In fact the RAID code seems to be orphaned. A while ago I noticed that when a single sector read error occurs on one of the mirrored partitions, the kernel takes the entire partition offline immediately. The user then has to do a "raidhotremove" and attempt a "raidhotadd", and the entire partition will have to be synchronized again.
I think it would be more reasonable in this case to read the sector from the other disk, attempt to write it on the disk that returned an error, read it again, and if that succeeds keep the disk online. There could be a counter that limits this procedure to a certain number of occurrences and then still take the partition offline as "unreliable". As it is now, the soft-failed disk is not kept in sync anymore, and when an error occurs on the other disk the system dies. A recovery attempt would increase the chances of a running system, especially when the disk has automatic bad sector remapping and will write the corrected sector at a different disk location.
I sent this suggestion to a mail address listed in the code and documentation. The reply was that he thought it was a good idea, but had transferred the maintenance to someone else. I sent the message to the new maintainer and heard nothing.
While I probably could write this enhancement myself and submit a patch, I think I have little chance when it does not go through an official kernel maintainer. So I did nothing yet.
With a RAID1 disk implemented in Linux software, there should be an additional level of optimization possible. Having a RAID1 of 2 disks is like having a dual-processor system: there are 2 disks, with 2 head positions, available to read data in parallel.
With good optimization techniques this should yield double the troughput of a single disk. "hardware raid" controllers seldomly make this optimization.
Of course the disks have to seek to the same position for writes, but for reads they are independent.
The big problem is that the US and some other countries fail so miserably in understanding their problem.
When terrorists want to destroy their country and "freedom", these governments don't sit down and asks themselves "why would they want to do that, are we maybe doing some things to them that they don't like?". No. They are going out to destroy them and enforce their "freedom" onto other people.
And that happens to be the original fact that the terrorists did not like. So, a "war on terrorism" is not going to end terrorism, it is just feeding it with new motifs and new people who also want to join in that war.
If the Americans and allies want to protect themselves and head-off terroristic attacks, they must first and foremost focus on themselves, not on the people that are trying to enter their country. They must change behaviour, not only be concerned about their own freedom but also that of other people (including the freedom not to want to join in their definition of a free world), and back down on their desire to control everything and everyone.
Depends on what DSL you have, and what you call "better".
Here in the Netherlands, standard DSL service via the ex-state-monopoly DSL provider pings at about 10ms nationally and 160ms to the USA (west coast).
This is not bad and completely unachievable via satellite, because of laws of physics.
For any geostationary satellite solution you need to add 260ms for one-way and double that for two-way solutions, plus any delays incurred by time-division multiplexing (if applicable). That is a huge increase compared to the above numbers.
The only satellite solution that can be faster is a LEO constellation.
Please note that the primary concern is not that the droplets (as a non-native English speaker I wonder if there would not be a different word than "droplets" to describe 1 to 3 inch bubbles) are radio active.
The primary concern is that they are space debris and might hit other spacecraft.
Now you should consider that these droplets are accidentally released (maybe by a badly designed spacecraft, but certainly not on purpose).
Then look at the USA. They PURPOSELY released a load of needles in space, with the naive idea that they would remain in a small cloud and could be used as a reflector for radio signals. This is also briefly mentioned in the article.
If anything can be called a mess, it was this "experiment". It has been a large contribution to space environment pollution.
>Why should it be OK for Open Office to require even more resources for acceptable performance than Microsoft's legendarily bloated Office?
It is not OK, but it is the way it is. With StarOffice 5.2 it was much worse. Apparently they are improving things.
But let's face it: Microsoft writes better applications for Windows than any competitor.
Office is much faster than OpenOffice. MSIE is faster than Mozilla, or at least has been for a long time. Windows media player is much faster than any competitor. etc etc...
This has been noticed by others as well, and investigations have been made to find if they abuse inside knowledge for anti-competitive measures.
This has been a problem for us as well, but I have been able to fix it on Windows machines.
It is possible to install OpenOffice.org as an administrative user, and then move the program menu items to "All Users". In that state, it is possible to share one OpenOffice installation amongst all users by making the "user" directory in the OpenOffice.org1.1.0 directory writable for everyone.
The downside of this is that all users will share the same preferences, on the machine. So when someone else logs on he/she will get the preferences of the previous user, will see what documents (filenames) that user worked on, etc.
To fix that, you need to make some simple changes to the installation that are similar to what a user install after a/net install does. But you can make these changes in a LOGON script.
After installation, in the file %PROGRAMFILES%\OpenOffice.org1.1.0\program\bootstr ap.ini change these lines: Location=$SYSUSERCONFIG/sversion.ini and UserInstallation=${$Location:$Section:$ProductKey }
In the file %USERPROFILE%\Application Data\sversion.ini (which will have to be created when it does not yet exist when a new user logs on) change this: OpenOffice.org 1.1.0=file:///C:/Documents%20and%20settings/Userna me/Appication%20Data/OpenOffice
Copy the user directory from the %PROGRAMFILES%\OpenOffice.org1.1.0 to that location when it does not yet exist.
Now the OpenOffice settings will be in the user profile and will roam to computers where this user logs on.
After putting these actions in LOGON scripts, we now have automatic installation and roaming of user data with OpenOffice.org
But I agree that this should be automatic, just as it is with Microsoft Office.
Like many rebuttal documents, it starts of to-the-point and with facts, and then it wanders off and shows how the writer of the document did not even understand the statement in the original document.
For example, look at the paragraph about "collaboration". With this, Microsoft want to discuss the calendaring and planning features of Exchange. He completely misses the point.
By coincidence today I filed a bug in Mozilla because it sends the domain name part of the sender mail address as argument to the SMTP EHLO command when sending mail. This should be the hostname of the sending machine.
As it turns out, this was a duplicate of a bug reported in februari 2001, that is still open! That bug in fact talked only about HELO so I did not find it using search. SMTP has been replaced by ESMTP in the meantime, but the issue remains.
Not that there were no contributions, there are 110 followups on the original report, many of them with good suggestions and patches. But in december 2003, things have again come to a standstill it seems.
With such issues, that should be trivial to fix in an evening, open for 3 years, I wonder how a change like this got checked in. Maybe the remarks about OSS made in another thread are sometimes not that far off the mark?
I don't think you got the message. Worms like this could run on your dedicated firewall box (like linksys or draytek). They don't require Windows or an insecure OS at all, they just require sloppy programming in any program that handles network packets.
When this SECAM system sucks so bad, then why are most consumer video recorders converting the video to what is essentially SECAM before recording it, and then converting back to NTSC or PAL when playing back?
This is not true at all! PAL B/G (625 lines 25 fps AM negative modulation with FM audio) is compatible with the existing BW standard in the countries where it is used.
What you are referring to is probably the fact that some TV stations used the opportunity of introduction of color for a system change. Some countries were using 405 or 819 lines in BW. A multisystem TV from those days had to be switched, but not because of the BW-color switchover.
It is funny to see how Americans discuss the fact that PAL is better than NTSC (which certainly is true) while in Europe the people who like good quality picture moved from PAL to DVB...
An external DVB-S or other DVB system receiver connected to the TV set using RGB signals is a lot better than PAL.
The "public broadcasting" in the USA is apparently not the same as in the UK (BBC) and other European countries.
While a lot of people watch commercial tv (and proportionally more in lower-educated classes), public broadcasting still has a very significant market share in this region. It isn't elitist at all.
> So, what's at address 0xfffffff0 when there are only, say, 512MB installed?
The BIOS. That is what this article is about. This used to be a ROM chip but today it is a Flash memory (a nonvolatile memory that is used mostly as a ROM but can be written by a BIOS update utility).
It does not matter how much memory is really in the machine, or how much the motherboard can physically hold. The processor can address 4GB using 32 address lines, and that defines an address space of 4GB. Newer processors address 64GB using 4 extra address lines but they are yet another add-on that has to be enabled lateron by special software support.
It is used as an email address lookup, yes. But it also works from addressbook for some other items. I am not aware of any special schema, does it exist? (we use standard schema with inetorgperson etc)
I guess you will have to live with the changes that occur over time (sometimes called progress). Today's PCs are powerful enough to run Mozilla, and it seems nobody cares to maintain a browser that still runs on a 486 with 16MB.
Given that the limited resources of developers can be spent only once, I rather have them spent on a browser that can keep up with the market demands, than on a browser for some NetBSD geeks that insist on using a 5 years old machine.
Well, it is often not known what type of OS is running on these boxes. I would not presume it is less fragile. Sure there is no user-installed software on it, but I doubt that is the primary reason why packetfilters fail at some point. More likely this results from design failures.
Some boxes run Linux, and some have taken the firewall functionality from other open source systems (like BSD) and put them in their image as an add-on. There certainly is room for an exploitable bug in there. And worse: once this is found, it is often considerably more difficult to update the firmware than it is to update a PC-resident piece of software. (at least for the non-tech-savvy user)
Let's at least keep attention on this, not assume "nothing can happen because I have this hardware firewall"...
> Close, but you're just pushing back the problem from a naming of sites to a naming of categories.
Ahh, I did not realize this could be a problem. With the BeNeLux trademark office, there is a fixed list of categories under which you can register your trademarks. So I assumed the category list would be obvious.
Slashdot Mirror
In fact the RAID code seems to be orphaned.
A while ago I noticed that when a single sector read error occurs on one of the mirrored partitions, the kernel takes the entire partition offline immediately. The user then has to do a "raidhotremove" and attempt a "raidhotadd", and the entire partition will have to be synchronized again.
I think it would be more reasonable in this case to read the sector from the other disk, attempt to write it on the disk that returned an error, read it again, and if that succeeds keep the disk online. There could be a counter that limits this procedure to a certain number of occurrences and then still take the partition offline as "unreliable".
As it is now, the soft-failed disk is not kept in sync anymore, and when an error occurs on the other disk the system dies. A recovery attempt would increase the chances of a running system, especially when the disk has automatic bad sector remapping and will write the corrected sector at a different disk location.
I sent this suggestion to a mail address listed in the code and documentation. The reply was that he thought it was a good idea, but had transferred the maintenance to someone else. I sent the message to the new maintainer and heard nothing.
While I probably could write this enhancement myself and submit a patch, I think I have little chance when it does not go through an official kernel maintainer. So I did nothing yet.
No disk ever guarantees that you can read it back in the future, and the interface does not influence that much.
Maybe you are referring to the bus parity on SCSI drives, but UDMA EIDE drives have block CRC.
Both SCSI and IDE drives have had FEC and bad block remapping for ages.
With a RAID1 disk implemented in Linux software, there should be an additional level of optimization possible.
Having a RAID1 of 2 disks is like having a dual-processor system: there are 2 disks, with 2 head positions, available to read data in parallel.
With good optimization techniques this should yield double the troughput of a single disk. "hardware raid" controllers seldomly make this optimization.
Of course the disks have to seek to the same position for writes, but for reads they are independent.
How much of that is implemented in 2.6 ?
The big problem is that the US and some other countries fail so miserably in understanding their problem.
When terrorists want to destroy their country and "freedom", these governments don't sit down and asks themselves "why would they want to do that, are we maybe doing some things to them that they don't like?". No. They are going out to destroy them and enforce their "freedom" onto other people.
And that happens to be the original fact that the terrorists did not like. So, a "war on terrorism" is not going to end terrorism, it is just feeding it with new motifs and new people who also want to join in that war.
If the Americans and allies want to protect themselves and head-off terroristic attacks, they must first and foremost focus on themselves, not on the people that are trying to enter their country.
They must change behaviour, not only be concerned about their own freedom but also that of other people (including the freedom not to want to join in their definition of a free world), and back down on their desire to control everything and everyone.
Depends on what DSL you have, and what you call "better".
Here in the Netherlands, standard DSL service via the ex-state-monopoly DSL provider pings at about 10ms nationally and 160ms to the USA (west coast).
This is not bad and completely unachievable via satellite, because of laws of physics.
For any geostationary satellite solution you need to add 260ms for one-way and double that for two-way solutions, plus any delays incurred by time-division multiplexing (if applicable).
That is a huge increase compared to the above numbers.
The only satellite solution that can be faster is a LEO constellation.
You mean the Toshiba satellites?
Please note that the primary concern is not that the droplets (as a non-native English speaker I wonder if there would not be a different word than "droplets" to describe 1 to 3 inch bubbles) are radio active.
The primary concern is that they are space debris and might hit other spacecraft.
Now you should consider that these droplets are accidentally released (maybe by a badly designed spacecraft, but certainly not on purpose).
Then look at the USA.
They PURPOSELY released a load of needles in space, with the naive idea that they would remain in a small cloud and could be used as a reflector for radio signals. This is also briefly mentioned in the article.
If anything can be called a mess, it was this "experiment". It has been a large contribution to space environment pollution.
So before you call the kettle black...
>Why should it be OK for Open Office to require even more resources for acceptable performance than Microsoft's legendarily bloated Office?
It is not OK, but it is the way it is.
With StarOffice 5.2 it was much worse. Apparently they are improving things.
But let's face it: Microsoft writes better applications for Windows than any competitor.
Office is much faster than OpenOffice.
MSIE is faster than Mozilla, or at least has been for a long time.
Windows media player is much faster than any competitor.
etc etc...
This has been noticed by others as well, and investigations have been made to find if they abuse inside knowledge for anti-competitive measures.
>This on a Pentium III 600MHz with 128MB of RAM.
The clock has ticked on. Today you buy a Pentium4 at 2.8 GHz with 512MB of RAM. OpenOffice flies on that.
This has been a problem for us as well, but I have been able to fix it on Windows machines.
/net install does. But you can make these changes in a LOGON script.
r ap.ini y }
a me/Appication%20Data/OpenOffice
It is possible to install OpenOffice.org as an administrative user, and then move the program menu items to "All Users".
In that state, it is possible to share one OpenOffice installation amongst all users by making the "user" directory in the OpenOffice.org1.1.0 directory writable for everyone.
The downside of this is that all users will share the same preferences, on the machine. So when someone else logs on he/she will get the preferences of the previous user, will see what documents (filenames) that user worked on, etc.
To fix that, you need to make some simple changes to the installation that are similar to what a user install after a
After installation, in the file %PROGRAMFILES%\OpenOffice.org1.1.0\program\bootst
change these lines:
Location=$SYSUSERCONFIG/sversion.ini
and
UserInstallation=${$Location:$Section:$ProductKe
In the file %USERPROFILE%\Application Data\sversion.ini (which will have to be created when it does not yet exist when a new user logs on) change this:
OpenOffice.org 1.1.0=file:///C:/Documents%20and%20settings/Usern
Copy the user directory from the %PROGRAMFILES%\OpenOffice.org1.1.0 to that location when it does not yet exist.
Now the OpenOffice settings will be in the user profile and will roam to computers where this user logs on.
After putting these actions in LOGON scripts, we now have automatic installation and roaming of user data with OpenOffice.org
But I agree that this should be automatic, just as it is with Microsoft Office.
I think there are more things wrong than that.
Like many rebuttal documents, it starts of to-the-point and with facts, and then it wanders off and shows how the writer of the document did not even understand the statement in the original document.
For example, look at the paragraph about "collaboration".
With this, Microsoft want to discuss the calendaring and planning features of Exchange. He completely misses the point.
By coincidence today I filed a bug in Mozilla because it sends the domain name part of the sender mail address as argument to the SMTP EHLO command when sending mail.
This should be the hostname of the sending machine.
As it turns out, this was a duplicate of a bug reported in februari 2001, that is still open!
That bug in fact talked only about HELO so I did not find it using search. SMTP has been replaced by ESMTP in the meantime, but the issue remains.
Not that there were no contributions, there are 110 followups on the original report, many of them with good suggestions and patches. But in december 2003, things have again come to a standstill it seems.
With such issues, that should be trivial to fix in an evening, open for 3 years, I wonder how a change like this got checked in.
Maybe the remarks about OSS made in another thread are sometimes not that far off the mark?
I don't think you got the message.
Worms like this could run on your dedicated firewall box (like linksys or draytek).
They don't require Windows or an insecure OS at all, they just require sloppy programming in any program that handles network packets.
When this SECAM system sucks so bad, then why are most consumer video recorders converting the video to what is essentially SECAM before recording it, and then converting back to NTSC or PAL when playing back?
This is not true at all! PAL B/G (625 lines 25 fps AM negative modulation with FM audio) is compatible with the existing BW standard in the countries where it is used.
What you are referring to is probably the fact that some TV stations used the opportunity of introduction of color for a system change. Some countries were using 405 or 819 lines in BW. A multisystem TV from those days had to be switched, but not because of the BW-color switchover.
It is funny to see how Americans discuss the fact that PAL is better than NTSC (which certainly is true) while in Europe the people who like good quality picture moved from PAL to DVB...
An external DVB-S or other DVB system receiver connected to the TV set using RGB signals is a lot better than PAL.
The "public broadcasting" in the USA is apparently not the same as in the UK (BBC) and other European countries.
While a lot of people watch commercial tv (and proportionally more in lower-educated classes), public broadcasting still has a very significant market share in this region. It isn't elitist at all.
> So, what's at address 0xfffffff0 when there are only, say, 512MB installed?
The BIOS. That is what this article is about.
This used to be a ROM chip but today it is a Flash memory (a nonvolatile memory that is used mostly as a ROM but can be written by a BIOS update utility).
It does not matter how much memory is really in the machine, or how much the motherboard can physically hold.
The processor can address 4GB using 32 address lines, and that defines an address space of 4GB.
Newer processors address 64GB using 4 extra address lines but they are yet another add-on that has to be enabled lateron by special software support.
Yes, the street address is a bug.
I think that the PO Box is shown as the street address, when present.
Will have a look at the mozilla schema...
It is used as an email address lookup, yes.
But it also works from addressbook for some other items.
I am not aware of any special schema, does it exist?
(we use standard schema with inetorgperson etc)
What is wrong? We use Mozilla with LDAP all the time.
I guess you will have to live with the changes that occur over time (sometimes called progress). Today's PCs are powerful enough to run Mozilla, and it seems nobody cares to maintain a browser that still runs on a 486 with 16MB.
Given that the limited resources of developers can be spent only once, I rather have them spent on a browser that can keep up with the market demands, than on a browser for some NetBSD geeks that insist on using a 5 years old machine.
Well, it is often not known what type of OS is running on these boxes. I would not presume it is less fragile. Sure there is no user-installed software on it, but I doubt that is the primary reason why packetfilters fail at some point. More likely this results from design failures.
Some boxes run Linux, and some have taken the firewall functionality from other open source systems (like BSD) and put them in their image as an add-on.
There certainly is room for an exploitable bug in there. And worse: once this is found, it is often considerably more difficult to update the firmware than it is to update a PC-resident piece of software. (at least for the non-tech-savvy user)
Let's at least keep attention on this, not assume "nothing can happen because I have this hardware firewall"...
> Close, but you're just pushing back the problem from a naming of sites to a naming of categories.
Ahh, I did not realize this could be a problem. With the BeNeLux trademark office, there is a fixed list of categories under which you can register your trademarks. So I assumed the category list would be obvious.