Sounds a lot like PCP, which SGI ported from IRIX to Linux a couple of years ago. I haven't used the Linux version, but the IRIX version of PCP gives point-and-click access to 1000's of kernel metrics.
In any case, it will be a nice addition to Solaris.
Unfortunately, there are still cases where having MS Windoze is necessary.
For example, my wife is a Mac user, but owns Virtual PC because a few of her upper level psychology courses use home-brewed simulation software that only runs under Windows. She avoids MS products like the plague, but the couses were degree requirements, so the options were:
Change majors.
Spend lots of time in the computer lab.
Suck it up, purchase VPC, and run the software on her laptop while sitting out on the deck wirelessly surfing over 802.11b in another window.
Sure, I won't deny that linux has it's problems, but I'd hardly call copy&paste one of them.
Unless of course you've used an OS that supports copy and paste of text, and styled text, and graphics. In which case the cut and paste capabilities of linux are extremely primitive to the point of useless.
Don't get me wrong, I've been successfully cuting and pasting ASCII text in X since before Linux. Cut and Past of ASCII works in X, or you've likely misconfigured something. But MacOS cut and paste is a whole different ballgame, and so far it's still on the ``things to do'' list for Linux and/or X11.
The rabid ``free as in the freedoms that the gpl forces you to give up so that the source will always be available'' crowd has often opined that companies could gpl their software while continuing to generate revenue by selling support (see also Cygnus). My reading of the story is ``we can fix your problem, but our support will cost you $50, or you can figure it out for yourself which only costs you your time'' which is exactly how it is supposed to work.
Yes it sucks that they don't provide free help, but if one is to take the rabid GNU crowd at its word, they are advocating Open Source not communism. If Open Source is going to be compatable with capitalism, people need to get paid somehow.
Alternately, the GPL/LGPL/... are really just about free-as-in-beer and all the scolding and idealism was just a front for people too cheap to buy software and too proud to admit that the BSD license really is better.
I'm using an 802.11b network with 128-bit encryption, meaningless passwords (not "admin" or "router"), and the WAP will recognize only the MAC of the portable (yes, that can be spoofed, but it keeps out random strangers). Finally, the access point is in the basement, so its reception zone is mostly up, not horizontal.
There could be specific weaknesses in my brands of hardware, but that's another problem.
Am I mistaken that this provides reasonably good security?
Assuming your neighbors are clueless luddites who
have to call you when their printer runs out of paper, WEP will prevent them from borrowing you Internet uplink bandwidth. Against a determined attacker, WEP, MAC filtering, and most of the other features built into modern 802.11a/b APs are ineffective.
On the other hand, you may not care.
Eg, my home machines are all secured and I do regular audits and scans. Any sensitive communication (eg, logging into a machine at work) happens over ssh and so is protected. So the only thing a script kiddie can do is watch my web traffic (which he is welcome to do), borrow my bandwidth (which would probably be noticed, and maybe try DoSing my home network (which is easy to fix).
All of the above was also true when my home network was wired. The move to 802.11b just traded a decrease in security for an increase in convenence (ah, reading/. while sitting on the deck).
As Schneier has said, security just buys you time. In the case of 802.11 (or for that matter, any wireless protocol), it takes significantly less time for the security to be breached than it would if the wired protocol was in use. If that worries you, don't use 802.11 networking, cordless phones, or cell phones, or adjust the sensitivity of your traffic to suit the medium.
When they say ``download full classic NES games to the Game Boy Advance'' they mean ``Download a port of an NES or SNES game from the GameCube Animal Crossing game to the GBA''. This has been covered in most of the console gaming publications, print and web.
Nintendo is not shipping a general NES emulator for the GBA. They are producing GC games that can use the GBA as an intelligent peripheral capable of disconnected use. Animal Crossing isn't the first game with this type of GC/GBA connectivity
So unless taco is planning on playing through the GC version of Animal Crossing, he will have to go blind the old fashioned way....
If your message means what I think it does, the answer is no.
An open SMTP port is an entirely different thing from an open mail relay. The first means that a host will (probably) be able to recieve mail. The second means that the host will recieve mail and then send it off to another host.
The mechanism you are proposing would prevent your host from receving mail from any host that could also recieve mail without discriminating between open relays and correctly configured mail servers.
The sendmail homepage has more info on how SMTP functions as well as how to block relay.
I'm sure I left serveral out. Checkpoint-restart has been part of the high-performance computing scene for years. Having been a systdmin on large, high-performance, computing platforms for the last few years of my professional life, my experiences with checkpoint-restart have been a mixed bag. All of the existing systems have limitations. Depending on the application, those limitations can be no problem, or they can be deal-breakers.
Infocom ``z-machine'' version 5. Interpreters are avilable for just about any platform you can think of. As usual, Google has a fairly complete list of what's out there.
Frotz and Zip are popular choices. I personally use Malyon in XEmacs...
The manual seemed fine, but I've been a sysadmin for around a decade, so I'm probably not the audience they were writing for. The SMC manual is about as ``windows centric'' as the Linksys was, if not more so. They occasionally mention Unix, but assume the user knows how to set up lpr printing. As you noted, there are some websites that fill the gap for Mac users.
The documentation included with the SMC is a quick start guide and a CDROM with a PDF version of the compelete manual. Both of those are available on-line on the drivers and downloads page for the 7004AWBR, so it's easy enough to compre for yourself.
The only things that I would consider ``difficult for the average user'' was getting lpr printing to work under MacOS 9 and WEP. Printing is point and shoot for MacOSX, but under 9 you'll need to use the desktop printer utility, not the chooser.
Under windows you'd need to install a driver, but I have an MS-free household so I can't comment on that process.
Getting WEP to work under MacOS (any flavor) depends requires knowing that MacOS Airport stuff uses the password you enter to generate a WEP key. If you want to enter a WEP key directly as a hexdec number, you'll need to prefix it with a $ in the MacOS dialog and without the $ in the SMC setup.
I owned the linksys
BEFW11S4 for about a week , after which I traded it in for the
SMC Barricade 7004AWBR . The SMC is a far nicer product.
Issues with the linksys hub were (in no particular order):
http interface did not work correctly under Netscape or other non-IE browsers.
802.3 and SNAP headers?? Sorry, I don't own any HP equipment....
Big honkin
security hole in the web interface that exposes the router and ISP passwords in cleartext should anyone wish to drive by with an access point....
In addition to having none of the above problems, the SMC also provides a parallel port with an lpr print server (which works fine for my environment of IRIX, Linux, and MacOSX boxen).
Upgrades to the SMC work fine via ftp and http from any of the above operating systems and have added additional features (restricting access to specific MAC addresses, supporting appletalk from wire to wireless, etc).
In short, I found the SMC superior to linksys and the Apple AP, and I'm not alone. After sending a note about my experiences to the author of this review...well, just read the Final update at the end of the page. Several other sysadmins at my site have also purchased the SMC and have similar praise for the unit.
Give scientists bigger computers and they'll tackle larger datasets. Big datasets means ``Big Vizualization'', so there is a branch of the Accellerated Strategic Computing Initiative called
VIEWS that deals with the design of systems and software that can do ``terascale'' visualization. This includes things like bertha displays, large power walls, big SGI systems, terabyte-sized filesystems, and the occasional linux cluster.
[Disclaimer: I'm a sysadmin with VIEWS (and we have an open sysadmin position).]
Mandrake has allowed RieserFS installs since at least version 7.1 in the ``expert'' install mode. [My laptop is currently drake 7.2 with ``RieserFS everwhere'' except for a 20M/boot.]
SGI provides a
Modified Red Hat 7.0 Installer that allows installing linux on top of their XFS filesystem. I haven't tried the XFS under Linux, but I've been impressed with XFS under IRIX. [One of the larger machines I currently admin has several terabyte-sized XFS filesystems.]
Gridware isn't all that new, and it isn't a reaction to Mosix or SETI@home.
Batch systems have been around a long time in the HPC world. Gridware was orginally developed by GENIAS Software GmbH. GENIAS produced a batch scheduler called Codine, which was a commercial version of
DQS. In fact, Sun's
Grid Engine FAQ even states that Sun Grid Engine is a new name for CODINE.
Of course, DQS/Codine/Grid isn't the only batch-scheduling/cycle-scavenging game around. Other players are:
Load Leveler (which IIRC is IBM's commercial implementation derived from Condor)
LSF which is the product Sun was previously co-marketing until they purchased Gridware (probably because of the high per CPU cost of LSF).
and lots of others that I've forgotten, many based on the once-common NQS/NQE batch system.
There are also systems like
Legion that represent a sort of ``next step'' computing enviroment.
Many of these predate newcomers like SETI@home and Mosix by serveral years. Most also provide hooks into parallel computing APIs like
MPI,PVM,
openMP, or
something similar.
Batch scheduling and cycle-scavening are old concepts. Having wasted away my years as a graduate student submitting large quantum chem jobs to Crays, it's nice to see lots of groups continuing to squeeze every useful cycle out of existing hardware. Sun's recent annoucements are just the latest update to an old product---not a new idea, and not a Mosix/SETI rip-off.
Slashdot Mirror
Sounds a lot like PCP, which SGI ported from IRIX to Linux a couple of years ago. I haven't used the Linux version, but the IRIX version of PCP gives point-and-click access to 1000's of kernel metrics.
In any case, it will be a nice addition to Solaris.
Not sure if the games are any good, but I love the idea.
Unfortunately, there are still cases where having MS Windoze is necessary.
For example, my wife is a Mac user, but owns Virtual PC because a few of her upper level psychology courses use home-brewed simulation software that only runs under Windows. She avoids MS products like the plague, but the couses were degree requirements, so the options were:
- Change majors.
- Spend lots of time in the computer lab.
- Suck it up, purchase VPC, and run the software on her laptop while sitting out on the deck wirelessly surfing over 802.11b in another window.
Which would you choose?Unless of course you've used an OS that supports copy and paste of text, and styled text, and graphics. In which case the cut and paste capabilities of linux are extremely primitive to the point of useless.
Don't get me wrong, I've been successfully cuting and pasting ASCII text in X since before Linux. Cut and Past of ASCII works in X, or you've likely misconfigured something. But MacOS cut and paste is a whole different ballgame, and so far it's still on the ``things to do'' list for Linux and/or X11.
I don't see this as ``getting around'' the gpl.
The rabid ``free as in the freedoms that the gpl forces you to give up so that the source will always be available'' crowd has often opined that companies could gpl their software while continuing to generate revenue by selling support (see also Cygnus). My reading of the story is ``we can fix your problem, but our support will cost you $50, or you can figure it out for yourself which only costs you your time'' which is exactly how it is supposed to work.
Yes it sucks that they don't provide free help, but if one is to take the rabid GNU crowd at its word, they are advocating Open Source not communism. If Open Source is going to be compatable with capitalism, people need to get paid somehow.
Alternately, the GPL/LGPL/... are really just about free-as-in-beer and all the scolding and idealism was just a front for people too cheap to buy software and too proud to admit that the BSD license really is better.
Take your pick. :-)
Short answer: Yes, you are mistaken.
Longer answers: Here, here, or here.
Assuming your neighbors are clueless luddites who have to call you when their printer runs out of paper, WEP will prevent them from borrowing you Internet uplink bandwidth. Against a determined attacker, WEP, MAC filtering, and most of the other features built into modern 802.11a/b APs are ineffective.
On the other hand, you may not care.
Eg, my home machines are all secured and I do regular audits and scans. Any sensitive communication (eg, logging into a machine at work) happens over ssh and so is protected. So the only thing a script kiddie can do is watch my web traffic (which he is welcome to do), borrow my bandwidth (which would probably be noticed, and maybe try DoSing my home network (which is easy to fix).
All of the above was also true when my home network was wired. The move to 802.11b just traded a decrease in security for an increase in convenence (ah, reading /. while sitting on the deck).
As Schneier has said, security just buys you time. In the case of 802.11 (or for that matter, any wireless protocol), it takes significantly less time for the security to be breached than it would if the wired protocol was in use. If that worries you, don't use 802.11 networking, cordless phones, or cell phones, or adjust the sensitivity of your traffic to suit the medium.
Take a deep breath and relax...
When they say ``download full classic NES games to the Game Boy Advance'' they mean ``Download a port of an NES or SNES game from the GameCube Animal Crossing game to the GBA''. This has been covered in most of the console gaming publications, print and web.
Nintendo is not shipping a general NES emulator for the GBA. They are producing GC games that can use the GBA as an intelligent peripheral capable of disconnected use. Animal Crossing isn't the first game with this type of GC/GBA connectivity
So unless taco is planning on playing through the GC version of Animal Crossing, he will
have to go blind the old fashioned way....
An open SMTP port is an entirely different thing from an open mail relay. The first means that a host will (probably) be able to recieve mail. The second means that the host will recieve mail and then send it off to another host. The mechanism you are proposing would prevent your host from receving mail from any host that could also recieve mail without discriminating between open relays and correctly configured mail servers.
The sendmail homepage has more info on how SMTP functions as well as how to block relay.
As the poster said, there are plenty of others:
I'm sure I left serveral out. Checkpoint-restart has been part of the high-performance computing scene for years. Having been a systdmin on large, high-performance, computing platforms for the last few years of my professional life, my experiences with checkpoint-restart have been a mixed bag. All of the existing systems have limitations. Depending on the application, those limitations can be no problem, or they can be deal-breakers.
Infocom ``z-machine'' version 5. Interpreters are avilable for just about any platform you can think of. As usual, Google has a fairly complete list of what's out there.
Frotz and Zip are popular choices. I personally use Malyon in XEmacs...
The manual seemed fine, but I've been a sysadmin for around a decade, so I'm probably not the audience they were writing for. The SMC manual is about as ``windows centric'' as the Linksys was, if not more so. They occasionally mention Unix, but assume the user knows how to set up lpr printing. As you noted, there are some websites that fill the gap for Mac users.
The documentation included with the SMC is a quick start guide and a CDROM with a PDF version of the compelete manual. Both of those are available on-line on the drivers and downloads page for the 7004AWBR, so it's easy enough to compre for yourself.
The only things that I would consider ``difficult for the average user'' was getting lpr printing to work under MacOS 9 and WEP. Printing is point and shoot for MacOSX, but under 9 you'll need to use the desktop printer utility, not the chooser. Under windows you'd need to install a driver, but I have an MS-free household so I can't comment on that process.
Getting WEP to work under MacOS (any flavor) depends requires knowing that MacOS Airport stuff uses the password you enter to generate a WEP key. If you want to enter a WEP key directly as a hexdec number, you'll need to prefix it with a $ in the MacOS dialog and without the $ in the SMC setup.
I owned the linksys BEFW11S4 for about a week , after which I traded it in for the SMC Barricade 7004AWBR . The SMC is a far nicer product.
Issues with the linksys hub were (in no particular order):
In addition to having none of the above problems, the SMC also provides a parallel port with an lpr print server (which works fine for my environment of IRIX, Linux, and MacOSX boxen). Upgrades to the SMC work fine via ftp and http from any of the above operating systems and have added additional features (restricting access to specific MAC addresses, supporting appletalk from wire to wireless, etc).
In short, I found the SMC superior to linksys and the Apple AP, and I'm not alone. After sending a note about my experiences to the author of this review...well, just read the Final update at the end of the page. Several other sysadmins at my site have also purchased the SMC and have similar praise for the unit.
killall under Solaris will terminate all processes, not kill processes by name. Add a linux sysadmin and hilarity ensues.
Me too.
Some minor tidbits:
- Yes, LLNL does have at least one of the ``big bertha'' displays.
- We also have a lot of computing power (including the
number 1 and number 4 machines on the current
Top500 supercomputers list.
Give scientists bigger computers and they'll tackle larger datasets. Big datasets means ``Big Vizualization'', so there is a branch of the Accellerated Strategic Computing Initiative called VIEWS that deals with the design of systems and software that can do ``terascale'' visualization. This includes things like bertha displays, large power walls, big SGI systems, terabyte-sized filesystems, and the occasional linux cluster.[Disclaimer: I'm a sysadmin with VIEWS (and we have an open sysadmin position).]
So:
Did I miss something?
Mandrake has allowed RieserFS installs since at least version 7.1 in the ``expert'' install mode. [My laptop is currently drake 7.2 with ``RieserFS everwhere'' except for a 20M /boot.]
SGI provides a Modified Red Hat 7.0 Installer that allows installing linux on top of their XFS filesystem. I haven't tried the XFS under Linux, but I've been impressed with XFS under IRIX. [One of the larger machines I currently admin has several terabyte-sized XFS filesystems.]
Batch systems have been around a long time in the HPC world. Gridware was orginally developed by GENIAS Software GmbH. GENIAS produced a batch scheduler called Codine, which was a commercial version of DQS. In fact, Sun's Grid Engine FAQ even states that Sun Grid Engine is a new name for CODINE.
Of course, DQS/Codine/Grid isn't the only batch-scheduling/cycle-scavenging game around. Other players are:
Many of these predate newcomers like SETI@home and Mosix by serveral years. Most also provide hooks into parallel computing APIs like MPI, PVM, openMP, or something similar.
Batch scheduling and cycle-scavening are old concepts. Having wasted away my years as a graduate student submitting large quantum chem jobs to Crays, it's nice to see lots of groups continuing to squeeze every useful cycle out of existing hardware. Sun's recent annoucements are just the latest update to an old product---not a new idea, and not a Mosix/SETI rip-off.