But is this really Google or is it Nest, AFAIK Nest is a separate company, although owned by Google but still governed by a separate CEO who is the one cited in TFA.
I actually think that we do know because if they (either NSA och the CIA) would ever have found anything it would have been posted all over the media. To really win over the population and get even more funds all they need is that one true case, that they haven't announced that tells me that they have none to show and instead they play the "if only we could tell you what we know" card.
Something tells me that the FSB know better than either you or me about this subject considering that they have been fighting the CIA and the NSA for decades. Now of course there is no proof that this new processor and OS have anything to do with the FSB but being able to produce 100% of the process in Russia is certainly something that the FSB would like very much. Of course the NSA will get their hands on it but I seriously doubt that they can influence the design or put their back doors into it (if the FSB is part of the project). During the Cold War the CIA never ever managed to get a single agent inside the KGB while the CIA itself where packed with KGB people and something tells me that Putin is more hard line than the old managers.
I think that you vastly underestimate the complexity of protecting a network for a full scale targeted attack by some one as NSA with all their resources. For one thing they would hardly use the regular network for this at all.
You do realise that this was affecting 15.10 right? Which is an intermediate/unstable version, the latest LTS from Ubuntu is 14.04. And there is nothing anti-UNIX with either systemd or GNOME 3, that you don't like a project does not automatically deem it anti Unix.
Because it's just one release in the v4 series, 4.1 has long term support, 4.2.8 was EOLd just before 4.4.0 was released and not when 4.5.0 is about to be released then 4.3.6 is EOLd. How many versions do you suppose that the kernel guys should support? At the moment they support 12 different kernels which they now cut down to 11.
And you as the end user is not the supposed user of these kernels anyways, they are meant for the different distribution maintainers who in turn will either switch kernels for you automatically or backport changes if they themselves support other kernels for long term (like Red Hat and Ubuntu does). This is not like Microsoft forcing you to upgrade to WIndows 10 by EOL:ing windows 7, it's more like when they EOLd Windows 7 Service Pack 1 when Windows 7 Service Pack 3 where releases (or what ever they call these things now a days, I haven't really used WIndows since Windows 2000 so this details might have changed since then but I guess that you get the analogy anyways).
Well most exploits are local anyway so unless he have other people logging into his machine or running software with exploits (and if so then those exploits in themselves is usually enough to gain root anyway) he has really nothing to fear here. Yes there have probably been remote exploits detected in the kernels that he runs, but they are then connected to certain remove file systems or communications protocols that he probably doesn't use either.
If some one gains access to your system it's 99% likely to be due to exploits in userspace and not the kernel.
Researchers found that when the seller of these popular items was self-identified as female, the auction got fewer bids and a lower final price.
So they where only looking at sellers that where clearly identified as females, I guess that also means that the comparison group then was both males and females but with user names and descriptions that didn't identify them as females.
People are not really upset that there where more versions really, they where more upset that the new versions where the only versions since Lucas deemed the older versions tabu.
While you are correct regarding the lengths of strings you are wrong about printf, it works just fine with Unicode/UTF-8.
That certain database vendors (read Microsoft) decided that you must declare strings as nvarchar instead of varchar and that you must prefix strings with N such as UPDATE x SET field=N'text'; hasn't really helped either, even though you can connect to the MSSQL with UTF-16.
Well that is your opinion which you of course are entitled to, however many in the embedded space are quite happy with systemd since it means that they can cut down on the number of old and unreliable services they used to rely on.
Where are those mythical hyper aggressive emotional campaign posts for systemd? All that I can see is hyper aggressive anti systemd trolls spewing the same false statements over and over (like throwing away stderr). Barsteward is right on the money, that AC post that was included i the questions where dead stupid and has been proved wrong millions of times. So if the anti systemd trolls can not come up with something other than what is known to be pure lies then yes they are well painted into a corner.
Another case in point is this whole thread about security, there has not been a single argument brought forward by the anti systemd crowd, not a single one. There have only been name callings and butt hurt comments. How is that for painting one self into a corner?
Of course not, we who do not hate systemd has no clue what so ever and is stupid beyond comprehension. Can you please dear oracle enlighten me on how all this affects OS security!
Then you should actually follow the StackOverflow link given by Matthew since that deals with exactly that (redirecting stderr to a file instead of to the journal). However as the AC points out, are you sure that you mean unbuffered? Because unbuffered output is only used for animations like progress bars and so on and is not well suited for either the journal, syslog or to a file but is intended to be displayed on a console.
If you however by unbuffered mean that you want to fetch the log even though it isn't complete due to a service crashing before it can produce a full line of text then that "should" work as is with the journal since a close of stderr should flush the buffer, also do remember that in this case the buffer is not in systemd but in libc (since the service has to call fcntl() in order to change from buffered to unbuffered since buffered output is the default in libc).
Apparantly. Considering that systemd is root (pid 1) that executes scripts that are supposed to run as root as well it will be fun to hear how you guys can see security nightmares here. Is the theory that systemd can get a root script more root than root?
The problem with that thinking is that if you plan to do the same job with a OpenBSD machine vs SomethingElse then both will probably run the exact number and type of services and applications.
While the initial version of systemd-resolvd didn't honor rfc5452 due to it beeing a stub resolver, it has since v223: https://lists.freedesktop.org/...
* systemd-resolved now implements RFC5452 to improve resilience against
cache poisoning. Additionally, source port randomization is enabled
by default to further protect against DNS spoofing attacks.
Slashdot Mirror
But is this really Google or is it Nest, AFAIK Nest is a separate company, although owned by Google but still governed by a separate CEO who is the one cited in TFA.
So publicity is only for the small plots and not the big ones? Why?
I actually think that we do know because if they (either NSA och the CIA) would ever have found anything it would have been posted all over the media. To really win over the population and get even more funds all they need is that one true case, that they haven't announced that tells me that they have none to show and instead they play the "if only we could tell you what we know" card.
Something tells me that the FSB know better than either you or me about this subject considering that they have been fighting the CIA and the NSA for decades. Now of course there is no proof that this new processor and OS have anything to do with the FSB but being able to produce 100% of the process in Russia is certainly something that the FSB would like very much. Of course the NSA will get their hands on it but I seriously doubt that they can influence the design or put their back doors into it (if the FSB is part of the project). During the Cold War the CIA never ever managed to get a single agent inside the KGB while the CIA itself where packed with KGB people and something tells me that Putin is more hard line than the old managers.
you have to start somewhere. The US didn't start the space race with Soviet by landing on the moon.
I think that you vastly underestimate the complexity of protecting a network for a full scale targeted attack by some one as NSA with all their resources. For one thing they would hardly use the regular network for this at all.
You do realise that this was affecting 15.10 right? Which is an intermediate/unstable version, the latest LTS from Ubuntu is 14.04. And there is nothing anti-UNIX with either systemd or GNOME 3, that you don't like a project does not automatically deem it anti Unix.
So in other words you have no clue what so ever how kernel.org works and what the people there do vs what is handled by the distributions?
Because it's just one release in the v4 series, 4.1 has long term support, 4.2.8 was EOLd just before 4.4.0 was released and not when 4.5.0 is about to be released then 4.3.6 is EOLd. How many versions do you suppose that the kernel guys should support? At the moment they support 12 different kernels which they now cut down to 11.
And you as the end user is not the supposed user of these kernels anyways, they are meant for the different distribution maintainers who in turn will either switch kernels for you automatically or backport changes if they themselves support other kernels for long term (like Red Hat and Ubuntu does). This is not like Microsoft forcing you to upgrade to WIndows 10 by EOL:ing windows 7, it's more like when they EOLd Windows 7 Service Pack 1 when Windows 7 Service Pack 3 where releases (or what ever they call these things now a days, I haven't really used WIndows since Windows 2000 so this details might have changed since then but I guess that you get the analogy anyways).
And it would webscale!
Well most exploits are local anyway so unless he have other people logging into his machine or running software with exploits (and if so then those exploits in themselves is usually enough to gain root anyway) he has really nothing to fear here. Yes there have probably been remote exploits detected in the kernels that he runs, but they are then connected to certain remove file systems or communications protocols that he probably doesn't use either.
If some one gains access to your system it's 99% likely to be due to exploits in userspace and not the kernel.
Wait there is a law against men selling kittens now? Wow the SJW's really have gotten too far now!
Well this was in TFA:
Researchers found that when the seller of these popular items was self-identified as female, the auction got fewer bids and a lower final price.
So they where only looking at sellers that where clearly identified as females, I guess that also means that the comparison group then was both males and females but with user names and descriptions that didn't identify them as females.
People are not really upset that there where more versions really, they where more upset that the new versions where the only versions since Lucas deemed the older versions tabu.
While you are correct regarding the lengths of strings you are wrong about printf, it works just fine with Unicode/UTF-8.
That certain database vendors (read Microsoft) decided that you must declare strings as nvarchar instead of varchar and that you must prefix strings with N such as UPDATE x SET field=N'text'; hasn't really helped either, even though you can connect to the MSSQL with UTF-16.
According to TFA they appear to be royaly fucked.
Well that is your opinion which you of course are entitled to, however many in the embedded space are quite happy with systemd since it means that they can cut down on the number of old and unreliable services they used to rely on.
https://systemd.events/systemd...
http://bec-systems.com/site/10...
Unfortunately not since it's not Linux, Zephyr RTOS is a completely different kernel, it's Wind Rivers old Rocket RTOS that has been renamed.
Where are those mythical hyper aggressive emotional campaign posts for systemd? All that I can see is hyper aggressive anti systemd trolls spewing the same false statements over and over (like throwing away stderr). Barsteward is right on the money, that AC post that was included i the questions where dead stupid and has been proved wrong millions of times. So if the anti systemd trolls can not come up with something other than what is known to be pure lies then yes they are well painted into a corner.
Another case in point is this whole thread about security, there has not been a single argument brought forward by the anti systemd crowd, not a single one. There have only been name callings and butt hurt comments. How is that for painting one self into a corner?
Of course not, we who do not hate systemd has no clue what so ever and is stupid beyond comprehension. Can you please dear oracle enlighten me on how all this affects OS security!
Then you should actually follow the StackOverflow link given by Matthew since that deals with exactly that (redirecting stderr to a file instead of to the journal). However as the AC points out, are you sure that you mean unbuffered? Because unbuffered output is only used for animations like progress bars and so on and is not well suited for either the journal, syslog or to a file but is intended to be displayed on a console.
If you however by unbuffered mean that you want to fetch the log even though it isn't complete due to a service crashing before it can produce a full line of text then that "should" work as is with the journal since a close of stderr should flush the buffer, also do remember that in this case the buffer is not in systemd but in libc (since the service has to call fcntl() in order to change from buffered to unbuffered since buffered output is the default in libc).
Apparantly. Considering that systemd is root (pid 1) that executes scripts that are supposed to run as root as well it will be fun to hear how you guys can see security nightmares here. Is the theory that systemd can get a root script more root than root?
Tried setting "StandardOutput=console" in the unit file?
The problem with that thinking is that if you plan to do the same job with a OpenBSD machine vs SomethingElse then both will probably run the exact number and type of services and applications.
While the initial version of systemd-resolvd didn't honor rfc5452 due to it beeing a stub resolver, it has since v223: https://lists.freedesktop.org/...
* systemd-resolved now implements RFC5452 to improve resilience against
cache poisoning. Additionally, source port randomization is enabled
by default to further protect against DNS spoofing attacks.