KCLUG

kclug@kclug.org

10 participants
3254 discussions

RE: Dell Linux Engineering
by Brian Densmore 07 Dec '04

07 Dec '04

I have no issue with Dell laptops. I have an IBM laptop. But see the TV-OUT thread ( http://kclug.org/pipermail/kclug/2004-December/024710.html ) for why I said what I said about Dell. It was partly tongue-in-cheek, but it sounds like they aren't very good in the support department. Something that isn't important until you need it. I like to test out the support lines for a company before I buy something from them, that way I know what to expect should I ever need help with their product. On a purchase such as a laptop, good support would be #1 in my requirements for a purchase. > -----Original Message----- > From: D. Hageman > > What in particular is your issue with them and with what models? > > I have had several different models of Dell laptops as well > as several > people that I can mention that you would probably recognize.

2 1

Re: Box stops communicating with switch on reboot
by Brian Kelsay 07 Dec '04

07 Dec '04

Did you try on the XP box "ipconfig /release" and "ipconfig /renew"? And on the Linux box "if eth0 up" or "ifconfig eth0", "dhcpcd" or "dhclient" if these are running DHCP. Give some other details. What happens if you just move the cable to another port without restarting the router? How about if you just wait longer than 2 seconds after boot? Does either ever come up? Brian Kelsay >>> Patrick M <> 12/07/04 08:15AM >>> Hi I have a couple of netopia routers with 4/8 port switches. When I take down my linux box and reboot, it won't connect unless I reset the switch. This in itself was livable I am the admin, I just have to remember to reset the switch or have someone else do it if I am not on site. Now, I reloaded an XP box on a linksys switch that is having the same problem. Is there something I am missing on this. do I need to open a port on the computers so the switches can see them easier? They end up working after resetting the switch, not something easy to do or remember every time you or your users reboot.

1 0

Box stops communicating with switch on reboot
by Patrick M 07 Dec '04

07 Dec '04

Hi I have a couple of netopia routers with 4/8 port switches. When I take down my linux box and reboot, it won't connect unless I reset the switch. This in itself was livable I am the admin, I just have to remember to reset the switch or have someone else do it if I am not on site. Now, I reloaded an XP box on a linksys switch that is having the same problem. Is there something I am missing on this. do I need to open a port on the computers so the switches can see them easier? They end up working after resetting the switch, not something easy to do or remember every time you or your users reboot.

1 0

RE: Dell Linux Engineering
by Brian Densmore 07 Dec '04

07 Dec '04

> -----Original Message----- > From: Jeremy Turner > > Hey all, > > In looking up kernel info, I noticed that Dell has a > Dell-related Linux > mailing list page: > > http://lists.us.dell.com/ > http://lists.us.dell.com/mailman/listinfo/ > That's pretty cool, now if they can only get their laptops to be stable...

2 1

Dell Linux Engineering
by Jeremy Turner 06 Dec '04

06 Dec '04

Hey all, In looking up kernel info, I noticed that Dell has a Dell-related Linux mailing list page: http://lists.us.dell.com/ http://lists.us.dell.com/mailman/listinfo/ They include daily-digests of the Linux kernel mailing list, along with some other mailing lists of Dell-specific Linux stuff. I found it interesting. Jeremy

1 0

RE: the venerable 2.0 kernel series
by Brian Densmore 06 Dec '04

06 Dec '04

> -----Original Message----- > From: Jeremy Turner > > > You can get Linux to run on just about anything. It just > depends on how > much time you'd like to spend on it. =) > Actually, I don't think the 2.6 will run on anything lower than 486 or eve Pentium class CPUs. At least for the Intel version. You'd have to go to 2.2 kernel to get 386.

1 0

RE: the venerable 2.0 kernel series
by Brian Densmore 06 Dec '04

06 Dec '04

> -----Original Message----- > From: David Nicol > > what are the advantages of using Linux 2.0 kernels? The series is The advantage to using 2.0 kernels is that you can run applications that require the 2.0.x series kernel. A lot of the 2.2 enhancements were backported to the 2.0 kernel. Some of the 2.4 enhancements were backported to the 2.0 kernel, but I doubt if any of the 2.6 enhancements were backported to the the 2.0 kernel. Another advantage is memory and disk requirements are much smaller. Still another advantage to the 2.0 kernel has been mentioned, size. It's a lot smaller, but not necessarily faster. Size and speed do not necessarily go hand in hand. The 2.6 kernel has some speed enhancing improvements. Although the size is by default much much larger, this can be changed to some degree by compiling a custom kernel and leaving out all of the code for hardware and features you do not use. If you aren't using/emulating scsi devices there is no reason to load scsi support/drivers in the kernel. Also compiling as much of the kernel to load as modules will reduce the kernel size. This is something everyone *should* do with their kernel. Also turning off debugging and stripping debugging code from the kernel and drivers will speed up the kernel. If your system doesn't panic or crash or lock up, there is no reason to have debugging code in there. Sure it's nice to have to give to the designers *should* the system fail. But there is no doubt in my mind that the 2.0 kernel would be faster on older hardware than 2.6. Some statistics on the various kernels in regards to size: Kernel | Source Size Version | (bz2) ===========+============= 1.0 | < 1MB 1.2 | 2.3MB 2.0 | < 6MB 2.2 | 15.4MB (almost tripled) 2.4 | 30.3MB (almost doubled again) 2.6 | 35.4MB So the 2.6 kernel source is almost six times larger in *compressed* size. This is inclusive of *official* modules. I couldn't find any comparison of the compiled size of the kernels. This information taken from the kernel archives. Personally I can't see using the 2.0 except in a classroom. I wouldn't go lower than 2.2. In fact until about 9 months ago I used 2.2 for my webserver. JMHO, Brian Densmore

3 2

RE: the venerable 2.0 kernel series
by Brian Densmore 06 Dec '04

06 Dec '04

> -----Original Message----- > From: Jonathan Hutchins > > On Monday 06 December 2004 09:55 am, Brian Densmore wrote: > > > > Although the size [of later kernels] is by default much much > > larger, this can be changed to some degree by compiling a > custom kernel and > > leaving out all of the code for hardware and features you > do not use. > > I haven't seen a monolithic kernel in all of the time I've > run Linux, starting > pretty early in the RedHat release era (2.1). Once the > kernel went modular, > all of the HOWTO's that started with "first recompile your > kernel" were > obsolete, and so were considerations of slimming down by > leaving out hardware > support. I never said monolithic, but if you go into your RH 2.1 default kernel compile you will see options that are compiled into the kernel (such as SCSI support and one SCSI driver are compiled directly into the kernel, not as modules). While some of the better distros do a good job of modularizing the kernel, there are still some pieces that are compiled in and some that must be compiled in. If you want sound, sound support must be compiled in then the sound drivers can be in modules. If you want to do I2C then support for I2C must be compiled in, and then drivers for I2C devices can be compiled as modules. etc. All I'm saying is if you don't need sound, scsi, I2C, webcams, shortwave, IPv6, etc support turn it off in the kernel. You will then remove the stub code and frequently a default built in module from your kernel. Generally if there is an option for hardware that you don't know what it is you don't need it in the kernel so remove it and make your kernel a lean mean computin' machine. Of course SCSI support probably not one of them if you have a CD, but the default SCSI device for the ACME Roadrunner X9000 SCSI controller could be deselected, which might save 10KB of space. So I say the "first recompile your kernel" is still a good recommendation. Brian D. <disclaimer: I'm not responsible should you remove support for some vital function in the kernel, like support for ext2 filesystems, and trash your machine.>

1 0

RE: TV-OUT on laptop causes major damage, fire / shock
by Brian Densmore 06 Dec '04

06 Dec '04

I say take the free laptop replacement and when it comes sell/trade it for something different. Or are they not really offering to replace it? > -----Original Message----- > From: Thomas Bruno > > I got the following e-mail reply from dell: > > > -----CUT--------- > > Thank you for contacting Dell Technical Support. > > Ronnie, if I have understood it correctly, you are concerned > about the > smoke coming from the Laptop while trying to connect TV > through TV Out > connector and also the RCA cable melted because of this. > > I apologize and truly regret any inconvenience this matter may have > caused. > > Please provide the information requested below so I can setup the > service to arrange for the system to be replaced. > > > -----CUT----------- > > This proves dell e-mail techs can't read, i posted the same story to > them, word for word here, I never said the laptop smoked. I > said the TV > caught on fire. and the wire melted >

5 6

RE: TV-OUT on laptop causes major damage, fire / shock
by Brian Densmore 06 Dec '04

06 Dec '04

> -----Original Message----- > From: Duane Attaway > > OK, I recreated the laptop video out to television experiment. Dude! You are seriously twisted. ;') > > The first video was connecting the RCA cable to house > current. It was > uneventful and promptly blew the 20A house breaker each time. So I > started up the generator and was able to use unfused power. > > Here is a frame by frame of the video of the outer (ground) > conductor of > the video cable melting under 120 volts at 7500 VA: > > http://dattaway.org/whitewire.jpg > > Closeup of melted RCA cable: > > http://dattaway.org/whitewire2.jpg > > Other side: > > http://dattaway.org/whitewire3.jpg > > It took about two seconds of short current to melt the wire > and break the > circuit. It flashed when it broke. > > So what does that mean? It takes a lot of current from a 120 > volt feed to > blow the tiny stranded wires of a RCA video cable. I believe > each one of > the copper strands alone, seperated in air, should take 5 > amps each. I > counted 15 strands. > > Could a capacitor do it as someone suggested? Well it could, > but it would > have to be instantaneous. I have some capacitors up to 8KV > if you want to > see more video and pictures. Well it depends on the type of capacitor. It would have to be a polarized capacitor. I remember when I used to test power supplies for a company I worked at. We would buy them for our products, but they needed to be tested before being installed in our equipment. Every once in a while a PS would have the capacitor in reverse polarity. Quite a show when a big capacitor blows. Less frequently the capacitor would take too much voltage with the same end result, explosion. why do you say it would have to be instantaneous? Although from the description of the event it was instantaneous. > > I'd say his power supply had a minor meltdown inside, causing > the AC to > couple with the DC side. This allowed current to flow from > the AC mains, > through his laptop, and return through the television back > into the wall. > Televisions are designed with a neutral grounded chassis and > this may have > allowed a possibly defective power supply to complete a current loop. I'd tend to agree he's got a current/voltage leak in the PS, which of course could lead to a capacitor explosion. Haven't heard whether the laptop is usable now or if the adaptor is usable. If a cap went the PC or adaptor wouldn't work anymore. He's lucky the TV still works.

4 3

Jump to page:

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

KCLUG