Please Scroll Down to See Forums Below
napsgear
genezapharmateuticals
domestic-supply
puritysourcelabs
UGL OZ
UGFREAK
napsgeargenezapharmateuticals domestic-supplypuritysourcelabsUGL OZUGFREAK

An ethical question

TheProject

New member
I'm having a disagreement with someone on my team regarding business ethics.

We have a server that has 3rd party memory in it. This server has been having memory problems for a while, which forces a reboot. We have replaced the memory, and the error persists. However, the server manufacturer refuses to service the box while the 3rd party memory is in it.

My idea was to temporarily put some of the manufacturer's memory in the box, get them to replace the motherboard, then put the other memory back in.

My coworker insists this is unethical, and that we are basically stealing from the manufacturer if we do this.

Thoughts?
 
It the 3rd party memory could have in any way caused the server problem, then it is unethical.

If not then you are just proving your case.
 
if the problem persists with manufacturers memory then no, it isnt IMHO, other than the fact ur lying to them.

if the problem doesnt p[ersist and u send it back then it is
 
besides, the motherboard is defective. they need to replace it. if putting in their own memory gets the desired result, I don't see how you're hurting anyone.
 
The server shipped with 256MB of RAM, and we put in 3rd party because it's a HELL of a lot cheaper.

Personally, after the troubleshooting we've done, I don't believe it's a problem with the memory because it doesn't follow the DIMM, it stays with the slot.

IMO, the manufacturer is just being difficult by saying, no, it doesn't void your warranty, we just won't fix it.

The other catch to all this is that I don't have enough mfr memory to do this right now, and would have to figure out a way to get some.
 
dude, you just waltz back into your coworker's office and tell them that multiple anonymous people from an internet chat board for weightlifting have spoken and the definitive word is that it is totally ethical.

also, you should pee on something in their office. it is a way of marking that you have won this battle.
 
I've been told that the server situation might go over some people's heads, so to put it another way:

Imagine you have a car, with Michelin tires, car is under warranty. Car dealer says, we won't fix the car without the original tires on it.

So, do you put the original tires back on the car long enough to get it fixed, then switch the tires back?
 
yeah - but my car has Bridgestone tires on it.
and they aren't what came with it.
they still fix everything.

hee hee :D


TheProject said:
I've been told that the server situation might go over some people's heads, so to put it another way:

Imagine you have a car, with Michelin tires, car is under warranty. Car dealer says, we won't fix the car without the original tires on it.

So, do you put the original tires back on the car long enough to get it fixed, then switch the tires back?
 
Put manufacturer memory back in and see if the problem persists. If so, then there would be nothing unethical about having them troubleshoot the server.

-Warik
 
HappyScrappy said:
dude, you just waltz back into your coworker's office and tell them that multiple anonymous people from an internet chat board for weightlifting have spoken and the definitive word is that it is totally ethical.

also, you should pee on something in their office. it is a way of marking that you have won this battle.

This is the same coworker that preaches the "Tony Gonzalez is the best player in the NFL" thing, and I actually gave him WODIN's list of tight ends, and he about blew a gasket.

He leaves earlier than I do, so I pee in his office all the time. Good thing he leaves his tupperware from lunch here on occasion.
 
HappyScrappy said:
dude, you just waltz back into your coworker's office and tell them that multiple anonymous people from an internet chat board for weightlifting have spoken and the definitive word is that it is totally ethical.

:FRlol:

Don't say "board for weightlifting" say "board for steroids". Its just that much more impressive.
 
Tony Gonzalez - the best player ever?!

OH MY GOD!!!

if I knew anything about football and who plays it today, then I might find some meaning in that.

you should toss back at him that Manute Bol is the best basketball player ever. I don't follow basketball at all, but Manute is the tallest black man I've ever seen brick a dunk multiple times in one game.
 
Y_Lifter said:
The Car example is known as the RICER RETURN

I just made that up off the top of my head, had no idea someone might actually have to do that.

Guess I'll be figuring out how to get some mfr memory in here...
 
Anyone that MOD's a vehicle knows that if you have problems during the vehicle manuf warranty, you must remove the Modifications before taking it in to the dealer for service.

Then put the stuff back on afterwards..

RICERS are notorious for this...
 
I work for a server manufacturer, and see problems like this frequently.
Server manufacturers design and test their equipment for specific configurations,
including memory from certain suppliers in certain lots known to be good.
There's a lot of variability in the quality of memory that has the same base specifications.
Your server probably needs the higher quality version to run properly.
If you put the manufacturers memory back in the machine and it functions properly, you are on the track to prove this point.
It may also be that you are having problems that only occur with the greater amounts of memory, and the problem goes away at lower amounts of memory.
I've seen this caused by both hardware and often software bugs.
Reboots are a common symptom of a software memory leak. Is you software up to recent maintenance levels?
It is very unlikely the problem is in the motherboard, and you will never find out until you put the manufacturers memory back in the machine and have them check it out.
 
john937 said:
I work for a server manufacturer, and see problems like this frequently.
Server manufacturers design and test their equipment for specific configurations,
including memory from certain suppliers in certain lots known to be good.
There's a lot of variability in the quality of memory that has the same base specifications.
Your server probably needs the higher quality version to run properly.
If you put the manufacturers memory back in the machine and it functions properly, you are on the track to prove this point.
It may also be that you are having problems that only occur with the greater amounts of memory, and the problem goes away at lower amounts of memory.
I've seen this caused by both hardware and often software bugs.
It is very unlikely the problem is in the motherboard, and you will never find out until you put the manufacturers memory back in the machine and have them check it out.

Well, here's the thing: I have two very similar servers, both running the same 3rd party memory. One works fine, the other reboots about once a month from this hardware error.

We bought the server with 3rd party RAM installed from the vendor, and I don't have manufacturer RAM to go back in it.

It just seems completely asinine to me to spend a few grand on more memory, when I'll lay odds that it won't fix the problem.
 
So, let me get this straight. I believe what I'm hearing is that the machine manufacturer sold the machine to a ISV (Independent Software Vendor) who creates/packages final turnkey solutions.
Yes ?
In this case, your warranty support is THRU the ISV, not the manufacturer.
The manufacturer has no idea what "customization" was done by the ISV.
If the ISV put unsupported low quality memory in the machine, then it is THEIR problem, not yours.
Once the ISV support is contacted, they will work with the manufacturer to get the problem resolved.
Manufacturers are highly motivated to keep their good relationships with their ISV's.
The manufacturer may/may not feel any responsibility directly to you the final user, except thru the ISV.

And it still sounds like it could be a software memory leak.
 
gonzo is the best tight end in football, easily.... perhaps he is a better tight end than any other player is at their position...

ie faulk at rb is awesome but you have holmes right there now too....

gonzo is further above the 2nd best tight end than anyone else is at their spot...


uh... tell your coworker the eagles play the gnats tonight and football is the most important thing in the world... fuck the server stuff... BBBBBBBBBBOOOOOOOOOOOOORRRRRRRRINNNNNNNNNNNNG
 
Current motherboards allow the RAM to be placed anywhere in the rows of DIMMs and you can have just one, or sets, and it is pleased either way.
But I recall back in the day when you needed to fill the banks in order, starting at the lowest position, and before that you had to do the same, but in pairs.
hot damn.

anyway - I had a workstation that I was doing a lot of 3d stuff on. I needed a lot of RAM for that ("a lot" at the time was half a gig - which was still pretty $$$ in '98).
my computer never crashed when it had less RAM - but when it got more, it would crash with memory errors all the time.

in the case of my problem, it was never noticed when there was less RAM because those banks were never filled.

my problem was much harder to test because I couldn't test all of the banks independantly via pulling out and interchanging RAM as easily as it can be today.

if the system works just fine with RAM in bank A, and/or C, but fails with it in bank B - which is what I thought you were saying - then it is sure as hell not a software error.
 
john937 said:
So, let me get this straight. I believe what I'm hearing is that the machine manufacturer sold the machine to a ISV (Independent Software Vendor) who creates/packages final turnkey solutions.
Yes ?
In this case, your warranty support is THRU the ISV, not the manufacturer.
The manufacturer has no idea what "customization" was done by the ISV.
If the ISV put unsupported low quality memory in the machine, then it is THEIR problem, not yours.
Once the ISV support is contacted, they will work with the manufacturer to get the problem resolved.
Manufacturers are highly motivated to keep their good relationships with their ISV's.
The manufacturer may/may not feel any responsibility directly to you the final user, except thru the ISV.

And it still sounds like it could be a software memory leak.

Actually, we bought the server from an authorized reseller, not a software vendor. We installed the software ourselves here.

I'd agree with you on the memory leak thing, except the server hardware monitoring always gives the error, and NT returns a hardware fault BSOD, not a software BSOD.
 
HappyScrappy said:
Current motherboards allow the RAM to be placed anywhere in the rows of DIMMs and you can have just one, or sets, and it is pleased either way.
But I recall back in the day when you needed to fill the banks in order, starting at the lowest position, and before that you had to do the same, but in pairs.
hot damn.

anyway - I had a workstation that I was doing a lot of 3d stuff on. I needed a lot of RAM for that ("a lot" at the time was half a gig - which was still pretty $$$ in '98).
my computer never crashed when it had less RAM - but when it got more, it would crash with memory errors all the time.

in the case of my problem, it was never noticed when there was less RAM because those banks were never filled.

my problem was much harder to test because I couldn't test all of the banks independantly via pulling out and interchanging RAM as easily as it can be today.

if the system works just fine with RAM in bank A, and/or C, but fails with it in bank B - which is what I thought you were saying - then it is sure as hell not a software error.

We have rotated the memory in the server itself, but not left that socket empty for any period of time. I could do that, but it will affect app performance, which will make me unpopular. The error stays with socket 3 consistently, but I can't make the error happen.
 
HappyScrappy said:

in the case of my problem, it was never noticed when there was less RAM because those banks were never filled.

my problem was much harder to test because I couldn't test all of the banks independantly via pulling out and interchanging RAM as easily as it can be today.

if the system works just fine with RAM in bank A, and/or C, but fails with it in bank B - which is what I thought you were saying - then it is sure as hell not a software error.

I agree completely, and have experienced this also.
You could have a problem with one of the DIMMS but not the others, or you could have a problem with a specific DIMM position.
But the most important thing you can do now, is to get you into a "supported configuration" whatever that takes.
You need to have either the ISV or the manufacturer agree you are in a supported configuration before they will take responsibilty for helping you.
 
TheProject said:


We have rotated the memory in the server itself, but not left that socket empty for any period of time. I could do that, but it will affect app performance, which will make me unpopular. The error stays with socket 3 consistently, but I can't make the error happen.

would they rather have a few hours of slower system performance, or continual crashes?

I'd think that they'd want the problem resolved and are already experiencing crappy performance (reboots), so I would wonder if they would even notice.
 
See, I have to tell the department what we're doing, and whether or not there is any actual performance decrease or not, they'll probably bitch about it. I'm trying to keep my bitch quotient as low as possible.

For other reasons that I can't discuss here, I can't contact the hardware vendor. The software vendor, sure, but they're useless to me in this situation.
 
is this big iron sorta stuff, or just a run of the mill server? how crucial is it to the day to day things? is it a redundant system?

I'm just curious - if it is redundant, then pull it out and work on it in isolation and don't tell them.

also, I'm curious why you can't contact the hardware people. you stole this from some kid didn't you? you beat him up and took it?
pathetic. that is just like you.
bastard.
 
HappyScrappy said:
is this big iron sorta stuff, or just a run of the mill server? how crucial is it to the day to day things? is it a redundant system?

I'm just curious - if it is redundant, then pull it out and work on it in isolation and don't tell them.

also, I'm curious why you can't contact the hardware people. you stole this from some kid didn't you? you beat him up and took it?
pathetic. that is just like you.
bastard.

We have no redundant servers, unfortunately, and it's pretty much mission critical.
 
There is nothing unethical about it Project.. i am CFO of a computer networking/systems integration firm and we do swaps like this all the time.. Especially with DELL. Our customer service rep at DELL actually suggested it to avoid any warranty issues.
 
TheProject said:
I'm having a disagreement with someone on my team regarding business ethics.

We have a server that has 3rd party memory in it. This server has been having memory problems for a while, which forces a reboot. We have replaced the memory, and the error persists. However, the server manufacturer refuses to service the box while the 3rd party memory is in it.

My idea was to temporarily put some of the manufacturer's memory in the box, get them to replace the motherboard, then put the other memory back in.

My coworker insists this is unethical, and that we are basically stealing from the manufacturer if we do this.

Thoughts?

DO IT! It is done all the time and if it is their motherboard that is messed up it doesn not matter whos memory is there. They are just being pricks.
 
Top Bottom