How did you manage to keep the spyware out? Zero (or highly restricted) network access?
I can see a good combination of keeping on top of virus defs and keeping a tight firewall stopping most or all of the virii, but the spyware is such a constantly evolving battle waged by each of your 1000 users that invite it into their computer, I don't see how you could possibly have had even a couple PCs with browsing access and no spyware on them, for even a couple weeks let alone 11 months.
Most spam engines do direct sending, because if they relay through an ISP's email server, that's a single server seeing thousands of outbounds from a single IP. And since that's where the point of control is, that can be easily detected and blocked. (tho admittedly that often doesn't happen, due to stupid ISPs) If you allow outgoing through 25, then the bot can just deliver mail directly to the victims' mailservers, a random spray of IPs. Harder for the ISP on either end to detect, and very difficult for the receiving mailserver to filter out.
Though I would mind the inconvenience of getting my mailserver blocked, I am fairly sure that I could straighten that out with one phonecall. (helps to have a good ISP)
Knowing the address of a (little known) open relay is handy for just such an emergency. Though finding an open relay has gotten a lot harder, and they usually disappear (spammied into oblivion) within a month or so.
I run a mailserver for my friends and family, and it's relay is authtenticated, to keep out the spammers. I've found my mailserver on more than one blacklist in the past, but so far it's been a case of a misconfiguration on MY part (no reverse DNS, oopsie!) and another for being in a DSL pool. (business class DSL mind you, a block of static IPs) Got off the lists easily enough, haven't had any problems since. The blacklist system seems to work well.
Flash based memory isn't faster than hard drives, it's much slower. There are different grades of flash memory with different speeds. (just shop on the net for CompactFlash cards, notice the huge price and speed differences) Even the fastet flash usb drive will only achieve 2mb (read) continuous in any real world test, and they are always slower on write. A 4200rpm (slow!) HD will easily run rings around any flash drive.
Power, heat savings, and shock resistance are the biggest advantages they have going for them now.
They don't have to do a random search actualy. As part of the BT protocol, when a peer joins a network they receive a peer list - the IP addresses of some of the peers in the swarm. (normally not all the peers, to prevent trackers from getting swamped by large swarms)
The new peer then attempts to make a connection with some or all of the peers in the list it received. Once communication is established between the new peer and one of the peers, both exchange a list of which pieces of the torrent they already posess. Anyone with 100% of the pieces is a "seed", and is a good candidate for being the original poster. This I know from having read one of the earlier drafts of the BT protocol.
Looking at current clients such as Bits on Wheels, I also am assuming that new information has been added to the exchange, and that peers also swap information such as how much total data they have upstreamed and downstreamed since they connected into the swarm. In a swarm with several seeds, this information might be used to determine which seed is the original seed, assuming they will have a total of 0 bytes downstreamed and most likely one of the largest byte totals upstreamed.
It's sort of like the mosquito problem... spray a lake with oil and you wipe out the mosquitoes in an area for awhile. Take away the lake, and now all you can do is swat them one by one... far less effective, and I think that's what the new BT is aiming for.
I also wonder where the line is drawn in copyright. I can take a sentence from a book and that's most likely not violating copyright. I can copy 5 notes from a song and that's not violating copyright. But if I take two chapters from a book, or a dozen notes from a song, THAT can violate copyright. Where do they draw the line for digital works? And what if a torrent's "piece size" is smaler than this amount? Technically you couldn't violate copyright by sending out five bytes of data, because that would be a common string that could occur in any number of things. Maybe it's arguing semantics, but that's what it always comes down to for the lawyers, because they have to draw the line in stone somewhere and it's always possible to be just on the other side of that line.
I work at a mac service center, and about 1/4 of the machines I see from day to day have netscape on them. People do use it, some prefer it. (it seems to be the browser of choice when compared to IE) Netscape is missing out on a market by not supporting macintosh for version 8, because there's a ton of mac users out there running 7.02 and looking for the next release.
It would seem like a futile thing to try to study older versions of your product - why aren't they just focusing on the most up-to-date patched versions of their various products? Anyone running an unpatched windows box is insane...
Macs do draw that low of power. My powerbook g4 draws at most 28 watts when charging the battery and playing UT 2004 at the same time. (graphics and CPU maxed out) It draws 18 watts under normal use, and 2 watts asleep OR off when not charging batteries.
Not sure why his unit only lasted 2 hrs though. The author mentioned the computer will run off from 12-20v, and it seems like he had enough cells to keep the voltage up until they wound down all the way. G4 processors are good at conserving energy, and run cool and lean when they're not busy doing things. (I never understood why do PCs run full tilt all the time?) It's possible he made his measurements with the machine idle, and then in the actual test was actively using the machine, increasing its power consumption. I can't fault your math, but 40w seems excessive for a mini. The mini's pack is rated 85w, but that's to power USB and firewire too.
From a liability standpoint, you might want to tell them about the back doors when you leave. Elsewise, if something blows up a week after you're gone, and they bring in someone with a cluebat and find your back doors, you will be Prime Suspect to the cause of the blowup, even if it was caused by someone else or simple system failure.
Bonus points for telling -everyone- on the I.T. staff about the back doors when you leave, rather than just your boss. In the event something does blow up a week later, they have a room full of suspects to cause internal friction, ("you didn't play with Bob's back doors did you?") and you look most innocent as the open source of this important information. (if you intended to use them, you would not have told anyone about them)
You must assume that the box is out of your control when an admin leaves.
This is theoretically good advice, but in reality recertifying a box as "secure" is a lot harder than changing the keys on a few doors. A system of significant complexity may be next to impossible (or at least, very expensive) to completely audit. When it comes down to it, the company has to rely to some degree on the good integrity of the ex-employee. This puts the company AND the employee in a disadvantageous position because the company fears retribution by the employee so they feel compelled to pink slip you at the door to protect themselves, which ends up shafting the employee. (which could arguably push an ex-employee "over the edge" into using back doors for retribution) So really in this game, everyone loses.
Oh it happens. Happened where I used to work - a new member of the staff (placed in a position of authority) found he had created an intensely hostile work environment. (by pissing off everyone in the building) He quit. I pointed out to our manager (more than once) that we needed to change passwords. "Oh, you don't need to worry about that." was the reply.
One morning two weeks later the supervisor passwords on all our novell servers suddenly stopped working. Cute trick. We had to hack our own servers to get back in, at all eight locations. Fortunately, only the supervisor accounts he knew about had been changed, which made getting passwords reset much easier because we had a few "service" accounts for our paid support people with supervisor privs.
Only after that did our I.T. manager agree we needed to change supervisor and dial-in passwords. *sigh*
It doesn't matter who it is that leaves/quits/fired/whatever, if they had access to passwords, those need to get changed, immediately. Just because a person held a position of authority does not mean there is any reason to trust them with company property after they are gone. Looking back on it, I forgive him for doing it, it was our fault and we got what we deserved by treating security so foolishly. The lesson could have been much more painful.
I'm against the concept of giving someone the boot without warning though. Funny how companies expect a 2 week (or longer) notice when you're going to cut out, but are perfectly ok with taking your badge at the front door when you come in on a Monday morning. Whenever an employer asks me how much notice I'll give them when I'm headed out, I always say "I'd never give you any less notice than I expect to receive from you." They smile, then they frown. They know how the game works.
If someone's got it in mind to sabotage the works before they leave, odds are good that they will smell the pink slip before it's handed out, and have ample opportunity to muck with things.
I've seen an ibook come in with the keyboard saturated with beer. Funny part about that was the customer had no idea what was wrong with it. (his roommate had tipped a can of bud into it the previous evening) But yes, the musk of beer on keyboard definitely is beaten hands down by chocolate cow. Not only does it jam up the keys, but it reeks to high heaven for a few weeks. That was on a pro keyboard... they asked me if it could be fixed. I said yep, we have new ones right on the shelf over there.
I've heard two reports of possibly worse though, thankfully experienced and repaired by other people - one had his cat piss on his ibook's keyboard, another fellow had a drunk visitor vomit on his powerbook. ewwwww
I'm not convinced that's a safer way to dispose of lithium... recall back a couple months the slashdot story about the guy that took a chunk of Li and dropped it into a lake via remote control - it spent the next couple minutes falling into the water, blasting itself OUT of the water, flying 40 feet, and repeating over again.
The reason Li batteries will "flame out" when ruptured is because the Li is reacting with the water (humidity) in the air. Surrounding Li with liquid water is extremely dangerous. The Li rips the oxygen off the hydrogen and gets very hot in the process, enough to say, ignite flamible gasses. Now, that hydrogen rises to the surface of the water and carries the lithium along for the ride, exposing a flammible gas (hydrogen) to oxygen, in the presense of extreme heat. Is anyone surprised we get a fireball?
I'm surprised the isp isn't doing the usual maneuver and crying us rivers for being abused. Kudos to them for sucking it up, taking their lumps, and attempting to do something about the problem they fostered instead of whining to us about it not being their problem.
If I were on their service I'd be a little peeved for having my email blacklisted, but I would know the correcct organization to direct my ill-content to: not the blacklister, but to the isp. Hopefully they can get a good percentage of their customers' owned machines back under control and get delisted soon.
Any isp that does not have detection systems to identify and either cap or disconnect machines that are clearly spam engines is doing the internet a dis-service, and as a result their customers are getting the shaft by blacklist. Best way to look at it is the ISP should be providing the "service" of not getting their customers blacklisted periodically by catching and dealing with clueless customers with owned machines. Customers going on the cheap with their ISPs risk not receiving this oft-overlooked "service" and finding their email roundfiled.
... by giving them any notice. Ask yourself, how much notice could you count on if they were going to let you go? I've found out first-hand on three separate occasions, that number is zero.
Assuming you have a position of functional importance (and uniqueness even) in your company... if your boss had his wits about him, (and it doesn't look that way from your description) he would immediately pull someone from somewhere else for you to go on full-time training. You wouldn't be doing any of your normal work, which would give management a good view of what things you do now that nobody else knows how to do right, so they can squeeze this information from you before you depart. You should be spending the next 2 weeks answering questions and walking someone through doing your day-to-day work.
Given that this is not happening, do not be surprised if the boss tries guilt, sense of duty, obligation, or fraud (withholding of pay) after you leave or are about to leave, in an attempt to get you to stay longer, answer phonecalls at all times of the day, night, or weekend, or just plain do free consulting for them. Dont' stand for it. If they want to be calling you and asking questions or have you come in 'for a few minutes" to show them how to do something, establish a firm rate, ($50/hr is a generous consulting rate) paid in advance. Don't let them take advantage of you.
If you're thinking about cutting out early, at least try to weigh in your impact on the other employees... you probably know what life will be like for them for the next 2-4 weeks as things are trying to get settled back down. Don't let one manager ruin your day AND theirs. The manager might have it coming to him, but he'll just spread the grief around as most PHB's tend to do.
It seems that in your area it's not possible to exercise effective control over the individual computers. You might be able to mitigate the damage caused by outbreaks by purchasing some managed switches to replace your dumb switches currently in use. Some switches can cap bandwidth, or disconnnect/throttle ports if they exceed usage limits for a set amount of time. Most of them are smart enough that you can specify the direction that the triggers are placed, so you can tell the switch basically, "if any user on this switch exceeds 100k/sec upstream for 5 seconds, cap their upstream to 10k/sec for the next five minutes." They also usually let you monitor current bandwidth usage, and some provide short term history for bandwidth or violations, either of which makes it fairly easy to spot owned machines. You can always just shut their port off remotely until the owner is so motivated by lack of internet connection that they clean up their machine. It's not a cure, but it can dull the symptoms. (it also cuts down on the foot traffic required by you)
The only major drawback is price. Managed switches can be expensive, and it sounds like you'll need a number of them. Best way to get funding for something like this approved is to throw together a report outlining how many man hours and how much network downtime this will save over say, the next 3 years. When they see the dollar difference and performance paybacks, they'll get out the checkbook. You can also roll it out incrementally, by getting a fast switch that's managed, as your main node, that initially feeds the existing dumb switches at each of the buildings. Watch how fast people fix their machines when they have like 11 other tennants on their back because everyone's internet in the building is down. Then take the noise from the other 11 tennants that's pointed to you and redirect that into a requisition for a managed switch in that building, killing several birds with one stone. By that route you'l eventually get your entire network to managed switches and your problem will be much easier to deal with. Happy managers (no complaining tennants), happy tennants (no downtime), and a few pissed of ppl with owned machines. Justice all-around.
I didn't see any reference to the actual posted speed limit in the article (might have missed it tho) but it said the estimated speed of the car was 50km/hr. That's clipping along pretty good for a sharply curved on-ramp, and the officers said they felt that the car was moving too fast to negotiate the onramp. The officers also said the car was going faster than any other car they'd seen that evening. Whether they were speeding or not, hard to say. Does sound like they were driving faster than was safe, and that's something you're more likely to see in a car-bomber than grandma getting groceries.
You also mention, Since everyone knows that US troops follow up a warning shot by spraying cars with hundreds of rounds in a panic, the driver sped up. but if you re-read the report, you'll see that the team encountered between 15 and 30 vehicles prior to this one, all of which reacted correctly and stopped their vehicle. Fortunately the people that live in a war zone react more sensibly to an armed roadblock than the "everyone" you are referring to.;-)
The driver screwed up, big time, and his passenger paid the price. There were mechanisms that should have made it possible to cover for his mistake, but they failed. That doesn't make it the soldiers' fault that the guy made a big mistake, it just means they failed to go the extra mile to compensate for a fool's behavior. You can't count on the soldiers to give you a wide margin of error in a warzone, because for them to give anyone a wide margin is suicide.
The 1-76 TOC had two means of communicating with 4th Brigade, its higher headquarters: Voice Over Internet Protocol (VOIP)2 and FM. The 1-76 FA Battle Captain was using only VOIP to communicate with 1-69 IN, but experienced problems with VOIP, therefore losing its only communication link with 1-69 IN, other than going through 4th Brigade. (Annex 97C). As a result, the Battle Captain was unable to pass updated information about the blocking mission either directly to 1-69 IN, or to 4th Brigade. He did not attempt to contact 4th Brigade via FM communications. (Annex 63C).
and
(U) Mr. Carpani told Sergeant First Class Feliciano who Ms. Sgrena was and that he was trying to get to the airport. He told Sergeant First Class Feliciano that he heard shots from somewhere, and that he panicked and started speeding, trying to get to the airport as quickly as possible. Mr. Carpani further told Sergeant First Class Feliciano that he continued to speed down the ramp, and that he was in a hurry to get to the airport. (Annexes 91C, 136C).
So it all came down to two issues.
(1) Failure to communicate. The car wasn't where it should be, wasn't informed of what was waiting ahead of them, gave its position but that information was not forwarded to the roadblock, so they were not expecting them.
(2) The driver then risked the lives of everyone in the vehicle by reacting with very bad judgement when he arrived at the roadblock. (accelerating the vehicle after he was spotlighted, laserpointered, and heard the warning shots)
Bad decisions by the driver of the vehicle, amplified by failure to communicate.
If you're behind a firewall with a proxy, you're going nowhere until you at least find the proxy. From there there's no guarantees of course, but it's a necessary first step. (worry about the key after you find the door)
establish a connection to something... anything that's persistent. (download a big file on www?)
then on the command line:
netstat -n | grep "ESTAB"
That'll kick out a list of established network connections, the IP they go to, and the port they're going out on. That should help you find your proxy server's IP and port number.
Slashdot Mirror
How did you manage to keep the spyware out? Zero (or highly restricted) network access?
I can see a good combination of keeping on top of virus defs and keeping a tight firewall stopping most or all of the virii, but the spyware is such a constantly evolving battle waged by each of your 1000 users that invite it into their computer, I don't see how you could possibly have had even a couple PCs with browsing access and no spyware on them, for even a couple weeks let alone 11 months.
Most spam engines do direct sending, because if they relay through an ISP's email server, that's a single server seeing thousands of outbounds from a single IP. And since that's where the point of control is, that can be easily detected and blocked. (tho admittedly that often doesn't happen, due to stupid ISPs) If you allow outgoing through 25, then the bot can just deliver mail directly to the victims' mailservers, a random spray of IPs. Harder for the ISP on either end to detect, and very difficult for the receiving mailserver to filter out.
Though I would mind the inconvenience of getting my mailserver blocked, I am fairly sure that I could straighten that out with one phonecall. (helps to have a good ISP)
Knowing the address of a (little known) open relay is handy for just such an emergency. Though finding an open relay has gotten a lot harder, and they usually disappear (spammied into oblivion) within a month or so.
I run a mailserver for my friends and family, and it's relay is authtenticated, to keep out the spammers. I've found my mailserver on more than one blacklist in the past, but so far it's been a case of a misconfiguration on MY part (no reverse DNS, oopsie!) and another for being in a DSL pool. (business class DSL mind you, a block of static IPs) Got off the lists easily enough, haven't had any problems since. The blacklist system seems to work well.
Tho the real thing to ponder is how someone with a UID so large got a nickname so small
Flash based memory isn't faster than hard drives, it's much slower. There are different grades of flash memory with different speeds. (just shop on the net for CompactFlash cards, notice the huge price and speed differences) Even the fastet flash usb drive will only achieve 2mb (read) continuous in any real world test, and they are always slower on write. A 4200rpm (slow!) HD will easily run rings around any flash drive.
Power, heat savings, and shock resistance are the biggest advantages they have going for them now.
Price, capacity, and speed are the big drawbacks.
anyone who's watched farscape already knows this...
They don't have to do a random search actualy. As part of the BT protocol, when a peer joins a network they receive a peer list - the IP addresses of some of the peers in the swarm. (normally not all the peers, to prevent trackers from getting swamped by large swarms)
The new peer then attempts to make a connection with some or all of the peers in the list it received. Once communication is established between the new peer and one of the peers, both exchange a list of which pieces of the torrent they already posess. Anyone with 100% of the pieces is a "seed", and is a good candidate for being the original poster. This I know from having read one of the earlier drafts of the BT protocol.
Looking at current clients such as Bits on Wheels, I also am assuming that new information has been added to the exchange, and that peers also swap information such as how much total data they have upstreamed and downstreamed since they connected into the swarm. In a swarm with several seeds, this information might be used to determine which seed is the original seed, assuming they will have a total of 0 bytes downstreamed and most likely one of the largest byte totals upstreamed.
It's sort of like the mosquito problem... spray a lake with oil and you wipe out the mosquitoes in an area for awhile. Take away the lake, and now all you can do is swat them one by one... far less effective, and I think that's what the new BT is aiming for.
I also wonder where the line is drawn in copyright. I can take a sentence from a book and that's most likely not violating copyright. I can copy 5 notes from a song and that's not violating copyright. But if I take two chapters from a book, or a dozen notes from a song, THAT can violate copyright. Where do they draw the line for digital works? And what if a torrent's "piece size" is smaler than this amount? Technically you couldn't violate copyright by sending out five bytes of data, because that would be a common string that could occur in any number of things. Maybe it's arguing semantics, but that's what it always comes down to for the lawyers, because they have to draw the line in stone somewhere and it's always possible to be just on the other side of that line.
I work at a mac service center, and about 1/4 of the machines I see from day to day have netscape on them. People do use it, some prefer it. (it seems to be the browser of choice when compared to IE) Netscape is missing out on a market by not supporting macintosh for version 8, because there's a ton of mac users out there running 7.02 and looking for the next release.
It would seem like a futile thing to try to study older versions of your product - why aren't they just focusing on the most up-to-date patched versions of their various products? Anyone running an unpatched windows box is insane...
Macs do draw that low of power. My powerbook g4 draws at most 28 watts when charging the battery and playing UT 2004 at the same time. (graphics and CPU maxed out) It draws 18 watts under normal use, and 2 watts asleep OR off when not charging batteries.
Not sure why his unit only lasted 2 hrs though. The author mentioned the computer will run off from 12-20v, and it seems like he had enough cells to keep the voltage up until they wound down all the way. G4 processors are good at conserving energy, and run cool and lean when they're not busy doing things. (I never understood why do PCs run full tilt all the time?) It's possible he made his measurements with the machine idle, and then in the actual test was actively using the machine, increasing its power consumption. I can't fault your math, but 40w seems excessive for a mini. The mini's pack is rated 85w, but that's to power USB and firewire too.
From a liability standpoint, you might want to tell them about the back doors when you leave. Elsewise, if something blows up a week after you're gone, and they bring in someone with a cluebat and find your back doors, you will be Prime Suspect to the cause of the blowup, even if it was caused by someone else or simple system failure.
Bonus points for telling -everyone- on the I.T. staff about the back doors when you leave, rather than just your boss. In the event something does blow up a week later, they have a room full of suspects to cause internal friction, ("you didn't play with Bob's back doors did you?") and you look most innocent as the open source of this important information. (if you intended to use them, you would not have told anyone about them)
You must assume that the box is out of your control when an admin leaves.
This is theoretically good advice, but in reality recertifying a box as "secure" is a lot harder than changing the keys on a few doors. A system of significant complexity may be next to impossible (or at least, very expensive) to completely audit. When it comes down to it, the company has to rely to some degree on the good integrity of the ex-employee. This puts the company AND the employee in a disadvantageous position because the company fears retribution by the employee so they feel compelled to pink slip you at the door to protect themselves, which ends up shafting the employee. (which could arguably push an ex-employee "over the edge" into using back doors for retribution) So really in this game, everyone loses.
Oh it happens. Happened where I used to work - a new member of the staff (placed in a position of authority) found he had created an intensely hostile work environment. (by pissing off everyone in the building) He quit. I pointed out to our manager (more than once) that we needed to change passwords. "Oh, you don't need to worry about that." was the reply.
One morning two weeks later the supervisor passwords on all our novell servers suddenly stopped working. Cute trick. We had to hack our own servers to get back in, at all eight locations. Fortunately, only the supervisor accounts he knew about had been changed, which made getting passwords reset much easier because we had a few "service" accounts for our paid support people with supervisor privs.
Only after that did our I.T. manager agree we needed to change supervisor and dial-in passwords. *sigh*
It doesn't matter who it is that leaves/quits/fired/whatever, if they had access to passwords, those need to get changed, immediately. Just because a person held a position of authority does not mean there is any reason to trust them with company property after they are gone. Looking back on it, I forgive him for doing it, it was our fault and we got what we deserved by treating security so foolishly. The lesson could have been much more painful.
I'm against the concept of giving someone the boot without warning though. Funny how companies expect a 2 week (or longer) notice when you're going to cut out, but are perfectly ok with taking your badge at the front door when you come in on a Monday morning. Whenever an employer asks me how much notice I'll give them when I'm headed out, I always say "I'd never give you any less notice than I expect to receive from you." They smile, then they frown. They know how the game works.
If someone's got it in mind to sabotage the works before they leave, odds are good that they will smell the pink slip before it's handed out, and have ample opportunity to muck with things.
Do they even make 7200rpm 2.5" HDDs? I've heard of the 4200's and the 5400's, but only seen 7200 and 10000 rpm ratings on 3.5" drives.
I've seen an ibook come in with the keyboard saturated with beer. Funny part about that was the customer had no idea what was wrong with it. (his roommate had tipped a can of bud into it the previous evening) But yes, the musk of beer on keyboard definitely is beaten hands down by chocolate cow. Not only does it jam up the keys, but it reeks to high heaven for a few weeks. That was on a pro keyboard... they asked me if it could be fixed. I said yep, we have new ones right on the shelf over there.
I've heard two reports of possibly worse though, thankfully experienced and repaired by other people - one had his cat piss on his ibook's keyboard, another fellow had a drunk visitor vomit on his powerbook. ewwwww
I'm not convinced that's a safer way to dispose of lithium... recall back a couple months the slashdot story about the guy that took a chunk of Li and dropped it into a lake via remote control - it spent the next couple minutes falling into the water, blasting itself OUT of the water, flying 40 feet, and repeating over again.
The reason Li batteries will "flame out" when ruptured is because the Li is reacting with the water (humidity) in the air. Surrounding Li with liquid water is extremely dangerous. The Li rips the oxygen off the hydrogen and gets very hot in the process, enough to say, ignite flamible gasses. Now, that hydrogen rises to the surface of the water and carries the lithium along for the ride, exposing a flammible gas (hydrogen) to oxygen, in the presense of extreme heat. Is anyone surprised we get a fireball?
It doesn't do any good unless you seed it.
I'm surprised the isp isn't doing the usual maneuver and crying us rivers for being abused. Kudos to them for sucking it up, taking their lumps, and attempting to do something about the problem they fostered instead of whining to us about it not being their problem.
If I were on their service I'd be a little peeved for having my email blacklisted, but I would know the correcct organization to direct my ill-content to: not the blacklister, but to the isp. Hopefully they can get a good percentage of their customers' owned machines back under control and get delisted soon.
Any isp that does not have detection systems to identify and either cap or disconnect machines that are clearly spam engines is doing the internet a dis-service, and as a result their customers are getting the shaft by blacklist. Best way to look at it is the ISP should be providing the "service" of not getting their customers blacklisted periodically by catching and dealing with clueless customers with owned machines. Customers going on the cheap with their ISPs risk not receiving this oft-overlooked "service" and finding their email roundfiled.
... by giving them any notice. Ask yourself, how much notice could you count on if they were going to let you go? I've found out first-hand on three separate occasions, that number is zero.
Assuming you have a position of functional importance (and uniqueness even) in your company... if your boss had his wits about him, (and it doesn't look that way from your description) he would immediately pull someone from somewhere else for you to go on full-time training. You wouldn't be doing any of your normal work, which would give management a good view of what things you do now that nobody else knows how to do right, so they can squeeze this information from you before you depart. You should be spending the next 2 weeks answering questions and walking someone through doing your day-to-day work.
Given that this is not happening, do not be surprised if the boss tries guilt, sense of duty, obligation, or fraud (withholding of pay) after you leave or are about to leave, in an attempt to get you to stay longer, answer phonecalls at all times of the day, night, or weekend, or just plain do free consulting for them. Dont' stand for it. If they want to be calling you and asking questions or have you come in 'for a few minutes" to show them how to do something, establish a firm rate, ($50/hr is a generous consulting rate) paid in advance. Don't let them take advantage of you.
If you're thinking about cutting out early, at least try to weigh in your impact on the other employees... you probably know what life will be like for them for the next 2-4 weeks as things are trying to get settled back down. Don't let one manager ruin your day AND theirs. The manager might have it coming to him, but he'll just spread the grief around as most PHB's tend to do.
Agents will be arriving in a few minutes to correct this. Just remain seated in front of your computer.
It seems that in your area it's not possible to exercise effective control over the individual computers. You might be able to mitigate the damage caused by outbreaks by purchasing some managed switches to replace your dumb switches currently in use. Some switches can cap bandwidth, or disconnnect/throttle ports if they exceed usage limits for a set amount of time. Most of them are smart enough that you can specify the direction that the triggers are placed, so you can tell the switch basically, "if any user on this switch exceeds 100k/sec upstream for 5 seconds, cap their upstream to 10k/sec for the next five minutes." They also usually let you monitor current bandwidth usage, and some provide short term history for bandwidth or violations, either of which makes it fairly easy to spot owned machines. You can always just shut their port off remotely until the owner is so motivated by lack of internet connection that they clean up their machine. It's not a cure, but it can dull the symptoms. (it also cuts down on the foot traffic required by you)
The only major drawback is price. Managed switches can be expensive, and it sounds like you'll need a number of them. Best way to get funding for something like this approved is to throw together a report outlining how many man hours and how much network downtime this will save over say, the next 3 years. When they see the dollar difference and performance paybacks, they'll get out the checkbook. You can also roll it out incrementally, by getting a fast switch that's managed, as your main node, that initially feeds the existing dumb switches at each of the buildings. Watch how fast people fix their machines when they have like 11 other tennants on their back because everyone's internet in the building is down. Then take the noise from the other 11 tennants that's pointed to you and redirect that into a requisition for a managed switch in that building, killing several birds with one stone. By that route you'l eventually get your entire network to managed switches and your problem will be much easier to deal with. Happy managers (no complaining tennants), happy tennants (no downtime), and a few pissed of ppl with owned machines. Justice all-around.
I didn't see any reference to the actual posted speed limit in the article (might have missed it tho) but it said the estimated speed of the car was 50km/hr. That's clipping along pretty good for a sharply curved on-ramp, and the officers said they felt that the car was moving too fast to negotiate the onramp. The officers also said the car was going faster than any other car they'd seen that evening. Whether they were speeding or not, hard to say. Does sound like they were driving faster than was safe, and that's something you're more likely to see in a car-bomber than grandma getting groceries.
;-)
You also mention, Since everyone knows that US troops follow up a warning shot by spraying cars with hundreds of rounds in a panic, the driver sped up. but if you re-read the report, you'll see that the team encountered between 15 and 30 vehicles prior to this one, all of which reacted correctly and stopped their vehicle. Fortunately the people that live in a war zone react more sensibly to an armed roadblock than the "everyone" you are referring to.
The driver screwed up, big time, and his passenger paid the price. There were mechanisms that should have made it possible to cover for his mistake, but they failed. That doesn't make it the soldiers' fault that the guy made a big mistake, it just means they failed to go the extra mile to compensate for a fool's behavior. You can't count on the soldiers to give you a wide margin of error in a warzone, because for them to give anyone a wide margin is suicide.
The 1-76 TOC had two means of communicating with 4th Brigade, its higher headquarters: Voice Over Internet Protocol (VOIP)2 and FM. The 1-76 FA Battle Captain was using only VOIP to communicate with 1-69 IN, but experienced problems with VOIP, therefore losing its only communication link with 1-69 IN, other than going through 4th Brigade. (Annex 97C). As a result, the Battle Captain was unable to pass updated information about the blocking mission either directly to 1-69 IN, or to 4th Brigade. He did not attempt to contact 4th Brigade via FM communications. (Annex 63C).
and
(U) Mr. Carpani told Sergeant First Class Feliciano who Ms. Sgrena was and that he was trying to get to the airport. He told Sergeant First Class Feliciano that he heard shots from somewhere, and that he panicked and started speeding, trying to get to the airport as quickly as possible. Mr. Carpani further told Sergeant First Class Feliciano that he continued to speed down the ramp, and that he was in a hurry to get to the airport. (Annexes 91C, 136C).
So it all came down to two issues.
(1) Failure to communicate. The car wasn't where it should be, wasn't informed of what was waiting ahead of them, gave its position but that information was not forwarded to the roadblock, so they were not expecting them.
(2) The driver then risked the lives of everyone in the vehicle by reacting with very bad judgement when he arrived at the roadblock. (accelerating the vehicle after he was spotlighted, laserpointered, and heard the warning shots)
Bad decisions by the driver of the vehicle, amplified by failure to communicate.
If you're behind a firewall with a proxy, you're going nowhere until you at least find the proxy. From there there's no guarantees of course, but it's a necessary first step. (worry about the key after you find the door)
establish a connection to something... anything that's persistent. (download a big file on www?)
then on the command line:
netstat -n | grep "ESTAB"
That'll kick out a list of established network connections, the IP they go to, and the port they're going out on. That should help you find your proxy server's IP and port number.