Yeah, HP OpenView. Twice the functionality and 40 times the work. Well, if you count them out of the box, What's Up has more functionality than HP OpenView. Buy some extra crap and get an expert to set it up, and you might be able to get to the same point after $50,000 and a few months with HP OpenView as it takes with What's Up and a week.
And open source beats the heck out of both on functionality and cost at least on the low to mid-range.
As for What's Up, I've never seen it used in a complex enterprise or carrier environment, HP OpenView, Tivoli, or SunNet Manager and the like are pretty much the standard there.
I've done some mid-size OpenView setups before and it isn't that hard to get working, particularly if what you are primarily interested in is SNMP monitoring.
I'll admit I haven't dug too deeply into What's Up but from my limited exposure it seemed like an overpriced toy. Perhaps that is due to it being very windows-centric when what I was trying to monitor was network gear and UNIX/Linux servers. Still that tends to make it very much not a cross-platform tool.
These days I'd likely use an open source monitoring solution unless I really needed some functionality (or integration of functionality) I simply couldn't get except by using a commercial tool.
What's Up Gold and spending $20,000 for configuration and buying it and setting it up in house.
Dear lord, on what planet is What's Up Gold considered a 'world-class' monitoring solution? Frankly it is a steaming pile of crap that most OSS monitoring solutions outclass easily.
If you want world-class cross-platform monitoring solutions you need to go with something like HP OpenView or Tivoli.
The average Westerner uses something on the order of 100-1000 times the water that someone in a developing country does. Now, imagine everyone in the world using water at developed-nation levels.
Not the same as oil, or many other resources, water generally isn't shipped long distances so it isn't as if the US and China are competing with each other for water.
Sure, water recycles, but not that fast. Not thousands of gallons a day for billions of people fast.
Within developed nations at least all you have to do is feed the output from the sewage treatment plant into the input of the water treatment plant. While you do lose some water along the way you can also dramaticly cut the net water consumption of say Santa Barbara.
Food is another biggie. Sure, there's plenty to feed the world now, but that's because few countries eat meat at the level that developed nations do. Doubling the amount of meat consumption worldwide would multiply world grain needs by a much larger factor (I think the equation is something like 1 pound of beef = 10 pounds of grain).
While the popularity of some particular foodstuffs may make them more expensive people will generally switch to something else. I think beef is currently overpriced so I eat Pork, Chicken, or Lamb. For that matter you don't have to raise many meat animals on grain. You can graze cattle on rangeland that really isn't suited to raising crops.
Until fairly recently men typically drove more miles per year than women on average.
While male drivers do seem more prone to engage in certain types of stupid behavior, the more you drive the more likely you are to get into an accident.
Megers, feh! There are only stupid takeovers and product assimilation in the business world. Someone is either buying your top talent (never happens) or your top product (to shut it down.) If you couldn't hire away the top talent before, they won't stay now. Product closures is just capitalism in reverse. There is less competition and far too often the inferior product is the only product left in the market. People that think mergers are like happy marriges of two companies have never personally seen the look of pure greed on a pre-merger CEO's face.
I'm not bitter though, it's just that the idea of mergers and the reality of mergers have nothing to do with each other.
Sometimes mergers do work, take Apple and NeXT, Price and Costco, or any of the number of companies that Berkshire Hathaway has bought.
Then again the corprate cultures of the merging companies in the first two examples were very similar and Berkshire tends to not mess much with the companies it buys leaving exsisting management and employees in place.
Many customers claim that Comcast overall has been an improvement over AT&T Broadband, Verizon an improvement over GTE or Airtouch, or Wells Fargo an improvement over First Interstate. There is some thought that for many customers Cingular will be an improvement over AT&T Wireless (which has been going downhill ever since most of the old McCaw executives left a few years ago).
Peet's was first being founded by Alfred Peet in 1966.
Starbucks was founded by Jerry Baldwin (current CEO of Peet's), Gordon Bowker (also a co-founder of Red Hook Ale and The Seattle Weekly), and Zev Siegel who were all fans of Peet's coffee. Peet essentially taught them everything they knew about buying, roasting, brewing, and selling gourmet specialty coffee.
Howard Schultz didn't join the company until 1982 when he became the company's marketing manager (he esssentially lobbied Baldwin, Bowker, and Siegel for the job).
In 1984 the owners of Starbucks jumped at the chance to buy Peet's (Peet was retiring).
In 1985 Schultz left Starbucks to pursue his vision of Milanese style espresso bars by founding Il Giornale. He received considerable help from Baldwin and Bowker, both financial and in retailing and marketing advice.
In 1987 Baldwin and Bowker decided they wanted to sell Starbucks. Bowker wanted to focus on Red Hook and Baldwin wanted to focus on Peet's. Schultz made the winning bid for Starbucks and merged it with Il Giornale. The rest they say is history.
Schultz hired the founder of Seattle's Cafe Allegro to be Il Giornale's retail operations manager. He continues to serve as a Starbucks executive to this day.
Jim Stewart founded The Wet Whisker in 1970, which was later known as Stewart Brother's Coffee, then SBC, then as Seattle's Best Coffee. SBC's parent company Seattle Coffee Company was acquired by Starbucks in 2003. Stewart was also influenced and mentored by Alfred Peet.
Jerry Baldwin continues to serve as Director of Peet's Coffee & Tea. Baldwin's prior association with Starbucks probably explains some of the ill will between the two companies.
Some independent coffee shops are very good though. Here in Seattle in addition to Zoka mentioned in the article we have Diva, Uptown, Vita, Allegro, among others. Any of them are a far cut above what Starbucks or the other chains put out.
The concern I have is: surely the next time you go for a piss someone will just rob your computer.
Fairly unlikely in that part of Seattle (not a lot of petty crime) and if someone did try it I suspect that co-workers, the other regulars, or the staff would try to stop the miscreant.
And how do you get on the Internet in a cafe?
Free wireless internet provided by the cafe.
It really is quite common here, not to mention all of the places that will provide wi-fi for a fee.
Shipley was a co-founder and former President of Omni I doubt a lack of control had anything to do with his leaving.
I suspect it had more to do with wanting to work on projects that the rest of Omni wasn't interested in pursuing plus an itch for the challenge of doing another startup.
I somehow doubt they'll have much trouble getting an office when and if the time comes for that.
Really, talk about the pot calling the kettle black.
If the datacenter that hosts Slashdot was to have a massive power failure how long would/. be down for?
That said my company has gear in the same datacenter as LJ, our servers were back up 10 minutes after power was restored. Then again we use Oracle on HP-UX with nice SAN RAID boxes for storage for our database. So our stuff tends to recover from a sudden power loss a little better than a MySQL derivative running on clone hardware.
Whatever are you on about? InteNAP isn't even in the top 10 spam hosting ISP's. They don't tolerate spammers on their network as a general rule.
On the other hand they aren't going to boot a customer just because some crank complains about getting a commercial email from that customer. It does happen too. I've seen people sign up for a product newsletter. Confirm the subscription and then complain that the newsletter was spam to every single place they could possibly think of reporting it.
For contrast compare them to MCI, XO, or Above.net. All of those providers are far more spam friendly.
Datacenters have to have a BRS by the exit doors both in case some tech shorts himself across a power feed and so the fire department can kill all of the power to the room as killing the power feed to the building won't do it.
That's debatable, and depending on what database you use, having more than one database server (or pool of database servers) in different physical locations that are kept in sync at all times is definitely possible. I'm not sure whether MySQL allows this, but I think if you have a site that has nearly 6 million users, more than 100000 of which are paying you for the service you provide (I'm one of those, one might add), then you really should look into doing just that - or at least I hope the LJ people will do now (I don't really want to blame them for the problem).
Even with Oracle or DB2 running database servers in multiple locations that need to stay in synch is a massive pain in the ass that you avoid if you can. Given that other problems are at least or more likely to cause your site to be unreachable to at least some customers for longer periods of time than problems with a decent co-location center it really isn't worth the bother for the majority of sites out there.
Hell last time I checked Slashdot still had all of their servers in a single datacenter.
For that matter I've seen times with my Verizon DSL line that I've been unable to reach Google, Yahoo, Amazon, Expedia or other internet sites with a high amount of redundancy including multiple datacenters for more than a day due to some problem on Verizon's network or with intermediate backbone.
But I don't blame say Google for the problems because they really have done everything they reasonably can to make sure their stuff is up and as many people as possible can reach their servers at any given point in time.
That being said, I think you didn't quite understand what I was trying to say. I really don't care whether they have "plenty of backup power", "plenty of generator capacity" and "top-of-the-line big datacenter grade stuff" (which really sounds more like a collection of buzzwords than anything else, anyway). If a wiring fault (of whatever kind) can bring up the entire UPS system as well as the "generator capacity behind that" and all other safeguards they supposedly had in place as well, then it's just worthless and a waste of money - a UPS is supposed to be an *uninterrupted* power supply.
By their nature multiplely redundant power systems are a tricky thing that even experts fail to get right. Even then someone hitting the BRS by the door that is required by fire code can still take power to the whole datacenter down. InterNAP is in the process of expanding the datacenter in question which means there are contractors running around all over the place. It is quite likely that someone miswired a new UPS. Shit happens, even to the datacenters of major banks, stock exchanges, or airline reservation systems.
And while I admit that it's not possible to guard against *all* problems, saying that the colo facility is "one of the most solid in the state" and supposedly can't be taken offline by something "short of a direct strike from a comet" is just silly when a "wiring failure" can bring down the whole thing, and even more so when it's not the first time that happens.
Really, this just stinks of an attitude that's all too prevalent in parts of the IT industry - just piecing together the components of a reliable system won't necessarily give you one, and if you can't build one properly, then don't go advertising that you have one. Don't you think the fact that the LJ people are now planning to buy their own UPS equipment to use on top of the facility's should tell you something?
The Fisher Plaza datacenter really is one of the best in the state. It is far from "pieced together". The InterNAP people aren't a bunch of clowns and do acutally know what they are doing (or at least as much as any other top-tier colocation vendor).
Especially in the cost vs. quality tradeoff InterNAP does really well which is why a fair number of companies around here buy rack space from them.
I do find it somewhat amusing that InterNAP's first datacenter, which really is
Maybe they just blew the rest of the year's allowable downtime.
I don't have our contract in front of me but I believe InterNAp just blew the next two years allowable unplanned downtime.
Fortunately for them their SLAs are only year to year or they would still be working off the time a customer hit the BRS thinking it would open the door.
The co-location facility in question has plenty of backup UPS power with plenty of generator capacity behind that. Supposedly there is enough generator capacity to fully power everything in the building including the network TV station even with one generator out.
The UPS gear in Internap's space is all top-of-the-line big datacenter grade stuff. Apparently there was some sort of wiring fault in one of the new UPSes they were bringing online that caused both building power to fail and the self-protection circuits in all of their UPSes to trip.
IOW it was either a faulty UPS or a faulty wiring job by the electrical contractor.
Livejournal isn't the only ones who got burned by this outage. The colocation facility in question is supposedly one of the most solid in the state and nothing short of a direct strike from a comet is supposed to be able to take it offline. My company was in the same boat as our gear is in the same facility as LiveJournal's.
Sure both LiveJournal and the company I work for could have hedged our bets by having redundant gear in another facility in another state, but that is a pain in the ass especially when backend databases are involved. To tell you the truth it probably isn't really worth the bother unless you truely have a need for six nines of uptime.
They helped kill two of my favorite local bookstores, leaving exactly none. I said "helped" because Barnes and Noble are comparably evil... but you asked. Also, please don't ask me about Starbucks!
That's funny I live in Seattle which is sort of ground-zero for both Amazon and Starbuck's (not to mention a certain large software monopoly). The thing is we still have plenty of non-chain bookstores here. If you ask the owners what the biggest threat to their business is they will usually mention the chain bookstores like B&N or Borders long before they mention Amazon.
As for Starbucks I really don't get the accusation that they kill off other coffee places. If anything Starbucks increases the local market for decent coffee by introducing people to it in the first place. In Seattle for example, there are several other medium-sized coffee chains that started in the shadow of Starbucks, along with several small local coffee chains and countless independant espresso bars and stands.
We saw the same problem a few years ago, in DB2 Support. Customer kept getting a system crash around the same time of day, nothing of consequence in either the db2diag.log or the system logs. The customer couldn't recreate, and traces were getting us nothing. Eventually, as the customer kept blaming us (of course) and went critsit, we asked him to physically monitor the system terminal to see what was going on. As he was on a conference call, a janitor came in and (IIRC, got this story second-hand) plugged his vacuum into the UPS circuit.
I've seen similar in person. At the university lab I used to spend way too much time in back in the late 80's and early 90's I watched a cleaner plug his vaccum into the same circuit as a bunch of the lab's X-terminals, macs, and NeXTs were on. There was a 'pop' sound and about 30 X terminals, 15 macs and 10 NeXTs all suddenly lost power. The resulting power surge killed a number of devices that happened to be on the same circuit along with many others that were coupled to them electricly via 10b2 cable outright. For at least a year afterward devices would die and when pulled apart would show evidence of arcing on the circuit board from a power surge.
Needless to say the university managed to rewire the lab areas in that building with outlets off of the big power conditioner in back for the datacenter gear in record time. Fortunately the janitors already knew better than to plug into the special orange outlets if they wanted to keep their jobs.
Slashdot Mirror
Yeah, HP OpenView. Twice the functionality and 40 times the work. Well, if you count them out of the box, What's Up has more functionality than HP OpenView. Buy some extra crap and get an expert to set it up, and you might be able to get to the same point after $50,000 and a few months with HP OpenView as it takes with What's Up and a week.
And open source beats the heck out of both on functionality and cost at least on the low to mid-range.
As for What's Up, I've never seen it used in a complex enterprise or carrier environment, HP OpenView, Tivoli, or SunNet Manager and the like are pretty much the standard there.
I've done some mid-size OpenView setups before and it isn't that hard to get working, particularly if what you are primarily interested in is SNMP monitoring.
I'll admit I haven't dug too deeply into What's Up but from my limited exposure it seemed like an overpriced toy. Perhaps that is due to it being very windows-centric when what I was trying to monitor was network gear and UNIX/Linux servers. Still that tends to make it very much not a cross-platform tool.
These days I'd likely use an open source monitoring solution unless I really needed some functionality (or integration of functionality) I simply couldn't get except by using a commercial tool.
What's Up Gold and spending $20,000 for configuration and buying it and setting it up in house.
Dear lord, on what planet is What's Up Gold considered a 'world-class' monitoring solution? Frankly it is a steaming pile of crap that most OSS monitoring solutions outclass easily.
If you want world-class cross-platform monitoring solutions you need to go with something like HP OpenView or Tivoli.
The average Westerner uses something on the order of 100-1000 times the water that someone in a developing country does. Now, imagine everyone in the world using water at developed-nation levels.
Not the same as oil, or many other resources, water generally isn't shipped long distances so it isn't as if the US and China are competing with each other for water.
Sure, water recycles, but not that fast. Not thousands of gallons a day for billions of people fast.
Within developed nations at least all you have to do is feed the output from the sewage treatment plant into the input of the water treatment plant. While you do lose some water along the way you can also dramaticly cut the net water consumption of say Santa Barbara.
Food is another biggie. Sure, there's plenty to feed the world now, but that's because few countries eat meat at the level that developed nations do. Doubling the amount of meat consumption worldwide would multiply world grain needs by a much larger factor (I think the equation is something like 1 pound of beef = 10 pounds of grain).
While the popularity of some particular foodstuffs may make them more expensive people will generally switch to something else. I think beef is currently overpriced so I eat Pork, Chicken, or Lamb.
For that matter you don't have to raise many meat animals on grain. You can graze cattle on rangeland that really isn't suited to raising crops.
It is usually better, since you get a much better deal.
Performance, quality, price. Pick any two.
It applies to pretty much everything from computer hardware and telecommunications to cars and airplanes.
Until fairly recently men typically drove more miles per year than women on average.
While male drivers do seem more prone to engage in certain types of stupid behavior, the more you drive the more likely you are to get into an accident.
I think instead of contacting your insurance company you should be contacting a divorce lawyer.
I suspect his wife must be both very hot and very good in bed.
Then again maybe not ...
Megers, feh! There are only stupid takeovers and product assimilation in the business world. Someone is either buying your top talent (never happens) or your top product (to shut it down.) If you couldn't hire away the top talent before, they won't stay now. Product closures is just capitalism in reverse. There is less competition and far too often the inferior product is the only product left in the market. People that think mergers are like happy marriges of two companies have never personally seen the look of pure greed on a pre-merger CEO's face.
I'm not bitter though, it's just that the idea of mergers and the reality of mergers have nothing to do with each other.
Sometimes mergers do work, take Apple and NeXT, Price and Costco, or any of the number of companies that Berkshire Hathaway has bought.
Then again the corprate cultures of the merging companies in the first two examples were very similar and Berkshire tends to not mess much with the companies it buys leaving exsisting management and employees in place.
Many customers claim that Comcast overall has been an improvement over AT&T Broadband, Verizon an improvement over GTE or Airtouch, or Wells Fargo an improvement over First Interstate. There is some thought that for many customers Cingular will be an improvement over AT&T Wireless (which has been going downhill ever since most of the old McCaw executives left a few years ago).
I think Wil explains it elsewhere in this thread.
Basicly it sounds like Wil and Mike wanted to pursue some things that the rest of Omni wasn't interested in so they spun off their own company.
I suspect that if Wil, Mike, and Tim wanted to return to Omni they would be welcomed back.
You have your history a bit wrong.
1 1e/case/starbucks.html
Peet's was first being founded by Alfred Peet in 1966.
Starbucks was founded by Jerry Baldwin (current CEO of Peet's), Gordon Bowker (also a co-founder of Red Hook Ale and The Seattle Weekly), and Zev Siegel who were all fans of Peet's coffee. Peet essentially taught them everything they knew about buying, roasting, brewing, and selling gourmet specialty coffee.
Howard Schultz didn't join the company until 1982 when he became the company's marketing manager (he esssentially lobbied Baldwin, Bowker, and Siegel for the job).
In 1984 the owners of Starbucks jumped at the chance to buy Peet's (Peet was retiring).
In 1985 Schultz left Starbucks to pursue his vision of Milanese style espresso bars by founding Il Giornale. He received considerable help from Baldwin and Bowker, both financial and in retailing and marketing advice.
In 1987 Baldwin and Bowker decided they wanted to sell Starbucks. Bowker wanted to focus on Red Hook and Baldwin wanted to focus on Peet's. Schultz made the winning bid for Starbucks and merged it with Il Giornale. The rest they say is history.
Schultz hired the founder of Seattle's Cafe Allegro to be Il Giornale's retail operations manager. He continues to serve as a Starbucks executive to this day.
Jim Stewart founded The Wet Whisker in 1970, which was later known as Stewart Brother's Coffee, then SBC, then as Seattle's Best Coffee. SBC's parent company Seattle Coffee Company was acquired by Starbucks in 2003. Stewart was also influenced and mentored by Alfred Peet.
Jerry Baldwin continues to serve as Director of Peet's Coffee & Tea. Baldwin's prior association with Starbucks probably explains some of the ill will between the two companies.
Links to more info:
http://www.mhhe.com/business/management/thompson/
http://www.mondaymemo.net/020923feature.htm
http://www.starbucks.com/aboutus/timeline.asp
Some independent coffee shops are very good though. Here in Seattle in addition to Zoka mentioned in the article we have Diva, Uptown, Vita, Allegro, among others. Any of them are a far cut above what Starbucks or the other chains put out.
The concern I have is: surely the next time you go for a piss someone will just rob your computer.
Fairly unlikely in that part of Seattle (not a lot of petty crime) and if someone did try it I suspect that co-workers, the other regulars, or the staff would try to stop the miscreant.
And how do you get on the Internet in a cafe?
Free wireless internet provided by the cafe.
It really is quite common here, not to mention all of the places that will provide wi-fi for a fee.
I haven't been to that location but it appears to be over in the U Village area near the Burke-Gillman trail.
Glad to see you still have your acid wit Wil.
I have rather fond memories of it back from the days of the UWbb.
Congradulations on Delicious Library, it looks like a nice program. Too bad I don't have a mac to run it on.
Shipley was a co-founder and former President of Omni I doubt a lack of control had anything to do with his leaving.
I suspect it had more to do with wanting to work on projects that the rest of Omni wasn't interested in pursuing plus an itch for the challenge of doing another startup.
I somehow doubt they'll have much trouble getting an office when and if the time comes for that.
Sorry I mean it was for saving the tech or for the fire department. Basicly any time you want to make sure everything is off RIGHT NOW.
Really, talk about the pot calling the kettle black.
/. be down for?
If the datacenter that hosts Slashdot was to have a massive power failure how long would
That said my company has gear in the same datacenter as LJ, our servers were back up 10 minutes after power was restored. Then again we use Oracle on HP-UX with nice SAN RAID boxes for storage for our database. So our stuff tends to recover from a sudden power loss a little better than a MySQL derivative running on clone hardware.
Whatever are you on about? InteNAP isn't even in the top 10 spam hosting ISP's. They don't tolerate spammers on their network as a general rule.
On the other hand they aren't going to boot a customer just because some crank complains about getting a commercial email from that customer. It does happen too. I've seen people sign up for a product newsletter. Confirm the subscription and then complain that the newsletter was spam to every single place they could possibly think of reporting it.
For contrast compare them to MCI, XO, or Above.net. All of those providers are far more spam friendly.
Datacenters have to have a BRS by the exit doors both in case some tech shorts himself across a power feed and so the fire department can kill all of the power to the room as killing the power feed to the building won't do it.
Well more to the point the problem LJ is having is due to a crappy database and not using decent storage systems.
We were able to bring our Oracle boxes back up quickly after power was restored.
That's debatable, and depending on what database you use, having more than one database server (or pool of database servers) in different physical locations that are kept in sync at all times is definitely possible. I'm not sure whether MySQL allows this, but I think if you have a site that has nearly 6 million users, more than 100000 of which are paying you for the service you provide (I'm one of those, one might add), then you really should look into doing just that - or at least I hope the LJ people will do now (I don't really want to blame them for the problem).
Even with Oracle or DB2 running database servers in multiple locations that need to stay in synch is a massive pain in the ass that you avoid if you can. Given that other problems are at least or more likely to cause your site to be unreachable to at least some customers for longer periods of time than problems with a decent co-location center it really isn't worth the bother for the majority of sites out there.
Hell last time I checked Slashdot still had all of their servers in a single datacenter.
For that matter I've seen times with my Verizon DSL line that I've been unable to reach Google, Yahoo, Amazon, Expedia or other internet sites with a high amount of redundancy including multiple datacenters for more than a day due to some problem on Verizon's network or with intermediate backbone.
But I don't blame say Google for the problems because they really have done everything they reasonably can to make sure their stuff is up and as many people as possible can reach their servers at any given point in time.
That being said, I think you didn't quite understand what I was trying to say. I really don't care whether they have "plenty of backup power", "plenty of generator capacity" and "top-of-the-line big datacenter grade stuff" (which really sounds more like a collection of buzzwords than anything else, anyway). If a wiring fault (of whatever kind) can bring up the entire UPS system as well as the "generator capacity behind that" and all other safeguards they supposedly had in place as well, then it's just worthless and a waste of money - a UPS is supposed to be an *uninterrupted* power supply.
By their nature multiplely redundant power systems are a tricky thing that even experts fail to get right. Even then someone hitting the BRS by the door that is required by fire code can still take power to the whole datacenter down. InterNAP is in the process of expanding the datacenter in question which means there are contractors running around all over the place. It is quite likely that someone miswired a new UPS. Shit happens, even to the datacenters of major banks, stock exchanges, or airline reservation systems.
And while I admit that it's not possible to guard against *all* problems, saying that the colo facility is "one of the most solid in the state" and supposedly can't be taken offline by something "short of a direct strike from a comet" is just silly when a "wiring failure" can bring down the whole thing, and even more so when it's not the first time that happens.
Really, this just stinks of an attitude that's all too prevalent in parts of the IT industry - just piecing together the components of a reliable system won't necessarily give you one, and if you can't build one properly, then don't go advertising that you have one. Don't you think the fact that the LJ people are now planning to buy their own UPS equipment to use on top of the facility's should tell you something?
The Fisher Plaza datacenter really is one of the best in the state. It is far from "pieced together". The InterNAP people aren't a bunch of clowns and do acutally know what they are doing (or at least as much as any other top-tier colocation vendor).
Especially in the cost vs. quality tradeoff InterNAP does really well which is why a fair number of companies around here buy rack space from them.
I do find it somewhat amusing that InterNAP's first datacenter, which really is
Maybe they just blew the rest of the year's allowable downtime.
I don't have our contract in front of me but I believe InterNAp just blew the next two years allowable unplanned downtime.
Fortunately for them their SLAs are only year to year or they would still be working off the time a customer hit the BRS thinking it would open the door.
The co-location facility in question has plenty of backup UPS power with plenty of generator capacity behind that. Supposedly there is enough generator capacity to fully power everything in the building including the network TV station even with one generator out.
The UPS gear in Internap's space is all top-of-the-line big datacenter grade stuff. Apparently there was some sort of wiring fault in one of the new UPSes they were bringing online that caused both building power to fail and the self-protection circuits in all of their UPSes to trip.
IOW it was either a faulty UPS or a faulty wiring job by the electrical contractor.
Livejournal isn't the only ones who got burned by this outage. The colocation facility in question is supposedly one of the most solid in the state and nothing short of a direct strike from a comet is supposed to be able to take it offline. My company was in the same boat as our gear is in the same facility as LiveJournal's.
Sure both LiveJournal and the company I work for could have hedged our bets by having redundant gear in another facility in another state, but that is a pain in the ass especially when backend databases are involved. To tell you the truth it probably isn't really worth the bother unless you truely have a need for six nines of uptime.
They helped kill two of my favorite local bookstores, leaving exactly none. I said "helped" because Barnes and Noble are comparably evil... but you asked. Also, please don't ask me about Starbucks!
That's funny I live in Seattle which is sort of ground-zero for both Amazon and Starbuck's (not to mention a certain large software monopoly). The thing is we still have plenty of non-chain bookstores here. If you ask the owners what the biggest threat to their business is they will usually mention the chain bookstores like B&N or Borders long before they mention Amazon.
As for Starbucks I really don't get the accusation that they kill off other coffee places. If anything Starbucks increases the local market for decent coffee by introducing people to it in the first place. In Seattle for example, there are several other medium-sized coffee chains that started in the shadow of Starbucks, along with several small local coffee chains and countless independant espresso bars and stands.
We saw the same problem a few years ago, in DB2 Support. Customer kept getting a system crash around the same time of day, nothing of consequence in either the db2diag.log or the system logs. The customer couldn't recreate, and traces were getting us nothing. Eventually, as the customer kept blaming us (of course) and went critsit, we asked him to physically monitor the system terminal to see what was going on. As he was on a conference call, a janitor came in and (IIRC, got this story second-hand) plugged his vacuum into the UPS circuit.
I've seen similar in person. At the university lab I used to spend way too much time in back in the late 80's and early 90's I watched a cleaner plug his vaccum into the same circuit as a bunch of the lab's X-terminals, macs, and NeXTs were on. There was a 'pop' sound and about 30 X terminals, 15 macs and 10 NeXTs all suddenly lost power. The resulting power surge killed a number of devices that happened to be on the same circuit along with many others that were coupled to them electricly via 10b2 cable outright. For at least a year afterward devices would die and when pulled apart would show evidence of arcing on the circuit board from a power surge.
Needless to say the university managed to rewire the lab areas in that building with outlets off of the big power conditioner in back for the datacenter gear in record time. Fortunately the janitors already knew better than to plug into the special orange outlets if they wanted to keep their jobs.