This was discussed on Linux-kernel just last week.
Question: Is there Linux support for the hardware
checksumming?
Answer: 3c590, 3c900, 3c905, 3c575 - they all have hardware support for Tx and
Rx checksums - UDP and TCP. Many other NICs do as well.
Linux uses the hardware checksum on receive, but not on transmit. The
cost of doing the Tx checksum is small compared to the cost of the
mem->mem copy, which is performed at the same time.
It's a great way to improve your Spanish reading ability, even if you're starting with none at all. Just grab a dictionary and start building your lexicon. You'd be surprised how quickly you can achieve 80 to 90 percent comprehension.
With the demographic changes expected in the US in the comming decades, being able to at least read Spanish is probably a good skill to pick up. And Barrapunto.com generally has a good set of stories which complement those on Slashdot. And instead of posting as an AC, you can post as a Pendejo Sin Nombre.
SPECweb99 is the next-generation SPEC benchmark for evaluating the performance of World Wide Web Servers. As the successor to SPECweb96, SPECweb99 continues the SPEC tradition of giving Web users the most objective and representative benchmark for measuring a system's ability to act as a web server. In response to rapidly advancing Web technology, the SPECweb99 benchmark includes many sophisticated and state-of-the-art enhancements to meet the modern demands of Web users of today and tomorrow:
Standardized workload, agreed to by major players in WWW market
Full disclosures available on this web site
Stable implementation with no incomparable versions
Measurement of simultaneous connections rather than HTTP operations
Simulation of connections at a limited line speed
Dynamic GETs, as well as static GETs; POST operations.
Keepalives (HTTP 1.0) and persistent connections (HTTP 1.1).
Dynamic ad rotation using cookies and table lookups.
File accesses more closely matching today's real-world web server access patterns.
An automated installation program for Microsoft Windows NT as well as Unix installation scripts.
Last year SGI sponsored a Linux University road tour, and one of the sessions was given by Intel, where the IA-64 roadmap the next few years was discussed at some length. The dek^Hsktop version of IA-64 is codenamed Deerfield and is scheduled for 2003 sometime.
I asked the Intel guy "Will Sledgehammer speed up the development of Deerfield?", and he got that "Deer in the headlights" look for a few seconds, and then replied that "competition always results in better products sooner than later".
I hope that Sledgehammer causes Intel to put Deerfield on the front burner.
I've read some well written posts which point out that open source may not be enough and that open design (what can we expect coming down the road) may be a good idea.
In this response, Al Viro's point 2 reads: 2. Keep your attributions straight, I don't think that journalling is a good idea at all. Personally, I prefer soft-updates.
If I understand the context, Al Viro does not support journalling. Could all those journalling fs authors and design teams have gotten it so wrong? Or, is one man's agenda making a complex situation unecessarily more difficult.
If I've misunderstood the context of your post, Al Viro, I apologize in advance. If, however, you really don't think that journalling is a good idea, then could you explain why and save the ext3fs, XFS, JFS, ReiserFS, et al developers some time while they continue to go down a blind alley.
The "planned power outages" were just turning the boxes off to see what happened. "Seeing is Believing", and I wanted to see what ReiserFS would really do. The longest that ReiserFS took was 18 seconds on a 27 GB Maxtor IDE 7200 rpm drive.
And, about 10 seconds after I made the "Grassroots" post, my two year old toddler managed to perform an "unplanned power outage" while I was distracted reading my post. In this case, ReiserFS reported "Replayed 12 transactions in 5 seconds", and this was on a 13 GB Western Digital IDE drive. Having the power button on the front of the computer is all too convenient for curious children, but ReiserFS makes this less of a problem than before.
I'd really like to see ReiserFS merged with the 2.4.0 kernel. I've been using ReiserFS 3.5.19 made available with the Linux-Mandrake 7.1 release, and I'm very happy with it. ReiserFS has survived several planned power outages and just keeps on ticking. By the way, the Linux-Mandrake installer even sets up the partition on which/boot exists with the notails option in/etc/fstab.
A journaling filesystem is a very high profile Killer Feature. Having journaling in 2.4.0 would make Linux an even more obvious choice where data integrity is of paramount importance.
Lets start a grassroots movement to have ReiserFS merged with 2.4.0!
While waiting for Slashdot to start responding again, I messed up the last paragraph of my previous comment. And then, I didn't look closely enough at the preview. The lost sentence is in italics.
The last paragraph should have read:
A brute force and sometimes quicker way of doing an upgrade is just to install over the old system, saving the/home directory by not formatting it and carefully saving other files you tweaked elsewhere. When I upgraded from Caldera 2.3 to Mandrake 7.0, I decided on this approach. I planned on saving the log files in/var/log by not formatting/var. Well, the 7.0 installer wouldn't let me get away with not formatting/var. Next time, I'll save my logs elsewhere.
I downloaded the ISO images last week and installed 7.1 on the following two systems:
A Dell 420 dual 733 P-III at work.
A Gateway 450 P-III at home.
The first thing I noticed is the Bogomips values reported at boot time are doubled for both machines over previous distributions, including Mandrake 7.0, Caldera 2.2 and 2.3 and RedHat 6.0 and 6.1.
My Dell box at work is no longer the strong but silent type; the sound card worked on the first boot up. With previous distibutions and also on an older Dell Gxi, it was always an annoying exercise to get sound working. I never had any trouble with the sound on the Gateway.
My main reason for installing 7.1 was to try the ReiserFS. I've set up both systems with ReiserFS on all partitions. I decided to give the ReiserFS a good test and after umounting the filesystems on a separate and still ext2 disk, I just pulled the plug and rebooted. It was the fasted reboot after a power outage that I've ever seen. Reiser took 18 seconds to do its thing on a 17 MB IDE disk, and then the system came up just perfectly.
I really like the new, improved menu system. So far, this is the best distribution I've used, and my Linux experience goes back to kernel 0.99 on Slackware.
This has undoubtably been mentioned elsewhere, but if you want to get Xfree86 4.0, you have to do the expert install.
Also, I usually make a separate/boot partition of about 20 MB, but when I tried to format it using ReiserFS, I got an error. I gave up and just made separate/,/usr,/var, and/home partitions of appropriate sizes and everthing worked fine. I made all these partitions ReiserFS. Does anyone have any suggestions about what an optimal setup might be? Could some partitions better be left ext2?
A brute force and sometimes quicker way of doing an upgrade is just to install over the old system, saving the/home directory by not formatting it and carefully saving other files you tweaked elsewhere. I planned on saving the log files in/var/log by not formatting/var. Well, the installer wouldn't let me get away with not formatting/var. Next time, I'll save my logs elsewhere.
Tycho Brahe was able to make fantastically accurate measurements with the naked eye. His super eyesight has been estimated to have been at least 20/10 and perhaps as good as 20/5.
I remember how my perception of the world changed when I first put on glasses; suddenly trees had leaves! Imagine how this kind of super sight would enable natural scientists to better observe nature.
A few months ago the Los Alamos National Laboratory magazine "BITS" featured an article on some of the people who now work for TurboLabs, and who were then working at LANL's Advanced Computing Lab. They produced a machine called Rockhopper, which had 128 nodes of dual-processor PIII machines IIRC. The networking architecture was characterized as a "superset of Beowulf", with much higher speed connections than mere 100Mbps ethernet and other architectural improvements. I believe their goal was to scale up this SuperBeowulf type of machine to much greater levels of performance. Then they quit the Lab and formed TurboLabs. I look forward to seeing some results someday soon from such a capable team.
I sincerely hope these folks are doing fine since we all got run out of town on May 10th by that damn fire.
Parallel computing: Beowulf is one example of the parallel-computing cluster, executing parallel applications written for libraries like the Message Passing Interface (MPI). Used mostly in scientific computing.
High availability: Dedicated to maintaining a high level of overall system availability, these clusters replace failing nodes with backup nodes as quickly as possible, so that in the worst case a node failure only creates a few seconds of downtime.
Load balancing: To improve overall system performance, the load-balancing cluster shares the application workload of all system users across the nodes of the cluster.
Resource sharing: This kind of cluster combines system resources into a central system or service so they can be accessed by all nodes of the cluster or by other computers.
Single system image: Here, the operating systems of several nodes are combined in a single system, so that it appears that all the users are running on a single large machine. This can ease system management and make it simpler to run applications for load balancing.
Although it's theoretically possible to build all these aspects into a single operating system, the practical issues of doing so are incredibly complex. While few operating system vendors have attempted to cover all the bases described above, one shining example does stand out in Compaq Tru64 Unix.
End of article snip...
If Compaq would only make the "Open" in OpenVMS mean GPL, then there could be some more great code to borrow.
At 3 PM today we got the word that White Rock residents could return. White Rock was threatened by the Cerro Grande fire, but due to heroic efforts, it was saved and suffered no damage. Of course, Los Alamos was not nearly so lucky as the coverage has clearly shown.
Wednesday afternoon, after "The Hill" was given the evacuation order, I had thirteen evacuees in my home in White Rock. They left for Albuquerque later that evening. I assured them that White Rock was safe, but they thought otherwise. You can imagine my surprise when the evacuation order came about 00:35 Thursday morning. We spent about 30 minutes packing up two vehicles, and then we were stuck for the next four hours in a massive traffic jam as 7,000 residents and 7,000 recent evacuees tried to get out on the one safe lane out of town. It was rather frightening when the birds started singing in the middle of the night, as the approaching fire simulated an early dawn.
Thanks to all the people who contributed to the Red Cross. And many, many thanks to the heroic efforts of the Fire crews who saved both White Rock and the major part of Los Alamos. Without their efforts, it would have been much, much worse.
Back in the good old days, people could sit on their front porch and chat with their neighbors as they rode by in their horse and buggy. The slow speed of the horse and the open buggy made this possible.
Today, we drive around with the windows rolled up and at speeds that make conversation with the neighbors impossible. So, the front porch is an icon of the past. A shame, but life goes on.
My "problem" with the Linux servers I have is that I never have to touch them and I lose my Linux skills from lack of practice! While this is not a legitimate complaint, I still nonetheless need to play (and screw-up;) on a non-production test machine to keep my Linux skills honed.
I know that Linux reliability can make the average Linux administrator feel like the Maytag repairman, so find yourself a Pentium-90 or equivalent gathering dust somewhere, install your favorite distro on it, and beat it to death. When you get bored with that, install another distro which you haven't used before and repeat. Install everything that you can find, and play, play, play. Find old hardware or borrow new hardware, install it and observe the results. Take lots of notes. Six months from now you might not remember how you resolved some tricky problem, but your notes will remember. Don't let the uptime on the crash test dummy exceed one week. Download new versions of the kernel and recompile often. Enjoy.
"How is it that you see the splinter in your brother's eye and are not aware of the beam in your own eye?"
With the police knocking down doors in New York at the behest of Microsoft to find a package that Microsoft themselves misaddressed, it seems that you have a pretty big plank in your own eye.
Do you think your Department of Justice is there to protect you? The Digital Millennium Copyright Act protects the few and powerful at the expense of the many.
Do you think you have the right to a fair trial and protection from cruel and unusual punishment? Wen Ho Lee has been held without bail in solitary confinement with his arms and legs shackeled, and this was after the US DOE said he passed his polygraph examination. The FBI "reinterpreted" the results, and Lee is now imprisoned without a trial. Sure, he'll get a trial someday, but only after he's rotted in the hole for good long while.
The road to hell is paved with good intentions. While Russia's destination on the path they're taking is obvious to most slashdot readers, the "good intentions" are obscure. Tax evasion is rampant in Russia, and the government needs money. This latest outrage may be nothing more than an ill-advised attempt to collect more taxes. But its probably more sinister, judging from the past.
Back at the ranch, you should get your own house in order and stop worrying about the affairs of others.
A sated man cannot understand a hungry man.
on
NASA Gets Smart
·
· Score: 1
Just as children growing up in a normal family have a hard time understanding the problems of children growing up with alchoholic parents, Americans are going to have difficulty understanding the problems confronting the Russian people.
Sometimes tough love is the best kind of love for a dysfunctional family, which is what Russia is now.
Giving money to a thoroughly corrupt and criminal system is the same as giving a liquor store gift certificate to an alchoholic. It only empowers the evil.
Perhaps someday Russia will improve, but it won't be soon.
...trying to get Arcserve to work on Netware with Win9x and WinNT clients.
I realize that this post is pretty redundant, but we spent so much precious time and energy trying to get that POS to work, that I just had to get my 0.02 Euros in.
Linux community, go invest in LinuxOne before you waste any time with Arcserve. Its truly a POS.
Title: Maximum Linux Security Publisher: SAMS Publishing Author: Anonymous ISBN: 0-672-31670-6 LOCCCN: 99-61434 First Printing: September 1999 Pages: 743
Micro-Review: Definitely worth buying, reading, and implementing.
Slashdot Mirror
Question: Is there Linux support for the hardware checksumming?
Answer: 3c590, 3c900, 3c905, 3c575 - they all have hardware support for Tx and Rx checksums - UDP and TCP. Many other NICs do as well.
Linux uses the hardware checksum on receive, but not on transmit. The cost of doing the Tx checksum is small compared to the cost of the mem->mem copy, which is performed at the same time.
You can read this thread in the archives here.
Meanwhile, consider the following:
- Novell bought Unix from ATT
- Ray Noorda was CEO of Novell
- SCO bought Unix from Novell
- Ray Noorda owns 83.5% of Caldera
Maybe Mr. Noorda just wants to get Unix back at a fire sale price.www.barrapunto.com offers stuff that matters in Spanish.
It's a great way to improve your Spanish reading ability, even if you're starting with none at all. Just grab a dictionary and start building your lexicon. You'd be surprised how quickly you can achieve 80 to 90 percent comprehension.
With the demographic changes expected in the US in the comming decades, being able to at least read Spanish is probably a good skill to pick up. And Barrapunto.com generally has a good set of stories which complement those on Slashdot. And instead of posting as an AC, you can post as a Pendejo Sin Nombre.
SPECweb99 is the next-generation SPEC benchmark for evaluating the performance of World Wide Web Servers. As the successor to SPECweb96, SPECweb99 continues the SPEC tradition of giving Web users the most objective and representative benchmark for measuring a system's ability to act as a web server. In response to rapidly advancing Web technology, the SPECweb99 benchmark includes many sophisticated and state-of-the-art enhancements to meet the modern demands of Web users of today and tomorrow:
ALongLongWayFromHome pointers.
This is like putting lipstick on a pig.
I asked the Intel guy "Will Sledgehammer speed up the development of Deerfield?", and he got that "Deer in the headlights" look for a few seconds, and then replied that "competition always results in better products sooner than later".
I hope that Sledgehammer causes Intel to put Deerfield on the front burner.
Thou shalt not reverse engineer the genome of man or any other creature.
In this response, Al Viro's point 2 reads:
2. Keep your attributions straight, I don't think that journalling is a good idea at all. Personally, I prefer soft-updates.
If I understand the context, Al Viro does not support journalling. Could all those journalling fs authors and design teams have gotten it so wrong? Or, is one man's agenda making a complex situation unecessarily more difficult.
If I've misunderstood the context of your post, Al Viro, I apologize in advance. If, however, you really don't think that journalling is a good idea, then could you explain why and save the ext3fs, XFS, JFS, ReiserFS, et al developers some time while they continue to go down a blind alley.
And, about 10 seconds after I made the "Grassroots" post, my two year old toddler managed to perform an "unplanned power outage" while I was distracted reading my post. In this case, ReiserFS reported "Replayed 12 transactions in 5 seconds", and this was on a 13 GB Western Digital IDE drive. Having the power button on the front of the computer is all too convenient for curious children, but ReiserFS makes this less of a problem than before.
Read the testimonials on the ReiserFS homepage.
A journaling filesystem is a very high profile Killer Feature. Having journaling in 2.4.0 would make Linux an even more obvious choice where data integrity is of paramount importance.
Lets start a grassroots movement to have ReiserFS merged with 2.4.0!
The last paragraph should have read:
A brute force and sometimes quicker way of doing an upgrade is just to install over the old system, saving the /home directory by not formatting it and carefully saving other files you tweaked elsewhere. When I upgraded from Caldera 2.3 to Mandrake 7.0, I decided on this approach. I planned on saving the log files in /var/log by not formatting /var. Well, the 7.0 installer wouldn't let me get away with not formatting /var. Next time, I'll save my logs elsewhere.
The first thing I noticed is the Bogomips values reported at boot time are doubled for both machines over previous distributions, including Mandrake 7.0, Caldera 2.2 and 2.3 and RedHat 6.0 and 6.1.
My Dell box at work is no longer the strong but silent type; the sound card worked on the first boot up. With previous distibutions and also on an older Dell Gxi, it was always an annoying exercise to get sound working. I never had any trouble with the sound on the Gateway.
My main reason for installing 7.1 was to try the ReiserFS. I've set up both systems with ReiserFS on all partitions. I decided to give the ReiserFS a good test and after umounting the filesystems on a separate and still ext2 disk, I just pulled the plug and rebooted. It was the fasted reboot after a power outage that I've ever seen. Reiser took 18 seconds to do its thing on a 17 MB IDE disk, and then the system came up just perfectly.
I really like the new, improved menu system. So far, this is the best distribution I've used, and my Linux experience goes back to kernel 0.99 on Slackware.
This has undoubtably been mentioned elsewhere, but if you want to get Xfree86 4.0, you have to do the expert install.
Also, I usually make a separate /boot partition of about 20 MB, but when I tried to format it using ReiserFS, I got an error. I gave up and just made separate /, /usr, /var, and /home partitions of appropriate sizes and everthing worked fine. I made all these partitions ReiserFS. Does anyone have any suggestions about what an optimal setup might be? Could some partitions better be left ext2?
A brute force and sometimes quicker way of doing an upgrade is just to install over the old system, saving the /home directory by not formatting it and carefully saving other files you tweaked elsewhere. I planned on saving the log files in /var/log by not formatting /var. Well, the installer wouldn't let me get away with not formatting /var. Next time, I'll save my logs elsewhere.
I remember how my perception of the world changed when I first put on glasses; suddenly trees had leaves! Imagine how this kind of super sight would enable natural scientists to better observe nature.
# shutdown -h mow
That must be tomorrow
Can't you wait till then?
A few months ago the Los Alamos National Laboratory magazine "BITS" featured an article on some of the people who now work for TurboLabs, and who were then working at LANL's Advanced Computing Lab. They produced a machine called Rockhopper, which had 128 nodes of dual-processor PIII machines IIRC. The networking architecture was characterized as a "superset of Beowulf", with much higher speed connections than mere 100Mbps ethernet and other architectural improvements. I believe their goal was to scale up this SuperBeowulf type of machine to much greater levels of performance. Then they quit the Lab and formed TurboLabs. I look forward to seeing some results someday soon from such a capable team.
I sincerely hope these folks are doing fine since we all got run out of town on May 10th by that damn fire.
From the article:
There are several different types of clustering:
Although it's theoretically possible to build all these aspects into a single operating system, the practical issues of doing so are incredibly complex. While few operating system vendors have attempted to cover all the bases described above, one shining example does stand out in Compaq Tru64 Unix.
End of article snip...
If Compaq would only make the "Open" in OpenVMS mean GPL, then there could be some more great code to borrow.
Wednesday afternoon, after "The Hill" was given the evacuation order, I had thirteen evacuees in my home in White Rock. They left for Albuquerque later that evening. I assured them that White Rock was safe, but they thought otherwise. You can imagine my surprise when the evacuation order came about 00:35 Thursday morning. We spent about 30 minutes packing up two vehicles, and then we were stuck for the next four hours in a massive traffic jam as 7,000 residents and 7,000 recent evacuees tried to get out on the one safe lane out of town. It was rather frightening when the birds started singing in the middle of the night, as the approaching fire simulated an early dawn.
Thanks to all the people who contributed to the Red Cross. And many, many thanks to the heroic efforts of the Fire crews who saved both White Rock and the major part of Los Alamos. Without their efforts, it would have been much, much worse.
Today, we drive around with the windows rolled up and at speeds that make conversation with the neighbors impossible. So, the front porch is an icon of the past. A shame, but life goes on.
I know that Linux reliability can make the average Linux administrator feel like the Maytag repairman, so find yourself a Pentium-90 or equivalent gathering dust somewhere, install your favorite distro on it, and beat it to death. When you get bored with that, install another distro which you haven't used before and repeat. Install everything that you can find, and play, play, play. Find old hardware or borrow new hardware, install it and observe the results. Take lots of notes. Six months from now you might not remember how you resolved some tricky problem, but your notes will remember. Don't let the uptime on the crash test dummy exceed one week. Download new versions of the kernel and recompile often. Enjoy.
unsigned short bugs_r_us;
the bug counter will soon go from 65535 to 0.
"How is it that you see the splinter in your brother's eye and are not aware of the beam in your own eye?"
With the police knocking down doors in New York at the behest of Microsoft to find a package that Microsoft themselves misaddressed, it seems that you have a pretty big plank in your own eye.
Do you think your Department of Justice is there to protect you?
The Digital Millennium Copyright Act protects the few and powerful at the expense of the many.
Do you think you have free speech? Check out this recent CNN story:
Park's Jesus statue is unconstitutional -- even on private land.
Do you think you have the right to a fair trial and protection from cruel and unusual punishment? Wen Ho Lee has been held without bail in solitary confinement with his arms and legs shackeled, and this was after the US DOE said he passed his polygraph examination. The FBI "reinterpreted" the results, and Lee is now imprisoned without a trial. Sure, he'll get a trial someday, but only after he's rotted in the hole for good long while.
The road to hell is paved with good intentions. While Russia's destination on the path they're taking is obvious to most slashdot readers, the "good intentions" are obscure. Tax evasion is rampant in Russia, and the government needs money. This latest outrage may be nothing more than an ill-advised attempt to collect more taxes. But its probably more sinister, judging from the past.
Back at the ranch, you should get your own house in order and stop worrying about the affairs of others.
Sometimes tough love is the best kind of love for a dysfunctional family, which is what Russia is now.
Giving money to a thoroughly corrupt and criminal system is the same as giving a liquor store gift certificate to an alchoholic. It only empowers the evil.
Perhaps someday Russia will improve, but it won't be soon.
You can read Linux news in Russian and see Tux wearing a suit of armor at www.linux.ru.net.
I realize that this post is pretty redundant, but we spent so much precious time and energy trying to get that POS to work, that I just had to get my 0.02 Euros in.
Linux community, go invest in LinuxOne before you waste any time with Arcserve. Its truly a POS.
Title: Maximum Linux Security
Publisher: SAMS Publishing
Author: Anonymous
ISBN: 0-672-31670-6
LOCCCN: 99-61434
First Printing: September 1999
Pages: 743
Micro-Review: Definitely worth buying, reading, and implementing.