that's cute, but often the ip you have is not the origin, but a hapless victim which is being used to launch the attack and/or hide the tracks of the real blackhat
by sending data back to that ip, you may be unwittingly being used to help the intruder hide and you may appear to be the intruder in the logs of the machine which the blackhat is using as a stepping stone
that's probably not what you are trying to do and that's why I just add those ips to a droplist instead of sending data back
Re:It's more difficult than they make it out to be
on
Building The Navy Intranet
·
· Score: 3, Interesting
Now that you've ordered through the GSA contract, you have to receive your goods. This takes a very long time. The terms for payment from the US Government is not what you would call favorable to the vendor. The stuff you've bought has to get sent to the GSA, then the GSA has to send it to you. Has anyone ever heard of efficiency in a government agency?
This is true.. When I was a contractor working at NASA Ames, I helped purchase some pretty large computer equipment which took about six months of meetings and such and really seemed pretty long and pointless.
But the really pointless part was that this gear was finally shipped to us, but sat in the shipping building on the base for several months because it was lost in there among all the other stuff that was bought a year ago.
Some of that stuff never makes it out of the building because the project it was purchased for has been cancelled or the staff working on it are no longer available, etc. There are no doubt dozens of these shipping wherehouses with orphaned obsolete computer gear all over the country.
But, when you work for the government theres really no incentive to rock the boat or streamline anything. It's like working for the post office.
> because anyone can admin them. Probably the biggest problem.
yep -- that's where to real problems come in because Fred from Accounting and Joe from the mail room both "fix" things that have unknown side effects like installing a worm, or broken driver or whatever. Soon, nobody knows quite how the windows server works, and everyone is afraid to upgrade or reboot it because it just might not come back up!
I think that goes for most things these days especially computer related stuff.
You can get all the scoop on apache, mysql, even advanced network design with vpns and ipsec all from google searches. Just a few years ago, this sort of info would only be available through textbooks, vendor manual, and trade publications.
I think we're getting to the point where C will be considered about as useful as assembler. That is to say, for some tasks it's essential. For most, though, it's more trouble than its worth.
Thank you for your expert opinon on language development. Having read your thoughts, I am convinced that it was more trouble than it was worth.:-P
maybe I'm just really opinionated, but adding object orientation to fortran 30 YEARS after it was designed just isn't going to work.
Look at OO in perl compared to python, java, or C++ for an example of objects scotch taped onto a working language as an afterthought. It looks like a cat ran across the keyboard with all the crazy => $ # and other syntax that makes it hard to read.
Also, this "old solid codebase" won't be used with the new object features -- that would just be a bad idea after another.
When I worked at NASA Ames in their supercomputing center in the mid 90s, I was always amazed at how those huge computers were just being used to work matix solutions in fortran all the time. That's really about all crays are used for, which is a shame.
but, like I said, that's just my opinion of fortran
if this works, lets get them to develop ALGOL 2002 or maybe RATFOR 2006?
no it isn't. the predictions were implying that the keyboard and monitor on computers will be an archaic memory like paper tape or punch cards. Some even predicted this would happen before the year 2000.
it didn't happen, it won't happen. it's much faster to type than to talk for technical things where accuracy is important.
but, anybody can be a 'Futurist', the trick is to make up something outlandish with a grain of possibility.
with the buffer overflow issues that ssh has has in the last year, I have blocked port 22 from the public side of my servers, and allow it from the internal network which is reachable via IPSEC.
this is more secure and stops ssh scans from public script kiddies.
there are lots of other reasons too
Re:all the isps already support linux
on
AOL's new Linux PC
·
· Score: 1
well, paul. I really don't see the difference.
It's not like the ISP comes to my house and helps me run Windows or something.
I said the ISP supports linux, not the ISP's help desk does. Most of them don't have a clue what that would mean, but don't want a million questions about linux when they don't really know how windows works either.
Re:all the isps already support linux
on
AOL's new Linux PC
·
· Score: 1
all you need to do is ifconfig your MTU down so it will pass through the hubs.
the fact that AOL happens to be pre-broken that way doesn't make it better than a real ISP, just lucky
Imagine is Microsoft did start to sell Word and PowerPoint and the rest for Linux os.
They could probably port all of it with the help of wine, cygwin, and their MS staff of developers, etc. in a relatively short amount of time.
What would that do to the momentum of the various projects? A lot of the projects seem to thrive off the anti-microsoft "let's make a better version of excell" type of enthusiasm. But if you could just go buy a copy of excel and run it on linux, a lot of people would.
Those microsoft apps would pretty much round out the rest of the linux as corporate desktop issues. So, a large flood of inexpensive desktop machines would be built for offices. They would probably all buy a copy of MS office over any staroffice/openoffice free download just on the MS name.
I guess what I'm thinking here is that ms apps for linux sounds stupid at first, but it might actually be a powerful approach for MS to use -- even though it seems like the opposite of what they would want to do.
a single-cdrom iso for installing a cluster of mandrake linux boxes, includes the s/w for the servers and clients (look for filenames like ganglia-monitor, clusterit, etc) at a cost of $0.00
that link is to a cluser of unix workstations vs a cluster of windows servers.
the unix workstations are not running on x86 hardware.
unix vendors ALWAYS screw their clients with outrageous support contracts that don't really help.
The prices you quote also include a 3 YEAR support package from both vendors.
almost all the cost of the unix solution is in that support contract, not the os or the hardware.
Here's the quote from the FAQ:
Don't be silly.:)
From this FAQ [tpc.org] you'll see:
In general, TPC benchmarks are system-wide benchmarks, encompassing almost all cost dimensions of an entire system environment the user might purchase, including terminals, communications equipment, software (transaction monitors and database software), computer system or host, backup storage, and three years maintenance cost. Therefore, if the total system cost is $859,100 and the throughput is 1562 tpmC, the price/performance is derived by taking the price of the entire system ($859,100) divided by the performance (1562 tpmC), which equals $550 per tpmC.
Most people would focus on the hardware cost, but in reality the highlighted maintenance cost took the precedence.
Most midrange UNIX server has outragous maintenance cost. The maintenance cost of a UNIX server in the third year could be exceeding the cost of the hardware itself. It's due to the fact that older parts are difficult to find, thus make maintaining older servers more difficult. Besides, they really want to cut older production lines in favor of newer servers production.
x86 platform is known to have flat and lower maintenance cost, due to the low cost hardware and high compability with older hardware, i.e. older parts can be found easily. That's why Microsoft could easily beat the TPC pissing races.
Well, actually, Star Office started as a commericial product from a company called Star Division. The company was eventually aquired by Sun who in turned offered Star Office as a free download and then open souced a version of it and has now gone back to selling branded versions of the open source project
And, of course, Balmer would have you believe that for every copy of staroffice sold, RMS gets a check because the GPL is a virus.:*)
I agree -- if microsoft sold software that simply did what it promised, they would control the market forever.
Instead, they are chasing away the power users and developers who are fed up with yet another bug from microsoft and yet another wacky api to deal with that cost lots of $$$ just to read the docs which aren't totally correct.
Compare that with the linux community and all the freely available help and software options, and it's really really tempting to try out linux.
Then after you try it a little, all that screechy hype from Microsoft seems hollow and expensive.
Slashdot Mirror
This is just another gyromouse story, isnt it?
what about the linux virtual server project?
or the distributed filesystem with network disk
block drivers
or the arch for iptables/netfilter, etc?
many of these things are innovative, some may have also been available
for other os at the same time or sooner, but still
that's cute, but often the ip you have is not the origin, but a hapless victim
which is being used to launch the attack and/or hide the tracks of the real blackhat
by sending data back to that ip, you may be unwittingly being used to help the intruder hide
and you may appear to be the intruder in the logs of the machine which the blackhat is using as a stepping stone
that's probably not what you are trying to do
and that's why I just add those ips to a droplist instead of sending data back
Now that you've ordered through the GSA contract, you have to receive your goods. This takes a very long time. The terms for payment from the US Government is not what you would call favorable to the vendor. The stuff you've bought has to get sent to the GSA, then the GSA has to send it to you. Has anyone ever heard of efficiency in a government agency?
This is true.. When I was a contractor working at NASA Ames, I helped purchase some
pretty large computer equipment which took about six months of meetings and such
and really seemed pretty long and pointless.
But the really pointless part was that this gear was finally shipped to us, but
sat in the shipping building on the base for several months because
it was lost in there among all the other stuff that was bought a year ago.
Some of that stuff never makes it out of the building because the project
it was purchased for has been cancelled or the staff working on it are no longer
available, etc. There are no doubt dozens of these shipping wherehouses with
orphaned obsolete computer gear all over the country.
But, when you work for the government theres really no incentive
to rock the boat or streamline anything. It's like working for
the post office.
Basically, a maximized emacs window with all the commands you can use without a mouse, and no bloat.
;)
I think that's the first time I've seen Emacs and no-bloat in the same sentence!
that site works just fine. the image paths are
bogus and have backslashes in them which is just
plain wrong.
that has NOTHING to do with ie or mozilla.
Every time I go to slashdot.info, i get a different new site. So far i've gotten The Register, CNN.com, and some other world news outlet.
yes, it seems to have the same articles as
slashdot, but fresher articles and no repeated
front page "stories" as news.
> because anyone can admin them.
Probably the biggest problem.
yep -- that's where to real problems come in
because Fred from Accounting and Joe from the
mail room both "fix" things that have unknown
side effects like installing a worm, or broken
driver or whatever. Soon, nobody knows quite
how the windows server works, and everyone is
afraid to upgrade or reboot it because it just
might not come back up!
I think that goes for most things these days
especially computer related stuff.
You can get all the scoop on apache, mysql,
even advanced network design with vpns and ipsec
all from google searches. Just a few years ago,
this sort of info would only be available through
textbooks, vendor manual, and trade publications.
I think we're getting to the point where C will be considered about as useful as assembler. That is to say, for some tasks it's essential. For most, though, it's more trouble than its worth.
:-P
Thank you for your expert opinon on language development. Having read your thoughts,
I am convinced that it was more trouble than
it was worth.
maybe I'm just really opinionated, but adding
object orientation to fortran 30 YEARS after it
was designed just isn't going to work.
Look at OO in perl compared to python, java, or C++
for an example of objects scotch taped onto a working
language as an afterthought. It looks like a cat
ran across the keyboard with all the crazy => $ #
and other syntax that makes it hard to read.
Also, this "old solid codebase" won't be used
with the new object features -- that would just
be a bad idea after another.
When I worked at NASA Ames in their supercomputing
center in the mid 90s, I was always amazed at how
those huge computers were just being used to work
matix solutions in fortran all the time. That's
really about all crays are used for, which is a
shame.
but, like I said, that's just my opinion of fortran
if this works, lets get them to develop ALGOL 2002
or maybe RATFOR 2006?
and all the RDM hardware plans will be dropped
as they aren't needed
no it isn't. the predictions were implying that
the keyboard and monitor on computers will be an
archaic memory like paper tape or punch cards.
Some even predicted this would happen before the
year 2000.
it didn't happen, it won't happen. it's much faster
to type than to talk for technical things where
accuracy is important.
but, anybody can be a 'Futurist', the trick is to
make up something outlandish with a grain of possibility.
with the buffer overflow issues that ssh has
has in the last year, I have blocked port 22 from
the public side of my servers, and allow it from
the internal network which is reachable via IPSEC.
this is more secure and stops ssh scans from public script kiddies.
there are lots of other reasons too
well, paul. I really don't see the difference.
It's not like the ISP comes to my house and helps
me run Windows or something.
I said the ISP supports linux, not the ISP's help
desk does. Most of them don't have a clue what
that would mean, but don't want a million questions
about linux when they don't really know how windows
works either.
all you need to do is ifconfig your MTU down
so it will pass through the hubs.
the fact that AOL happens to be pre-broken that
way doesn't make it better than a real ISP, just lucky
gentoo users, try 'emerge celestia'
there is a ebuild file ready and waiting to go
it will fetch and build all the required parts
and then build celestia -- nice!
I've still never seen an ISP that doesn't support
linux. Sometimes they don't understand that they
do, but it's just DHCP or pppoe.
also, aol is a terribly expensive isp with lots
of unneeded extra baggage
Imagine is Microsoft did start to sell Word and
PowerPoint and the rest for Linux os.
They could probably port all of it with the help of wine,
cygwin, and their MS staff of developers, etc. in
a relatively short amount of time.
What would that do to the momentum of the various
projects? A lot of the projects seem to thrive off
the anti-microsoft "let's make a better version of
excell" type of enthusiasm. But if you could just
go buy a copy of excel and run it on linux, a lot
of people would.
Those microsoft apps would pretty much round out the rest
of the linux as corporate desktop issues. So, a
large flood of inexpensive desktop machines would
be built for offices. They would probably all buy
a copy of MS office over any staroffice/openoffice
free download just on the MS name.
I guess what I'm thinking here is that ms apps for
linux sounds stupid at first, but it might actually
be a powerful approach for MS to use -- even though
it seems like the opposite of what they would want to do.
There's barely $200,000 worth of price difference there. 10% of the total saved by using Linux. Worth it?
What kind of question is that?
um. yes. I would prefer to keep my extra $200,000 thank you.
...
:)
a single-cdrom iso for installing a cluster of mandrake linux boxes, includes the s/w for the servers and clients (look for filenames like ganglia-monitor, clusterit, etc) at a cost of $0.00
I don't see how they're going to beat this
heh.
I guess they make up the difference in volume.
that link is to a cluser of unix workstations vs
:)
a cluster of windows servers.
the unix workstations are not running on x86
hardware.
unix vendors ALWAYS screw their clients with
outrageous support contracts that don't really help.
The prices you quote also include a 3 YEAR support
package from both vendors.
almost all the cost of the unix solution is in
that support contract, not the os or the hardware.
Here's the quote from the FAQ:
Don't be silly.
From this FAQ [tpc.org] you'll see:
In general, TPC benchmarks are system-wide benchmarks, encompassing almost all cost dimensions of an entire system environment the user might purchase, including terminals, communications equipment, software (transaction monitors and database software), computer system or host, backup storage, and three years maintenance cost. Therefore, if the total system cost is $859,100 and the throughput is 1562 tpmC, the price/performance is derived by taking the price of the entire system ($859,100) divided by the performance (1562 tpmC), which equals $550 per tpmC.
Most people would focus on the hardware cost, but in reality the highlighted maintenance cost took the precedence.
Most midrange UNIX server has outragous maintenance cost. The maintenance cost of a UNIX server in the third year could be exceeding the cost of the hardware itself. It's due to the fact that older parts are difficult to find, thus make maintaining older servers more difficult. Besides, they really want to cut older production lines in favor of newer servers production.
x86 platform is known to have flat and lower maintenance cost, due to the low cost hardware and high compability with older hardware, i.e. older parts can be found easily. That's why Microsoft could easily beat the TPC pissing races.
ouch -- that is expensive, and doesn't include the
clustering option, whatever that would cost too.
Where's the VALUE in that?
Well, actually, Star Office started as a commericial product from a company called Star Division. The company was eventually aquired by Sun who in turned offered Star Office as a free download and then open souced a version of it and has now gone back to selling branded versions of the open source project
:*)
And, of course, Balmer would have you believe
that for every copy of staroffice sold, RMS gets a check because
the GPL is a virus.
I agree -- if microsoft sold software that simply
did what it promised, they would control the market
forever.
Instead, they are chasing away the power users and
developers who are fed up with yet another bug
from microsoft and yet another wacky api to deal
with that cost lots of $$$ just to read the docs
which aren't totally correct.
Compare that with the linux community and all the
freely available help and software options, and
it's really really tempting to try out linux.
Then after you try it a little, all that screechy
hype from Microsoft seems hollow and expensive.