10% failure on powerdown? lets just hope you don't have to do that too often, and that the ups and generators are REALLY good!
To be honest - yes they are. Redundant electricity providers (seperate cables to seperate substations) and UPS / generators mean this really doesn't happen unless we want it to. Unfortunately, we do this from time to time essentially for upgrades of the SAN itself.:-/
Sorry - but this is the kind of thing that just doesn't stand up. If you replace the Windows boxes in an IT dept. you're gonna have to update Standard Operating Procedures, Business Continuity Plans and licence documents. I don't know why IT techys would be allowed to handle the MS licencing issues all by themselves - this is definitely in PHB territory. (OK - Mom's basement may not be such an exacting environment;-) ) And all of the time we're talking *infrastructure* services, Domain Controllers, File/Print servers etc - you certainly can't do this realistically without having to tell anyone on *application* servers. This kind of approach is typically the kind of thing that gets you hated when you leave a position - they suddenly discover that you have told them nothing and they are left second guessing a whole host of things that you might have set up. It's a nightmare- and just about the worst way, IMNSHO, to get any companies to get a handle on the real reasons why FOSS might be a good idea - you're going to leave them with an undocumented mess, which will cost them 10 times what "doing it properly" would have done, and the impression that FOSS is just a bunch of undisciplined geeks that can't behave themselves.
Agreed. We have around 50 TByte of data in one of our datacenters and it's great, but the number of disks that fail when you have to restart the systems (SAN fabric firmware install ) is just scary. Even on the system disks of the Wintel servers (around 400) which are DAS, around 10% fail on Datacenter powerdowns. That's where you pray that statistics are kind and you have no more failures on any one box than you have hot spares+tolerance:-) Last time one server didn't make it back up because of this.... though it was actually strictly speaking the PSUs that let go, it would appear.
I personally would be glad to be checked out by the police, because then I know they're doing their job. Privacy and all that stuff are important, but not more important than people's lives.
After just about every terrorist attack, this thinking is used to chip away at civil liberties. Governments convince the "General Public" that it's the price to pay to be safe. It started with the IRA attacks and continues. Are we safer? Are there no more terrorist attacks? No - we've given up civil liberties for the promise it would help stop this, but it doesn't - we still get attacked, we just are a little less free when we do....
My vote is for LDAP. You can do so much with it - authenticating users on your web apps is a cinch, directory lookups are easy, it integrates with every piece of mail client software, and it's free. Just my $.02.
$0.02? I think you're being generous with yourself:-) LDAP is of course the protocol, this is like saying "I'd recommend HTTP for web serving , and SMTP/IMAP for mail - they're great and FREE":-) There are many (very good) commercial LDAP offerings which are not free (as in beer or speech), though of course there are free directory servers that support it too.
Now what's not political about this whole thing?? The root cause is about the only thing that is no longer political. This sickens me, and I'll burn karma to prove it : the so-called "president" just doesn't care as much about thousands of americans dying as he does about being a "war president" (reference in Michael Moore's film Farenheit 9/11). Sorry folks, but this time you did vote for him, so I guess you're happy.
Completely agree. There's nothing about mainframes that decent graduate couldn't pick up with training. If companies can't find exactly the profiles they're after, they're going to have to broaden their horizons or outsource the support (to IBM, say) and make it the vendor's problem to get the staff. This always happens when tech specialists become hard to find in specific domains.
Methinks you've confused reinstalling (which takes time) to upgrading (which is quick and painless.)
Is it? Not always in my experience. Major distro updates are often quicker and easier with a reinstall, YMMV of course, but I've tried upgrading tens of PCs from SuSE 7.3 to 8.1 and I gave up and reinstalled after the third of fourth, just keeping the partition layout from the old install (homedirs were on the network). I'd need a lot of convincing that upgrading distros is easier and faster than reinstalling from a fresh MS Windows master install. Don't get me wrong - I think Windows is fairly braindead, but I just think the arguments in the post I replied to aren't the most important ones (and possibly aren't true) There are plenty of real reasons to question MS Windows utilisation without starting to save 60$ p.a. on resinstalls, because to be frank, as I said before, if that's why you're doing it, no-one will care enought to give you the resources to do it...
Only if you value your time at zero. Which is probably the case.
To be honest if an OS saves me an hour at install time, that's worth sixty bucks to me. If an OS is stable and doesn't need re-installing every year, that's sixty bucks every time.
This argument comes up regularly here, but I have 2 comments. You've just saved 60 bucks initial install and 60 bucks per year. The install fee and the 60 bucks per year per PC, unfortunately don't even hit the radar in terms of money saving on most IT budgets. 60 bucks per user is less than the electricity they use - it won't draw attention.
And to be honest, and probably far more significant, so many applications your users will want will be unsupported by the vendor or the FOSS developers (try getting current versions of most FOSS linux apps running on 8 year old distros....) on your legacy OS that you'll burn far more than 60 bucks having to support your apps yourself. You're more than welcome to run NT4 if you prefer, but if you're doing it for these reasons, I think you have the wrong reasons. Of course on one PC for grandma using apps Grandma knows and knowing she'll never want more recent apps, it works. Anywhere else it's just unrealistic, IMNSHO;-)
I don't think you comprehend what it means to be part of a very large organization.
LOL. We have 4 satellite receivers and an atomic clock on our private network. Trust me - I do. All I'm saying is that they should do something - either open the firewall OR multiple internal time sources (in LargeOrganisations(TM) and for SOX404 you need redundancy, and anyway from such a large network you probably don't want all your Tier 2 pointing at the same Tier 1). The Tier 1's ahould sync amongst themselves and with their attached clocks and use DNS to point clients at their nearest Tier 2.
From TFA : By making the rockets from shuttle parts, the new plan would draw on the shuttle's existing network of thousands of contractors and technologies, in theory speeding its completion and lowering its price.
Yes thereby ensuring that all the "keep prices down" corner cutting that got the shuttle where it is today doesn't go to waste either:-)
My boss is well aware of the problem and his boss is well aware of the problem but isn't in the position to do anything about it.:-(
Then go to someone who can. This is a no-brainer - it's probably worth speaking to your IT Security Head - any application logging is worthless unless you can rely on the timestamp. If your company is quoted on the NYSE - they'll have to answer questions like this soon anyway because of Sarbanes-Oxley etc.
Last night something went wrong temperature wise in my machine room.
One of the build alphas is now dead.
I think Theo should also ask for aircon. I'm willing to help but 1U boxes tend to get hot, and I see no point in all chipping in for a new Alpha box to see it go pop again in 2 weeks time. Theo, tell us what went wrong and what you've done to fix it or what we can do to help you fix it. Then we can worry about replacing the hardware - otherwise I think it's probably just as well to ask for Alpha hardware and rackspace in a reliable colo as send the hardware back to the same place.
Don't blame Itanium that you picked the wrong chip for your needs.
And if you're spending that kind of money without getting a benchmarking machine from the company you're buying it from before you sign - you deserve to get stung:-)
A little strange you are responsible for kernel dev. and "embedded distros" and struggle come up with an engineering report to decide on which dektop distro to use.
Choices such as these are ones companies make on a very regular basis without ever having to Ask Slashdot. Come up with your requirements, come up with a test plan for those requirements and tick the boxes. Once all the boxes are ticked (or not:-) ) pick the winner or, as some companies do, pick the one you wanted to pick anyway whether or not it won:-)
tell me why firing her saves their asses? If I were the owner I'd fire them regardless..
There are unspoken rules in this kind of thing. In my experience, the only risk of a high ranking exec losing his/her job over this is if there is someone at a similar or higher rank who had it in for them anyway.
How does this work? Someone calls a meeting to explain they've just lost 300 million. In most cases they will all agree that the most important thing to do is to find who seems to be directly responsible and fire them, and once that is done, come up with an "Action Plan" to avoid it happening again. The Action Plan bit is where they fix the real issue, and they will then congratulate themselves on having addressed the issue promptly and efficiently. The only time the question of why "promptly and efficiently" didn't mean before it happened is if the implementation was the work of a single person or small group of people that could also be made into blame carriers. The higher up the ladder those people are, the less likely there's someone higher to point it out, and so often it's the shop floor worker who pushed the light green button and not the dark green one, when in fact the buttons should have been RED and GREEN. This kind of thing happens at all levels of hierarchy, but of becomes more obvious once you reach the levels that can "fire at whim".
If the owners are stockholders, it's even simpler - they won't even care if you can present the bottom line performance in such a way that this doesn't even appear (it's probably a relatively small sum for such a place..) - or better still you can present the action plan to the stockholders in such a way as it will stop this kind of mistake and also make the company more reliable, efficient and therefore competitive and the stockholders won't even care who actually is responsible. SOX404, which has very much a pyramid of responsibilities for many things, including "operational risk control" (of which this is a case) may change this, but we'll have to wait and see. Right now all it's doing is costing companies a great deal of money, and is unfortunately too often being approached in a "cover your ass" manner, rather than as proper risk control.
She's the scape goat - if she doesn't get fired one of the higher execs will. I work in a large bank and the answer to this problem is business processes. The "head honchos" don't want to get fired for putting place an IT system and a business process that allows a single individual to do this kind of thing by mistake, and so are firing her to save their bacon.
I haven't seen any software system that can reliably scale to that level and still make any kind of sense for someone that wants to find a piece of data in that haystack,
Haven't you? Have you ever worked with real archiving before? IBM have some nice solutions that allow us to stock on disk and a WORM library (Tivoli Storage Manager) and index in a (large) Oracle DB - they work and scale just fine (our experience over a couple of hundred teras). You probably wouldn't want all that data in a single archive anyway, but i'd guess you'd know that if you'd ever archived anything....
There are very few cases where integrating a multi-threaded handler into a progrom doesn't introduce a formidable degree of complexity.
I would not be so categoric. It's a design issue - making a program that was designed from the ground up with single thread of "logic" play nicely with many different threads is stupidly complex and usually winds up being very kludgy - much of the threaded advantage is eaten away by the hacks that are needed to make it work. Design it from scratch to work this way, however, and the multi threading may not be simple, but it is at least "obvious", and that makes for good efficient threaded code. Lot's of tasks can be broken up quite easily and once the designer has understood inter-process communication and its constraints and overhead, the decision to create a new thread for a particular task or keep it in the exisiting one, is often far more straightforward than you make out, and yields good results.
And therefore it's vitally important that FIFTY PERCENT of everything is made in the US? This makes no sense whatsoever.
Agreed - I could come up with something that is 99% made in the US except for the chip that transmits keystrokes to North Korea. This is just an attempt to pass legislation that the WTO souldn't like by disguising it as security...
Strictly speaking, lawyers who are not barristers may be legal executives and not solicitors, if you see what I mean. All Lawyers who aren't barristers are not solicitors - though solicitors will try and tell you otherwise:-)
From the ILEX site : "Legal Executives are qualified lawyers specialising in a particular area of law"
Slashdot Mirror
10% failure on powerdown? lets just hope you don't have to do that too often, and that the ups and generators are REALLY good!
:-/
To be honest - yes they are. Redundant electricity providers (seperate cables to seperate substations) and UPS / generators mean this really doesn't happen unless we want it to. Unfortunately, we do this from time to time essentially for upgrades of the SAN itself.
Sorry - but this is the kind of thing that just doesn't stand up. If you replace the Windows boxes in an IT dept. you're gonna have to update Standard Operating Procedures, Business Continuity Plans and licence documents. I don't know why IT techys would be allowed to handle the MS licencing issues all by themselves - this is definitely in PHB territory. (OK - Mom's basement may not be such an exacting environment ;-) ) And all of the time we're talking *infrastructure* services, Domain Controllers, File/Print servers etc - you certainly can't do this realistically without having to tell anyone on *application* servers. This kind of approach is typically the kind of thing that gets you hated when you leave a position - they suddenly discover that you have told them nothing and they are left second guessing a whole host of things that you might have set up. It's a nightmare- and just about the worst way, IMNSHO, to get any companies to get a handle on the real reasons why FOSS might be a good idea - you're going to leave them with an undocumented mess, which will cost them 10 times what "doing it properly" would have done, and the impression that FOSS is just a bunch of undisciplined geeks that can't behave themselves.
Agreed. We have around 50 TByte of data in one of our datacenters and it's great, but the number of disks that fail when you have to restart the systems (SAN fabric firmware install ) is just scary. Even on the system disks of the Wintel servers (around 400) which are DAS, around 10% fail on Datacenter powerdowns. That's where you pray that statistics are kind and you have no more failures on any one box than you have hot spares+tolerance :-) Last time one server didn't make it back up because of this.... though it was actually strictly speaking the PSUs that let go, it would appear.
I personally would be glad to be checked out by the police, because then I know they're doing their job. Privacy and all that stuff are important, but not more important than people's lives.
After just about every terrorist attack, this thinking is used to chip away at civil liberties. Governments convince the "General Public" that it's the price to pay to be safe. It started with the IRA attacks and continues. Are we safer? Are there no more terrorist attacks? No - we've given up civil liberties for the promise it would help stop this, but it doesn't - we still get attacked, we just are a little less free when we do....
My vote is for LDAP. You can do so much with it - authenticating users on your web apps is a cinch, directory lookups are easy, it integrates with every piece of mail client software, and it's free. Just my $.02.
:-) :-) There are many (very good) commercial LDAP offerings which are not free (as in beer or speech), though of course there are free directory servers that support it too.
$0.02? I think you're being generous with yourself
LDAP is of course the protocol, this is like saying "I'd recommend HTTP for web serving , and SMTP/IMAP for mail - they're great and FREE"
I would hardly call BeOS "dead".
;-)
Yes, my Commodre64 still works. Doesn't stop it being dead
Yes No doubt he's right. Now CNN isn't normally a news source I'd quote but read :
THIS
Now what's not political about this whole thing?? The root cause is about the only thing that is no longer political. This sickens me, and I'll burn karma to prove it : the so-called "president" just doesn't care as much about thousands of americans dying as he does about being a "war president" (reference in Michael Moore's film Farenheit 9/11). Sorry folks, but this time you did vote for him, so I guess you're happy.
Completely agree. There's nothing about mainframes that decent graduate couldn't pick up with training. If companies can't find exactly the profiles they're after, they're going to have to broaden their horizons or outsource the support (to IBM, say) and make it the vendor's problem to get the staff. This always happens when tech specialists become hard to find in specific domains.
I like my books, I like hiding and having fun in the isles , don't take away my fun plllease.
:-)
You should try the aisle with the dictionaries in - it's a blast
Methinks you've confused reinstalling (which takes time) to upgrading (which is quick and painless.)
Is it? Not always in my experience. Major distro updates are often quicker and easier with a reinstall, YMMV of course, but I've tried upgrading tens of PCs from SuSE 7.3 to 8.1 and I gave up and reinstalled after the third of fourth, just keeping the partition layout from the old install (homedirs were on the network). I'd need a lot of convincing that upgrading distros is easier and faster than reinstalling from a fresh MS Windows master install. Don't get me wrong - I think Windows is fairly braindead, but I just think the arguments in the post I replied to aren't the most important ones (and possibly aren't true) There are plenty of real reasons to question MS Windows utilisation without starting to save 60$ p.a. on resinstalls, because to be frank, as I said before, if that's why you're doing it, no-one will care enought to give you the resources to do it...
Only if you value your time at zero. Which is probably the case. To be honest if an OS saves me an hour at install time, that's worth sixty bucks to me. If an OS is stable and doesn't need re-installing every year, that's sixty bucks every time.
;-)
This argument comes up regularly here, but I have 2 comments. You've just saved 60 bucks initial install and 60 bucks per year. The install fee and the 60 bucks per year per PC, unfortunately don't even hit the radar in terms of money saving on most IT budgets. 60 bucks per user is less than the electricity they use - it won't draw attention.
And to be honest, and probably far more significant, so many applications your users will want will be unsupported by the vendor or the FOSS developers (try getting current versions of most FOSS linux apps running on 8 year old distros....) on your legacy OS that you'll burn far more than 60 bucks having to support your apps yourself. You're more than welcome to run NT4 if you prefer, but if you're doing it for these reasons, I think you have the wrong reasons. Of course on one PC for grandma using apps Grandma knows and knowing she'll never want more recent apps, it works. Anywhere else it's just unrealistic, IMNSHO
I don't think you comprehend what it means to be part of a very large organization.
LOL. We have 4 satellite receivers and an atomic clock on our private network. Trust me - I do. All I'm saying is that they should do something - either open the firewall OR multiple internal time sources (in LargeOrganisations(TM) and for SOX404 you need redundancy, and anyway from such a large network you probably don't want all your Tier 2 pointing at the same Tier 1). The Tier 1's ahould sync amongst themselves and with their attached clocks and use DNS to point clients at their nearest Tier 2.
From TFA : By making the rockets from shuttle parts, the new plan would draw on the shuttle's existing network of thousands of contractors and technologies, in theory speeding its completion and lowering its price.
:-)
Yes thereby ensuring that all the "keep prices down" corner cutting that got the shuttle where it is today doesn't go to waste either
My boss is well aware of the problem and his boss is well aware of the problem but isn't in the position to do anything about it. :-(
Then go to someone who can. This is a no-brainer - it's probably worth speaking to your IT Security Head - any application logging is worthless unless you can rely on the timestamp. If your company is quoted on the NYSE - they'll have to answer questions like this soon anyway because of Sarbanes-Oxley etc.
Last night something went wrong temperature wise in my machine room.
One of the build alphas is now dead.
I think Theo should also ask for aircon. I'm willing to help but 1U boxes tend to get hot, and I see no point in all chipping in for a new Alpha box to see it go pop again in 2 weeks time. Theo, tell us what went wrong and what you've done to fix it or what we can do to help you fix it. Then we can worry about replacing the hardware - otherwise I think it's probably just as well to ask for Alpha hardware and rackspace in a reliable colo as send the hardware back to the same place.
This is Microsoft Tuesday
:-)
Perhaps they should make that Microsoft Tuesday (TM) like Microsoft Windows (TM), Microsoft Office (TM) etc
Don't blame Itanium that you picked the wrong chip for your needs.
:-)
And if you're spending that kind of money without getting a benchmarking machine from the company you're buying it from before you sign - you deserve to get stung
A little strange you are responsible for kernel dev. and "embedded distros" and struggle come up with an engineering report to decide on which dektop distro to use.
:-) ) pick the winner or, as some companies do, pick the one you wanted to pick anyway whether or not it won :-)
Choices such as these are ones companies make on a very regular basis without ever having to Ask Slashdot. Come up with your requirements, come up with a test plan for those requirements and tick the boxes. Once all the boxes are ticked (or not
tell me why firing her saves their asses? If I were the owner I'd fire them regardless..
There are unspoken rules in this kind of thing. In my experience, the only risk of a high ranking exec losing his/her job over this is if there is someone at a similar or higher rank who had it in for them anyway.
How does this work? Someone calls a meeting to explain they've just lost 300 million. In most cases they will all agree that the most important thing to do is to find who seems to be directly responsible and fire them, and once that is done, come up with an "Action Plan" to avoid it happening again. The Action Plan bit is where they fix the real issue, and they will then congratulate themselves on having addressed the issue promptly and efficiently. The only time the question of why "promptly and efficiently" didn't mean before it happened is if the implementation was the work of a single person or small group of people that could also be made into blame carriers. The higher up the ladder those people are, the less likely there's someone higher to point it out, and so often it's the shop floor worker who pushed the light green button and not the dark green one, when in fact the buttons should have been RED and GREEN. This kind of thing happens at all levels of hierarchy, but of becomes more obvious once you reach the levels that can "fire at whim".
If the owners are stockholders, it's even simpler - they won't even care if you can present the bottom line performance in such a way that this doesn't even appear (it's probably a relatively small sum for such a place..) - or better still you can present the action plan to the stockholders in such a way as it will stop this kind of mistake and also make the company more reliable, efficient and therefore competitive and the stockholders won't even care who actually is responsible. SOX404, which has very much a pyramid of responsibilities for many things, including "operational risk control" (of which this is a case) may change this, but we'll have to wait and see. Right now all it's doing is costing companies a great deal of money, and is unfortunately too often being approached in a "cover your ass" manner, rather than as proper risk control.
"Why is she getting fired and not trained?"
She's the scape goat - if she doesn't get fired one of the higher execs will. I work in a large bank and the answer to this problem is business processes. The "head honchos" don't want to get fired for putting place an IT system and a business process that allows a single individual to do this kind of thing by mistake, and so are firing her to save their bacon.
I haven't seen any software system that can reliably scale to that level and still make any kind of sense for someone that wants to find a piece of data in that haystack,
Haven't you? Have you ever worked with real archiving before? IBM have some nice solutions that allow us to stock on disk and a WORM library (Tivoli Storage Manager) and index in a (large) Oracle DB - they work and scale just fine (our experience over a couple of hundred teras). You probably wouldn't want all that data in a single archive anyway, but i'd guess you'd know that if you'd ever archived anything....
There are very few cases where integrating a multi-threaded handler into a progrom doesn't introduce a formidable degree of complexity.
I would not be so categoric. It's a design issue - making a program that was designed from the ground up with single thread of "logic" play nicely with many different threads is stupidly complex and usually winds up being very kludgy - much of the threaded advantage is eaten away by the hacks that are needed to make it work. Design it from scratch to work this way, however, and the multi threading may not be simple, but it is at least "obvious", and that makes for good efficient threaded code. Lot's of tasks can be broken up quite easily and once the designer has understood inter-process communication and its constraints and overhead, the decision to create a new thread for a particular task or keep it in the exisiting one, is often far more straightforward than you make out, and yields good results.
And therefore it's vitally important that FIFTY PERCENT of everything is made in the US? This makes no sense whatsoever.
Agreed - I could come up with something that is 99% made in the US except for the chip that transmits keystrokes to North Korea. This is just an attempt to pass legislation that the WTO souldn't like by disguising it as security...
At least this (even the bit about Safari feeling faster due to caching) sounds more plausible than a Pentium making teh intarweb go faster :-)
Strictly speaking, lawyers who are not barristers may be legal executives and not solicitors, if you see what I mean. All Lawyers who aren't barristers are not solicitors - though solicitors will try and tell you otherwise :-)
From the ILEX site :
"Legal Executives are qualified lawyers specialising in a particular area of law"