Slashdot Mirror
Equipment Suppliers You Can Trust?
Steve Gray asks: "It has happened to all of us at some time or another. You're two weeks from deploying an application, but suddenly your testbed server falls over, and just won't get back up. After fighting with a variety of companies to try and get parts delivered for Tuesday, I'm finding that most companies will stall your order for days for reasons from random extra checks through to migration of lesser known species of Vole, business needs be damned! Who do Slashdot readers turn to when technology goes wrong? Do you trust them to deliver by tommorow, without fail?"
If something critical breaks and we need standard(ish) parts then we bite the bullet and drive 30 miles to Scan, who in general have most stuff in stock.
Yeah, you pay a slight premium but it's worth it. I suppose you may want to consider next day on site repairs from the manufacturer as part of an extended warranty or service agreement.
PocketGamer.org - For the gamer on the go!
I prefer local small businesses, they need you maybe more than you need them.
perpetually dwelling in the -1 pits
If you pay, the offensive amounts of money they ask, they even will code for you...
On the other hand, Keep a small stock to be out of troubles your self.
2 o 3 spare hard disk, 1 GB ram, the hardware you need and the bugdet you have...
If is that important backup equipment, redundancy, etc, and always, have 2 or 3 plans of action. Even if you get a 100% next day whatever-you-need replacement, you still need the plan b, and c...
Check tue bussines continuity plans and risk managment theory, you will get pretty good ideas of what to do... isnt so hard.
Â_Â
Agreed. It's foolish to try to run a shop without spare parts on hand, especially for anything remotely critical. Time and experience has taught me over and over, that if you are not prepared, it will be made known to those who you'd rather it not be made known to.
What that overnight shipping costs on some parts would pay for the part itself. Keep spares on hand.
- Eric
We used to have actual Field Service contracts which guaranteed two hour response time, and that meant someone was on site in two hours, not returning a call within that time
Well I don't trust these field service contracts too much, unless I know the supplier has local or regional stock. I've seen it way to often recently, these companies (HP, Dell, EMC) can get you an engineer on site in 2/4 hours, but the spare parts might take a lot longer than the agreed time.
Hot spare.
I know IBM can do it, if they're paid enough.
My company (130,000 people) outsourced it's hardware support to IBM. Just at my location, depending on severity, we've had response times of less than a half-hour (when our IBM 3174 failed to reboot after a power failure, cutting off half our building), to days (when a single monitor released it's magic smoke).
Yup, I agree with the above. In fact, I would go further and say that you have to regularly practice stuff like replacing a drive, or restoring a database to a backup server, to make sure your knowledge and procedures are up to snuff and documented.
Let the Astroturfing BEGIN!
Repetition does not transform a lie into the truth. - FDR
I wonder how many of these posts will be lacking a 'Full Disclosure' announcement... or have a false one?
I guess this is a Create-Your-Own-Slashvertisement?
When our computer equipment breaks down, I like to go to a specific local store. They're 5 minutes away, carry quality parts at very reasonable prices, cheap "off the boat" parts are nowhere to be seen, they have a good return policy, and they speak ENGLISH. (This is more of a concern than you'd imagine, in a big city.)
My boss, on the other hand, likes to go to Tiger Direct and buy the cheapest crap they have on the shelf.
Does it make you happy you're so strange?
Recently I ordered a whole media box for a customer worth around $2000 from tigerdirect and I needed it fast fast fast. It came on time, but my heart attack came when I checked my bank. They charged twice, docking my account almost $4000 (they were nice enough not to include shippping in one). After going through many zombies I finally got a rep that could tell me what the hell was going on. Apparently it is a hefty sum and they decided to 'freeze' the sum of my purchase and then proceeded to charge for the same sum + shipping. I had to mess with this for a week before I could pay my damn bills. However this taught me a good lesson that I should have already had in high school. Never underestimate the value of human contact. Yeah we're all nerds and want to stay away from the Worst Buy commission gangs on sugar highs trying to sell you this nice wireless bluetooth toothbrush with mp3 player and a free headset, but when it comes down to it, somebody has to be responsible for whatever happens to your order. Can't blame the server BSODing in the middle of your order and charging you again after coming back online because someone didn't know what a friggin mutex is. But the best way to avoid all this is to forget the online stores and get the critical stuff locally, personally. If anything happens, you got someone to point your finger(choose carefully) at, even if its yourself.
"You're everywhere. You're omnivorous."
I wonder what the monetary value of this story is? It's essentially free advertising for companies on a website filled with nerds who order lots of equipment online and have no qualms about doing so.
I like newegg.com - and I wonder how much revenue they get directly attributable to this story and this comment.
we've ordered from newegg and received it the next day. newegg must have some really interesting deals with their shippers. weird, wild stuff.
although ever since newegg opened the separate warehouse on the east coast, their shipping has gotten a lot slower. it's typical to have half your order shipped from CA and the rest from the east coast 3 days later.
2 o 3 spare hard disk, 1 GB ram, the hardware you need and the bugdet you have...
With the possible exception of hard disks, the part that is [overwhelmingly] the most likely to fail, and, several years down the road, among the most difficult to replace [because form factors will have moved on to new standards] is the power supply.
Always purchase several extra power supplies for any mission critical system.
Technology costs money. If you require a fix within a certain time, are you paying someone to provide that fix within a certain amount of time? If not, then you have failed to plan. If you fail to plan, then of course you've planned to fail.
If you have no money, but you still want to be able to restore your system from disaster within a certain timeframe, you must of course ensure you are able to do that yourself with the parts and equipment you have on hand.
I am government man, come from the government. The government has sent me. -- G.I.R.
What I have found to make the difference is relationships.
If you know someone closer to your end of things, and you can work with that person, you will get far better service. In support, it's the guy who says "here's my pager number in case you have trouble with this" even if he doesn't want you to call him every time you have trouble. The flip side of this is that eventually you know which guys break more than they fix, or close tickets without even calling. Knowing the local service manager or dispatcher is a real help here, or more accurately, the more people you know, the better it gets.
In sales, you need a Rep who will work with you, and has some power. I mean the guy who says "I'll get you some of those tomorrow" and you may not even see a bill for them (although you also might be billed at the real value - you NEEDED those, right?) This is the guy you buy your redundant supplies from when things are calm, so you don't always have to rely on him dropping everything for you. This is not the guy who won't lift a finger without a signed PO.
Contracts aren't worth as much as you'd like.
I found IBM four hour turnaround time to be an exception even in the early nineties, and it hasn't gotten better. Admittedly, we were the low end of the market, but we still had a four hour contract with IBM, and it was honored almost exclusively in the breach. I have not seen anyone significantly better since then either. It just doesn't happen. I have occasionally gotten stellar support from IBM, Dell, HP, Compaq and Cisco, but that was always completely localized, never reliable with any single vendor. FedEx has built their reputation on promptness and reliability, not becasue it's easy or common, but rather because it's difficult and rare.
Let's not talk about contractors. Some kind souls cannot be bought or bound by a piece of paper. Those things only enable them to help you, as demonstrated by random arbitrary work interruptions. You may not see them for weeks at a time in the middle of an urgent job, but remember that these kind souls, martyrs really, help you stave off catastrophe out of the goodness of their hearts alone.
Ultimately, it's the people who make it happen, like the FedEx driver who scanned my package at 6:04 last night as he got into his truck, and waited while I went inside to get a piece of tape from the the counter guy who told me I was too late.
I hope you get lots of good recommendations for companies that will deliver quickly and reliably, and I'll keep an eye on this thread to see what people have to say. Meanwhile, be nice to your office manager.
Assembly is the reverse of disassembly.
If your company can't afford extended downtime, then your company can't afford to not have a service contract on your hardware. The service contract is, of course, only as good as the company behind it. That's one of the reasons for buying gear from the grown-up companies.
Most of our gear is Sun (~100 mid-sized servers, say 6CPU each on average), and production is under expensive service contracts. When something goes boom, Sun is onsite, diagnosing as necessary and repairing ASAP. Parts orders are delivered in one hour. This is how you run a business.
It's not expensive service, it's cheap insurance for the company.
"People who do stupid things with hazardous materials often die." -- Jim Davidson on alt.folklore.urban
At my last assignment, we used AVID video editors exclusively. We had a 24/7 next day delivery contract with them (12K per workstation!!!)
When our server died, (more accurately, the JBOD case) we had a new one the next day.
I guess it depends on what you're doing, but in that circumstance, it pays to use the industry leader - they're number one for a reason.
We're talking about an office environment here, not a home.
Office:
From Jan. 2002 - Jan 2005 we had OptiPlex GX110 desktops. 733mhz, 512mb and 40gb hdd's. Nice little PC's, and they served us well. In Jan 2005, out rolled the OptiPlex GX270's - 2.8ghz, 1gb and 80gb hdd's.
Initially, i thought 'whoah, this is a lot of PC'. Three months later, a new version of an application that runs on 85% of the desktops was released. Minimum specs: 2.0ghz and 1gb.
Home:
My desktop rig is a 1800+ w/ 1gb that i've had for almost 5 years. Still got the p3/500 w/ 512 that it replaced. That served as a replacement for the p120 w/ 256 that I was running debian on.
As for my car it's hardly under warranty - 142,000mi and still runs like new. You might get a kick out of the fact that I just put new brakepads on it in my garage last month!
If a power supply croaks or a processor fries at home, there is minimal cost associated with it. You buy a new part, and replace it. No rush other than our geeky sense of pride.
When something dies in a business, it costs money. Not only does the problem need to be fixed, but the employee who's machine went down can not do their job - thats where the real cost is.
"Gawds. We used to have actual Field Service contracts which guaranteed two hour response time, and that meant someone was on site in two hours, not returning a call within that time."
Back then hardware was reliable enough that the manufacturers could afford that luxury for the few times things did break down. Nowadays, they want to cut costs to stay "competitive," and the first thing to go it seems is reliability.
There is no other way than to keep spares on hand.
Someone will claim you can't keep a backup of a big database server or other huge machine and the solution to that is redesign the problem so it uses several smaller and cheaper servers.
Another solution is run your disaster recovery site live.
I actually had a rep apologize for not being able to get a 2 hour response time. (An odd add in raid controller failed... at least they claimed it was part of the raid controller setup)
Not sure how much the vendor paid for that particular IBM contract, but the service level is quite nice.
The JIT model isn't so bad and it would seem some companies are building around that. I had some time to chat with the service tech and he was telling me about the shipping setup various companies have. Dell actually had a facility nearby that warehoused and shipped out parts as needed. (Not anywhere close to a Dell facility, but just a warehouse/shipping rig) It would seem he wasn't just an IBM remote tech, but actually was shared among several companies.
So this fellow can actually have parts ordered and drove out on a moments notice from at least IBM and Dell.
"You should always go to other people's funerals; otherwise, they won't come to yours." -- Yogi Berra
I work for a large aerospace company which has 2-hour service response contracts with all major hardware vendors - HP, IBM, Sun, SGI, Dell, etc. The service is not what it used to be. Before we actually had tech reps on site. Or at least they would come over within the 2-hour window. They usually would be carrying replacement parts. The right parts.
These days our admins consider it luck if within two hours they get a service call from India. And then its the game of "find that part number in your half-assed outsourced overseas database of spare parts from every vendor in the world." They always want to know if they can just mail the part so you install it yourself, or if you want an actual field tech to come out, since you have that fancy "platinum" support plan anyway. And then they ask you how does next Thursday sound. Motherf...
Having spares on site is a good idea, but with the variety of hardware we have, it would be too expensive to cover all critical systems (and according to our DBAs and users every last stinking workstation needs to be 24x7). And even having the right spare doesn't always save the day. Here's a fun little story: A couple years ago we got a few Sun A3500 arrays (may they burn in hell). I insisted we also buy a couple replacement disks in case shit. A month later we lost three hard drives in less than 40 minutes. Go figure. After much whining Sun agreed to test the drives and found a defect.
Service is goin down; hardware quality is going to hell; prices for both are going up; and only my salary stays the same.
"JIT" is where the company attempts to predict exactly how many parts it will need tomorrow and only order that number of parts from its vendors today.
Those vendors also practice "JIT" with the vendors supplying them with parts.
So, it all breaks down on those days when the demand is higher than any of the companies anticipated.
Warehouses cost money, storing parts that aren't needed today costs money. JIT is supposed to save all of that money by predicting exactly what will be needed and how long it will take to get it and then having the part arrive at the company Just In Time to be shipped out to you.
"When you perform miracle after miracle to save their asses time after time, they get lazy and start believing that that is the natural order of things."
Yes! I enjoy my job and I'm always very helpful to everyone around me, including the dev team, end-users, and other MIS folks. I get it done fast because I know that's how I like it when I ask for something, too. Unfortunately, it does bite you in the ass. You do need to eventually push back a little bit, or else you'll end up in your scenerio. It sucks.
When someone asks me to do something for a third time, I'll give them some tongue. One of the help desk guys is now into getting user profiles re-created to fix just about EVERYTHING. Since only myself and a couple others have access to do it, I've had to fuck with roaming profiles all day instead of doing my project work. So yesterday, I bitched him out about it. I said "What's the problem?" and I fixed it in four minutes without touching the user profile. I then proceeded to lecture him on how it should now be considered a last resort.
I usually tell my manager about such things, in the event that someone complains that I'm not doing my job, even though I've done it better then the last three admins in this place. Unfortunately, if it's management porking you, you can't do anything about it. Then, you have to decide whether or not to find a new job. Fortunately, in IT, one of the only ways to advance your career is to change jobs, so it's not like finding a new job is anything new to most of us =)
- It's not the Macs I hate. It's Digg users. -