|
Devlin Gallant
Thought Police
Join date: 18 Jun 2003
Posts: 5,948
|
01-01-2005 14:13
Right on! (Dating myself again)
_____________________
I LIKE children, I've just never been able to finish a whole one.
|
|
Alan Edison
Ty Zvezda
Join date: 28 Jun 2004
Posts: 420
|
01-01-2005 14:16
From: Tread Whiplash The worst part is thinking you know what might go wrong - because then entirely DIFFERENT things go wrong!  At least, that's my experience in 8+ years of web/software dev...  Look, you can be pro-active; but proactive DOESN'T mean that you can successfully anticipate EVERY problem - or successfully plan the resolution to every problem that you DO anticipate! If it HAD turned out to be a simple "lack of spare parts" issue that killed SL for the last few days, I might be tempted to scold LL for not having a backup plan. But it turned out to be a nasty lurking software bug that was only triggered by the hardware failure. It always helps to have the FULL STORY, before you pass judgement. No user should be happy about the down-time; but considering that only poor server performance brought out this problem, I'm actually inclined to feel good about LL because this means their servers have traditionally performed rather well! I mean think about it: If their servers had been slow/unresponsive a lot in the past, this problem would have been found and killed a Looooong time ago! Take care, --Noel "HB" Wade (Tread Whiplash) But LL even admited that they COULD have prevented this by having the newer parts on stand by.
_____________________
Ty Zvezda
|
|
Devlin Gallant
Thought Police
Join date: 18 Jun 2003
Posts: 5,948
|
01-01-2005 14:19
Are you sure that's ALL they said, or was there a little more too it?
_____________________
I LIKE children, I've just never been able to finish a whole one.
|
|
Tread Whiplash
Crazy Crafter
Join date: 25 Dec 2004
Posts: 291
|
Did They??
01-01-2005 14:23
Sorry Alan, I didn't see that... I was on when the trouble started. I got knocked off for a couple of hours. They got the system limping and I was able to get back on - but I didn't stay on long because I didn't want to load the system and didn't have anything important to do. The next morning, the outtage was scheduled - and at THAT point things got bad, because apparently they tried to push up the time-tables on implementing a new system that wasn't ready. That caused problems, which led to them going BACK to the "gimped" old system, which brought out the "race-condition" software bug. At least that's my understanding of the sequence of events. I thought their talk about a "new system" implied both a hardware AND software changes. IF some of this was just a lack of hardware parts, well, I'll bet that won't happen again as long as they're a solvent company. I'd bet a lot of money that NONE of the LL folks are eager to repeat this debacle. Take care, --Noel "HB" Wade (Tread Whiplash) |