out of a million) but how many of those spurned customers come back? Most of them! Me, I can't get my book at Amazon, I just do something else and come back. ditto for my memory upgrade at Micron, or my tech info at Metalink. This "lost business" argument
is weak or NONEXISTENT in EVERY instantiation I have seen of it.

Also, a site being down can be anything...network...front line web servers...'
back end databases....intermediate LDAP servers....and the user ( that's you and I ) have NO WAY OF KNOWING for sure what failed. Ok...Ebay went down, repeatedly. They have IIS front end servers (which have not failed) and backend oracle databases on Sun E10K (which did). NASDAQ's reconciliation system just went down a few weeks ago ( Unix ) But that is a case where I have a mix of good press and backend information. As you note, most sites won't fess up.

I happen to work for a government client where we have aging Unix database servers of about five or six different flavors ( Siemens, DEC, Sun, Sequent, etc.)
that are pushed to their limits, feebly configured, and poorly maintained (due to
damagement "downtime" procedures) but very tightly maintained NT servers (due to
my company's downtime procedures ) and know what? My desktop has gone down ONCE
in two years. The mail servers for a 1000 user exchange system with 50 Mbytes per
user mailboxes has NEVER gone down in two years. The unix boxes have hiccuped on
disk...on memory...on oracle bugs.....

It's just too easy ( and too wrong ) to say "NT Sucks" or "Solaris Rules" or somesuch. (Not that you are, but....sadly, many do....)

Bottomline, I agree with you: If Management REALLY wants "24x7", then I just smile, and explain the costs to them. Before you know it, there are scheduled
hardware maintenance windows, oracle tuning/patching downtime, etc.

-----Original Message-----
Sent: Tuesday, June 26, 2001 9:58 AM
To: Mohan; Ross; Multiple recipients of list ORACLE-L

Well, I guess so if that was the only occurrence. I'll never know and I doubt
that they will fess-up.

At any rate, If one wants to use NT or any other OS for that matter in a 24x7
guaranteed manner then one should look into making as much as possible redundant. Back in my Blue Suit days we did a lot of cause and effect analysis,
particularly on Nuclear stuff, to insure that if one component failed there was
a redundant part to take over the tasks of the failed unit. We also did analysis to determine what the likelihood of the failure was and what the cost/benefit of having the redundant part was. Basically, if you can expect say
1 failure every 8544 hours and it will take less than 1 hour to correct the failure, is it worth the expense to have redundant hardware for that failure?
It's one of those things that needs to be evaluated on a case by case basis. In
the case of NT, you'd need a separate server and be running OPS. What is the
cost, what is the expected frequency, and is the loss >= the cost??

Good questions, but only you can provide the answers. In the case we have here,
out HP's fail once every 4 years on average over the 10+ years of history we have with HP. And each failure takes about 2 hours to fix. Now at $1000 per
minute of lost revenue that comes to $120,000. A dual server and OPS architecture would cost $190,000 just to acquire the hardware and software.

Definitely not worth the expense since all of the failures we've had have been
soft ones anyway.

Dick Goulet

____________________Reply Separator____________________
Author: "Mohan; Ross" <MohanR_at_STARS-SMI.com>
Date:       6/25/2001 4:56 PM

Wow. They must have known it
was you, Dick! <G>

so...."last April"....proceeding
scientifically, that's less than
one crash a year...better than
five nines, right?

;->

-----Original Message-----
Sent: Monday, June 25, 2001 4:47 PM
To: Mohan; Ross; Multiple recipients of list ORACLE-L

Ross,

I've had Dell's site crash on me before, last April right in the middle of
customizing a system. They apologized, but I went with Gateway anyway.

Dick Goulet

____________________Reply Separator____________________
Author: "Mohan; Ross" <MohanR_at_STARS-SMI.com>
Date:       6/25/2001 1:12 PM

Somebody should let Dell know. www.dell.com

They run on NT. When's the last time you heard about their site being out?

A $40 Billion company can't be all wrong about NT, can it?

-----Original Message-----
Sent: Monday, June 25, 2001 4:58 PM
To: Multiple recipients of list ORACLE-L

OK, after my vacation, I'll wade back into the fray!!

Ron,

To start with I do not believe it possible to guarantee that NT will be up
24x7, never mind Oracle. That is the main reason that we use Oracle ONLY on Unix (in one flavor or another) here. All of our NT servers require a periodic
unscheduled reboot, otherwise they do the unscheduled crash under Murphy's rules.

Dick Goulet

____________________Reply Separator____________________
Author: "Kevin Kostyszyn" <kevin_at_dulcian.com>
Date:       6/25/2001 12:31 PM

Wow what a can of worms that has just been opened!!! KK:)

-----Original Message-----
L.
Sent: Monday, June 25, 2001 4:07 PM
To: Multiple recipients of list ORACLE-L

I have a treasury application that needs to be up 24 x 7 except for scheduled downtime. Is there any way to guarantee an app will be available 24 x 7 on NT? Is anyone faced with this?

Ron Smith
Database Administration
rlsmith_at_kmg.com

--

Please see the official ORACLE-L FAQ: http://www.orafaq.com
--

Author: Smith, Ron L.
INET: rlsmith_at_kmg.com

Fat City Network Services    -- (858) 538-5051  FAX: (858) 538-5051
San Diego, California        -- Public Internet access / Mailing Lists
--------------------------------------------------------------------

To REMOVE yourself from this mailing list, send an E-Mail message to: ListGuru_at_fatcity.com (note EXACT spelling of 'ListGuru') and in the message BODY, include a line containing: UNSUB ORACLE-L (or the name of mailing list you want to be removed from). You may also send the HELP command for other information (like subscribing).
--

Please see the official ORACLE-L FAQ: http://www.orafaq.com
--

Author: Kevin Kostyszyn
INET: kevin_at_dulcian.com

Fat City Network Services    -- (858) 538-5051  FAX: (858) 538-5051
San Diego, California        -- Public Internet access / Mailing Lists
--------------------------------------------------------------------

Fat City Network Services    -- (858) 538-5051  FAX: (858) 538-5051
San Diego, California        -- Public Internet access / Mailing Lists
--------------------------------------------------------------------

Fat City Network Services    -- (858) 538-5051  FAX: (858) 538-5051
San Diego, California        -- Public Internet access / Mailing Lists
--------------------------------------------------------------------

Fat City Network Services    -- (858) 538-5051  FAX: (858) 538-5051
San Diego, California        -- Public Internet access / Mailing Lists
--------------------------------------------------------------------

Fat City Network Services    -- (858) 538-5051  FAX: (858) 538-5051
San Diego, California        -- Public Internet access / Mailing Lists
--------------------------------------------------------------------