The last numbers I saw were a few hundred years' worth, at ~1995's consumption rates. Sadly, I can't for the life of me find that document.
Given what, say, Chinese demand is doing in other areas, what's going to happen to those uranium reserves when *everybody* starts building new nuclear power plants?
Looking at the web page, it's based on a utility written by Van Jacobson, which seems to give it a pretty good pedigree. That having been said, the README says it still relies on ICMP replies, and will suffer similar problems as mentioned above.
The simplest way to measure network latency and lag is probably to control both ends of a TCP connection, and send timestamped data down it with Nagle disabled.
Rsync also works nicely for "upgrading" CD images of beta Ubuntu releases to the final version, and for, say, making a Kubuntu Live CD out of the normal GNOME-based Ubuntu one. It has the advantage that it can spot blocks that have moved around in the new version but are still the same, even if they're no longer on block boundaries.
I've been using Linux for about 14 years (and DOS and AmigaDOS before that), so I'm not exactly an anti-CLI zealot, but I have to say.. I do find GUIs are easier for exploring an unfamiliar program or system. (Imagine being plonked at a UNIX command prompt with no prior knowledge. What are valid commands? "help" won't tell you much, at least in bash.) Once I've gained familiarity, of course, I'm itching for something I can script.
The answer seems to be to provide multiple interfaces to any sizeable program or system component. With good design, it's not exactly hard.
Purely by chance, I noticed the release before it was announced, and rsync'd my beta image to the final version -- and then helped the seed the torrent, of course.
dnsmasq claims to be able to convert these bogus A records back to NXDOMAIN errors, at least for a single IP address (see the --bogus-nxdomain option.)
Alternatively, it couldn't be that hard to a resolv.conf option to something similar, could it?
Exactly. A SYN flood DoS attack isn't trying to establish real connections, so what good would generating RSTs do? More to the point, TFA seems to say that the RSTs are generated on other, real connections, not the "flooded", awaiting-ACK ones.
As I read the article, the RSTs and other anomalies were affecting other connections, not the (pseudo) connections that the flooded SYNs would have set up (had they been aimed at a real host.)
That's not exactly an effective strategy for cutting off a DoS. If they really thought a DoS attack was in progress and wanted to do something about it, I'd expect the outbound SYN rate to be throttled back, or the total outbound packets, or even to disconnect the link altogether.
In any serious nuclear war, the enemy is going to be targeting known missile sites. Presumably the assumption is that a significant %age of weapons may be destroyed before they can be deployed, therefore build in some factor of redundancy. Whether the factor is in fact 56, and if so, whether it's sensible, I couldn't say;-)
Ideally I would run the scan by unplugging the network cable and booting from directly the malware-scanner CD. Unfortunately nobody makes such a thing
TRK? Admittedly, the virus scanners it can pull down and run probably aren't as ideal as dedicated anti-malware/spyware software would be, but it's a start.
Quite. Before I read the linked articles, I did think that maybe it was emailing him a config file, or an internal state dump, or something like that. But looking at the screenshots of the Gmail account, it was just the username / password. I really can't think of any excuses.
Poking at it, they seem to use RTMP, which after a brief googling I couldn't find any rippers for. I've kludged up some code that succeeds (mostly) at at parsing the packets, but I've not got it decoding AMF or any dumping any A/V data yet.
I have a powerstrip where each socket has an individual switch, and the powerstrip is fairly easy to get at. Admittedly, I have to bend down to flip the switches, but I'd have to do that to press buttons on the printer anyway. If I was less/more lazy, I'd mount the strip on the wall..
This is the difference between software and manufacturing. In the software world, a specific implementation is copyright-able, as the code is the implementation, and the code is trivially copyable, like a work of literature. In physical manufacturing, the implementation isn't copyright-able, so patents exist to provide innovators with some protection. In software design, copyright already does this job, patents aren't needed, and are misused to protect abstract ideas, and not their implementation, which was never the intention.
Regarding pharmaceuticals, I feel patents are probably the more acceptable model, as modern drugs are definitely designed, and duplication / manufacturing is non-trivial and not zero-cost. Which isn't to say that there aren't arguments for tweaking the parameters of the system (patent duration length), or introducing public service obligations on pharmaceutical companies. Not to mention making clear that discoveries aren't patentable -- let's eliminate patenting of bits of the genome, or medicinal plants (but by all means patent methods for manufacturing the compounds found in those plants)..
Note that djb says an *empty* directory. Assuming that you can't mkdir() in that directory, then there's nowhere to move the chroot to. I presume that's why he explicitly mentioned the emptiness:)
Another trick I seem to recall using back when I still did Windows was to start taskmgr from an at job, it would end up running as SYSTEM and thereby be capable of killing anything. That little trick probably dates back to NT4 though, so may not work these days..
It's even worse than that. At no point does TFA provide evidence that the text "New Version" is related to Gmail, and not some other Google product.
Apart from the bit that says "Translate phrase for Gmail UI"?
(It could even be outdated text referring to a "New Version" of a product that has already been released.)
Possibly. But agaom. if you do look at the image in TFA, it says "newer UI", and AFAIK Gmail has only ever had one (excluding variations such as the simple HTML mode, etc.)
Slashdot Mirror
.net was originally for network infrastructure organizations (like ISPs), I believe, and it's a lot older than .info.
The last numbers I saw were a few hundred years' worth, at ~1995's consumption rates. Sadly, I can't for the life of me find that document. Given what, say, Chinese demand is doing in other areas, what's going to happen to those uranium reserves when *everybody* starts building new nuclear power plants?
Looking at the web page, it's based on a utility written by Van Jacobson, which seems to give it a pretty good pedigree. That having been said, the README says it still relies on ICMP replies, and will suffer similar problems as mentioned above.
The simplest way to measure network latency and lag is probably to control both ends of a TCP connection, and send timestamped data down it with Nagle disabled.
http://en.wikipedia.org/wiki/X86-64
Rsync also works nicely for "upgrading" CD images of beta Ubuntu releases to the final version, and for, say, making a Kubuntu Live CD out of the normal GNOME-based Ubuntu one. It has the advantage that it can spot blocks that have moved around in the new version but are still the same, even if they're no longer on block boundaries.
I've been using Linux for about 14 years (and DOS and AmigaDOS before that), so I'm not exactly an anti-CLI zealot, but I have to say .. I do find GUIs are easier for exploring an unfamiliar program or system. (Imagine being plonked at a UNIX command prompt with no prior knowledge. What are valid commands? "help" won't tell you much, at least in bash.) Once I've gained familiarity, of course, I'm itching for something I can script.
The answer seems to be to provide multiple interfaces to any sizeable program or system component. With good design, it's not exactly hard.
Purely by chance, I noticed the release before it was announced, and rsync'd my beta image to the final version -- and then helped the seed the torrent, of course.
dnsmasq claims to be able to convert these bogus A records back to NXDOMAIN errors, at least for a single IP address (see the --bogus-nxdomain option.)
Alternatively, it couldn't be that hard to a resolv.conf option to something similar, could it?
Exactly. A SYN flood DoS attack isn't trying to establish real connections, so what good would generating RSTs do? More to the point, TFA seems to say that the RSTs are generated on other, real connections, not the "flooded", awaiting-ACK ones.
As I read the article, the RSTs and other anomalies were affecting other connections, not the (pseudo) connections that the flooded SYNs would have set up (had they been aimed at a real host.) That's not exactly an effective strategy for cutting off a DoS. If they really thought a DoS attack was in progress and wanted to do something about it, I'd expect the outbound SYN rate to be throttled back, or the total outbound packets, or even to disconnect the link altogether.
In any serious nuclear war, the enemy is going to be targeting known missile sites. Presumably the assumption is that a significant %age of weapons may be destroyed before they can be deployed, therefore build in some factor of redundancy. Whether the factor is in fact 56, and if so, whether it's sensible, I couldn't say ;-)
Ideally I would run the scan by unplugging the network cable and booting from directly the malware-scanner CD. Unfortunately nobody makes such a thing
TRK? Admittedly, the virus scanners it can pull down and run probably aren't as ideal as dedicated anti-malware/spyware software would be, but it's a start.
Random geekiness: you can use PAL colour encoding with a 30fps broadcast system...
http://en.wikipedia.org/wiki/PAL-M
Quite. Before I read the linked articles, I did think that maybe it was emailing him a config file, or an internal state dump, or something like that. But looking at the screenshots of the Gmail account, it was just the username / password. I really can't think of any excuses.
According to TFA, we've already got the gene, the problem is switching it on. You probably also ought to read this comment, though.
Poking at it, they seem to use RTMP, which after a brief googling I couldn't find any rippers for. I've kludged up some code that succeeds (mostly) at at parsing the packets, but I've not got it decoding AMF or any dumping any A/V data yet.
This is the 2nd space station...
2nd?
I have a powerstrip where each socket has an individual switch, and the powerstrip is fairly easy to get at. Admittedly, I have to bend down to flip the switches, but I'd have to do that to press buttons on the printer anyway. If I was less/more lazy, I'd mount the strip on the wall..
This is the difference between software and manufacturing. In the software world, a specific implementation is copyright-able, as the code is the implementation, and the code is trivially copyable, like a work of literature. In physical manufacturing, the implementation isn't copyright-able, so patents exist to provide innovators with some protection. In software design, copyright already does this job, patents aren't needed, and are misused to protect abstract ideas, and not their implementation, which was never the intention.
Regarding pharmaceuticals, I feel patents are probably the more acceptable model, as modern drugs are definitely designed, and duplication / manufacturing is non-trivial and not zero-cost. Which isn't to say that there aren't arguments for tweaking the parameters of the system (patent duration length), or introducing public service obligations on pharmaceutical companies. Not to mention making clear that discoveries aren't patentable -- let's eliminate patenting of bits of the genome, or medicinal plants (but by all means patent methods for manufacturing the compounds found in those plants)..
Note that djb says an *empty* directory. Assuming that you can't mkdir() in that directory, then there's nowhere to move the chroot to. I presume that's why he explicitly mentioned the emptiness :)
Is there any advantage to parking heads without spinning down? Shock protection?
What ever happened to early /. UIDs?
:-)
Still working
Another trick I seem to recall using back when I still did Windows was to start taskmgr from an at job, it would end up running as SYSTEM and thereby be capable of killing anything. That little trick probably dates back to NT4 though, so may not work these days ..
It's even worse than that. At no point does TFA provide evidence that the text "New Version" is related to Gmail, and not some other Google product.
Apart from the bit that says "Translate phrase for Gmail UI"?
(It could even be outdated text referring to a "New Version" of a product that has already been released.)
Possibly. But agaom. if you do look at the image in TFA, it says "newer UI", and AFAIK Gmail has only ever had one (excluding variations such as the simple HTML mode, etc.)
Sadly, rockbox hasn't ever supported the 2nd gen nano, don't know if it's likely to support these new models.