|
|
Post by Robin Hood on Oct 14, 2005 22:11:49 GMT -6
Went out for another of the Wal-Mart upgrades last night, got onsite at 8pm, first server upgrade went without a hitch, the second though was a different story. The company I am doing these for sent me a bad hard drive, which in itself isn't enough reason to stop the upgrade, but when the 2nd drive on this server decide to die, they wanted me to do a back out, at this point the fuckin server itself DIED!!  Started getting parity errors on the motherboard. So we had to leave the site half-way upgraded, with plans to return to replace the motherboard in this server ASAP!! Nothing like leaving a business critical system tetering on the brink of failure. Basically if this one server fails they are out of business untill it is fixed. I finally left the site at 8am this morning. I was talking to some of the tech guys from this company and 75% of these upgrades on these IBM servers are failing. Not something that inspires confidence in a company. |
|
|
|
Post by Merlot Joe on Oct 14, 2005 23:01:58 GMT -6
Hopefully the third time will be a charm and everything will go okay.
I know how you feel. Last Sunday night while picking we blew a hydraulic motor in the middle of the night. 2 hours and $700. bucks later we where running again. The next night we blew a main belt drive. 3 hours and lot of cursing later we are running again.
When things go wrong at night, it really makes a long evening of it. Those 12 to 15 hours seem like days.
I know it's not the same as your server problems but it's similar.
Good luck guy!!!!!!!!!!!!!!!!
Joe.
|
|
|
|
Post by Robin Hood on Oct 15, 2005 1:34:21 GMT -6
In a previous life I worked as a millwright and spent many hours troubleshooting all kinds of machinery, it don't make any difference what you are working on it will go to pot in a heartbeat. I have installed HUGE (1,000 tons +) hydroelectric turbines for hydoelectric plants, and I have worked on some of the smallest hydralic valves you can imagine, if it is mechanical or electrical it is gonna fuck up at some point.
|
|
|
|
Post by Robin Hood on Oct 15, 2005 1:35:03 GMT -6
BTW, if it is both mechanical and electrical you have just doubled you odds at it fucking up!!
|
|
|
|
Post by guest on Oct 15, 2005 8:57:32 GMT -6
BTW, if it is both mechanical and electrical you have just doubled you odds at it fucking up!! uuuuhhh, no. Think about it. If you have certain odds of something fucking up with electrical and certain odds of something fucking up with mechanical, then your odds of something fucking up have not merely doubled. Do the math.....this *is* the pocket protector palace forum, isn't it?  |
|
|
|
Post by Chicago Jake on Oct 15, 2005 11:23:30 GMT -6
If Pa is the probability of item "a" failing, and Pb is the probability of item "b" failing, then the probability of either one of them failing is Pa + Pb - (Pa x Pb), assuming that Pa and Pb are independent of each other.
Or, look at it this way: the probability of NEITHER of them failing is (1 - Pa)(1 - Pb). Amounts to the same thing.
Trust me on this one, I teach Reliability Engineering......Jake
|
|
|
|
Post by Merlot Joe on Oct 15, 2005 11:54:57 GMT -6
Trust me on this one, Jake The only thing that is reliable is Murphy's Law. "What can go wrong will go wrong" or close to that. ;D And it frigin does. Trust me on this one ;D ;D Joe. |
|
|
|
Post by guest on Oct 15, 2005 12:18:32 GMT -6
If Pa is the probability of item "a" failing, and Pb is the probability of item "b" failing, then the probability of either one of them failing is Pa + Pb - (Pa x Pb), assuming that Pa and Pb are independent of each other. Or, look at it this way: the probability of NEITHER of them failing is (1 - Pa)(1 - Pb). Amounts to the same thing. Trust me on this one, I teach Reliability Engineering......Jake Exactly. And one needn't be a prof in RE to know this. Innumeracy is a greater problem than illiteracy in the developed world, in my opinion. |
|
Started getting parity errors on the motherboard. So we had to leave the site half-way upgraded, with plans to return to replace the motherboard in this server ASAP!! Nothing like leaving a business critical system tetering on the brink of failure. Basically if this one server fails they are out of business untill it is fixed. I finally left the site at 8am this morning. I was talking to some of the tech guys from this company and 75% of these upgrades on these IBM servers are failing. Not something that inspires confidence in a company.
