If it is Sandvine using heuristics to badly determine that google is P2P, possibly it is because of Google Web Accelerator, how the google extension pre-downloads the first result of a google search, or the network.prefetch-next setting in firefox. I have not heard anyone write about how they are configured related to those issues.
Also main() is supposed to take 2 arguments, in the posted version it takes a variable number of arguments. main() and main(void) are not necessarily interchangeable because the first means a function that can take an arbitrary number of args and second means a function that takes none.
Is the compiler going to figure-out that it really uses none? What about running this on an architecture that passes var args differently than a handful of arguments. Is _start() going to call it correctly? What if you pass args to the program when running it?
Heap-based buffer overflow in the NAT networking components vmnat.exe and vmnet-natd in VMWare Workstation 5.5, GSX Server 3.2, ACE 1.0.1, and Player 1.0 allows remote authenticated attackers, including guests, to execute arbitrary code via crafted (1) EPRT and (2) PORT FTP commands.
Unless when you say "encrypted traffic" you mean IPsec this will not do you any good long term. Comcast is sending RST packets so unless both ends set-up their firewalls to ignore these packets the OS is going to tear-down the connection. Possibly for now they are looking into the payload to figure-out this is torrent traffic, but quite easily they could determine that something from one tcp port to another is torrent traffic simply by looking at ratios for up load and download and numbers of connections after say 2MB or so.
They are not allowed to do this. I was surprising how quickly comcast disabled the encryption of the channels that were available as OTA network in my area when I called. One day I alled, the next day they were back. It was the only decent experience I had with that company.
I got an antenna the first time comcast raised my bill above $50 and cancellled cable. My parents and uncle are paying $100 and $120 per month. I think they recoup costs quickly.
I have only done anon edits to wikipedia also. It has been to correct facts in computer related articles though. I have always left a comment in the discussion with links and an explanation. I have never had an edit reverted, often my edits get rearranged and my links become citations at the end of the article. I have not registered because I do not have the time to deal with all of the style sorts of issues that are important to the registered people. I like it that they take care of that for me, it is very symbiotic actually.
There are schools in very sparsely populated areas that still use this. Primarily they use it for tele-teaching types of things where the student sits in front of a TV while the teacher on the TV is giving a lesson to the entire district or even state. It should not just be taken away from them. These places often have no other way to do something like this. They have been investing into this infrastructure for decades. If the spectrum is taken away from them, then they should be paid so that they can create other forms of distance learning. Verizon doesn't want to pay for this, but they just can't wait for when the same schools will pay them for the services that they will provide over that spectrum later.
If that is true it is bad and it is simply a limitation of Windows. For a long time x86 has offered PAE (Physical Address Extension) which gives 36 address lines. In fact PAE is also used for 2M page size which would be useful to use for a frame buffer. You map a reasonable window for frame buffer, another sliding window for textures, and a small window for CSRs for each card and still have plenty of VA space left over.
On the PPC side since the 745x at least there has been support for Extended Addressing (HID0[XAEN]) which gives 36 bit addresses. In fact it is much simpler as the PHB often has support to automatically remap for you too.
Sure, if you have terrorists in caves and detonate a bomb like this above the caves the combustion will suck the oxygen out of caves in addition to the cave-in of a typical detonation.
Yeah and here in the US there is a mandate that all ATSC broadcasters must send program information for at least 12 hours. This is done with EIT (Event Information Tables). EIT-0 (first three hours) is to be sent twice per second, EIT-1 (next three hours) once every three seconds, and EIT-2 and EIT-3 (final 6 hours) every minute. Guess what my set says for pretty much every channel?
"No program information"
Even the PBS station which used to be very good about it somehow stopped sending the EITs. It just shows how much the broadcasters pay attention to the rules. The details are in document A/65C at the ATSC website (http://www.atsc.org/standards.html).
$ ls -l/bin/awk -r-xr-xr-x 2 root bin 85368 May 19 2006/bin/awk* $ ldd/bin/awk libm.so.1 => /usr/lib/libm.so.1 libc.so.1 => /usr/lib/libc.so.1 libdl.so.1 => /usr/lib/libdl.so.1 /usr/platform/SUNW,Ultra-4/lib/libc_psr.so.1
I can almost always get away with an awk script instead of a perl script, sometimes gawk or nawk is better. Plus awk is great for few liners from the command line too. Awk and sed together are even better.
Here is perl BTW:
$ du -k/usr/perl5/5.6.1 | tail -1 27935 /usr/perl5/5.6.1
Yeah by vote for favorite unbloatware is awk. I have used it for unusual things like better hex dumps and pstree-like scripts.
You realize that they can send the RST to both ends right? You realize that unless you are using something like RAW sockets your app won't have a chance to see the RST? Do you really want to re-create a TCP IP stack in your bit torrent client?
Also, it does not help much if a remote server sends a 1500 byte packet your way (especially not with the DO_NOT_FRAGMENT bit set in the TCP headers). Then you get latency increases (or worse - dropped packets), and not just the extra header overhead due to smaller packet size.
If you are talking TCP, the MSS option in the initial TCP handshake is incredibly common, this should prevent datagrams being so large that they get fragmented on the way back to you.
Also the DF bit is in the IPv4 header not TCP. You most often see that set in PMTUD (Path MTU discovery), ie the OS is trying to figure-out the maximum MTU so that nothing is larger than the largest fragment size anywhere back to you.
You have never seen terrible fragmentation latency until you did IP over ATM.
I'm not using them as a service provider due to them using PPPoE, which increases packet fragmentation and reduces speed).
Whenever I run into PPPoE, I just decrease my MTU by a bit. Then the hit is so small I cannot notice. It keeps my fragments going out small enough so that they do not get broken-up later.
Slashdot Mirror
If it is Sandvine using heuristics to badly determine that google is P2P, possibly it is because of Google Web Accelerator, how the google extension pre-downloads the first result of a google search, or the network.prefetch-next setting in firefox. I have not heard anyone write about how they are configured related to those issues.
Because if you compare my copy of SMB to yours they are not equal and in fact are encrypted.
Also main() is supposed to take 2 arguments, in the posted version it takes a variable number of arguments. main() and main(void) are not necessarily interchangeable because the first means a function that can take an arbitrary number of args and second means a function that takes none.
Is the compiler going to figure-out that it really uses none? What about running this on an architecture that passes var args differently than a handful of arguments. Is _start() going to call it correctly? What if you pass args to the program when running it?
CVE-2005-4459:
You don't even need to be root in the guest VM.
No because it was Sony Europe, not Take Two or Rock Star.
Unless when you say "encrypted traffic" you mean IPsec this will not do you any good long term. Comcast is sending RST packets so unless both ends set-up their firewalls to ignore these packets the OS is going to tear-down the connection. Possibly for now they are looking into the payload to figure-out this is torrent traffic, but quite easily they could determine that something from one tcp port to another is torrent traffic simply by looking at ratios for up load and download and numbers of connections after say 2MB or so.
They are not allowed to do this. I was surprising how quickly comcast disabled the encryption of the channels that were available as OTA network in my area when I called. One day I alled, the next day they were back. It was the only decent experience I had with that company.
I got an antenna the first time comcast raised my bill above $50 and cancellled cable. My parents and uncle are paying $100 and $120 per month. I think they recoup costs quickly.
I have only done anon edits to wikipedia also. It has been to correct facts in computer related articles though. I have always left a comment in the discussion with links and an explanation. I have never had an edit reverted, often my edits get rearranged and my links become citations at the end of the article. I have not registered because I do not have the time to deal with all of the style sorts of issues that are important to the registered people. I like it that they take care of that for me, it is very symbiotic actually.
There are schools in very sparsely populated areas that still use this. Primarily they use it for tele-teaching types of things where the student sits in front of a TV while the teacher on the TV is giving a lesson to the entire district or even state. It should not just be taken away from them. These places often have no other way to do something like this. They have been investing into this infrastructure for decades. If the spectrum is taken away from them, then they should be paid so that they can create other forms of distance learning. Verizon doesn't want to pay for this, but they just can't wait for when the same schools will pay them for the services that they will provide over that spectrum later.
Yeah $300k plus $300 per month assorted condo fees and there are some pretty bad parts of San jose where you would want your kid growing-up.
If that is true it is bad and it is simply a limitation of Windows. For a long time x86 has offered PAE (Physical Address Extension) which gives 36 address lines. In fact PAE is also used for 2M page size which would be useful to use for a frame buffer. You map a reasonable window for frame buffer, another sliding window for textures, and a small window for CSRs for each card and still have plenty of VA space left over.
On the PPC side since the 745x at least there has been support for Extended Addressing (HID0[XAEN]) which gives 36 bit addresses. In fact it is much simpler as the PHB often has support to automatically remap for you too.
"I hate Illinois Nazis."
Do all your calculations in cents.
Sure, if you have terrorists in caves and detonate a bomb like this above the caves the combustion will suck the oxygen out of caves in addition to the cave-in of a typical detonation.
Yeah and here in the US there is a mandate that all ATSC broadcasters must send program information for at least 12 hours. This is done with EIT (Event Information Tables). EIT-0 (first three hours) is to be sent twice per second, EIT-1 (next three hours) once every three seconds, and EIT-2 and EIT-3 (final 6 hours) every minute. Guess what my set says for pretty much every channel?
"No program information"
Even the PBS station which used to be very good about it somehow stopped sending the EITs. It just shows how much the broadcasters pay attention to the rules. The details are in document A/65C at the ATSC website (http://www.atsc.org/standards.html).
Argh! actually this is better:
$ while read v; do echo "$v" foo; done >foo
Here's the 'real' way to do what you want from /bin/sh:
$ while read v; do echo "$v" >> foo; done
Perl not bloated!?
/bin/awk on a solaris box:
/bin/awk /bin/awk* /bin/awk /usr/lib/libdl.so.1
/usr/platform/SUNW,Ultra-4/lib/libc_psr.so.1
/usr/perl5/5.6.1 | tail -1
Here is
$ ls -l
-r-xr-xr-x 2 root bin 85368 May 19 2006
$ ldd
libm.so.1 => /usr/lib/libm.so.1
libc.so.1 => /usr/lib/libc.so.1
libdl.so.1 =>
I can almost always get away with an awk script instead of a perl script, sometimes gawk or nawk is better. Plus awk is great for few liners from the command line too. Awk and sed together are even better.
Here is perl BTW:
$ du -k
27935 /usr/perl5/5.6.1
Yeah by vote for favorite unbloatware is awk. I have used it for unusual things like better hex dumps and pstree-like scripts.
You realize that they can send the RST to both ends right? You realize that unless you are using something like RAW sockets your app won't have a chance to see the RST? Do you really want to re-create a TCP IP stack in your bit torrent client?
They are already available:
m inione-specs-revealed-its-actually-got-some-juice- 261388.php
http://gizmodo.com/gadgets/like-the-iphone/meizu-
If you are talking TCP, the MSS option in the initial TCP handshake is incredibly common, this should prevent datagrams being so large that they get fragmented on the way back to you.
Also the DF bit is in the IPv4 header not TCP. You most often see that set in PMTUD (Path MTU discovery), ie the OS is trying to figure-out the maximum MTU so that nothing is larger than the largest fragment size anywhere back to you.
You have never seen terrible fragmentation latency until you did IP over ATM.
I second this, the example pictures someone else posted were incredible.
Whenever I run into PPPoE, I just decrease my MTU by a bit. Then the hit is so small I cannot notice. It keeps my fragments going out small enough so that they do not get broken-up later.