[expert] tv tuner, sound card, ethernet won't work together
Hi all. I've got a bt848 tuner, soundblaster live 5.1, and 3com cyclone ethernet card on an Asus nforce2 board (A7N8X) with a Geforce2mx. On board sound and ethernet are disabled in the bios. If the tv tuner is not in use (i.e. I haven't started xawtv or motv) everything's fine. A few seconds after I start up motv/xawtv either the tv tuner, sound card or ethernet will stop working. The tuner's image will freeze, the sound card will start playing static or the ethernet just won't connect (I'm on broadband). I'm running mandrake 9.1 with a stock kernel with these parameters: hdc=ide-scsi hdd=ide-scsi pci=noapic noapic acpi=off Not sure if the pci=noapic does anything worthwhile, but my system is completely useless without the acpi=off line. Moving the card around seems to change which device goes dead, but no matter where I put it one card will go dead. There are no IRQ conflicts, and I don't think there are any I/O conflicts (but my Hex isn't good enough to be sure). I'm pretty sure the card is good (worked on my old board, and will work fine when another card goes dead). Reloading the driver doesn't help, only rebooting. I have no idea what the cause/solution might be. Can someone with more hardware experience give me some insight? Thanks again :). Jeremy Gregorio Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [expert] tv tuner, sound card, ethernet won't work together
On Wed, 2003-06-18 at 03:58, Jeremy Gregorio wrote: Hi all. I've got a bt848 tuner, soundblaster live 5.1, and 3com cyclone ethernet card on an Asus nforce2 board (A7N8X) with a Geforce2mx. On board sound and ethernet are disabled in the bios. If the tv tuner is not in use (i.e. I haven't started xawtv or motv) everything's fine. A few seconds after I start up motv/xawtv either the tv tuner, sound card or ethernet will stop working. The tuner's image will freeze, the sound card will start playing static or the ethernet just won't connect (I'm on broadband). I'm running mandrake 9.1 with a stock kernel with these parameters: hdc=ide-scsi hdd=ide-scsi pci=noapic noapic acpi=off Not sure if the pci=noapic does anything worthwhile, but my system is completely useless without the acpi=off line. Moving the card around seems to change which device goes dead, but no matter where I put it one card will go dead. There are no IRQ conflicts, and I don't think there are any I/O conflicts (but my Hex isn't good enough to be sure). I'm pretty sure the card is good (worked on my old board, and will work fine when another card goes dead). Reloading the driver doesn't help, only rebooting. I have no idea what the cause/solution might be. Can someone with more hardware experience give me some insight? Thanks again :). Jeremy Gregorio does your video card have an irq reserved in bios? Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [expert] MP3 to Wave
On Sunday August 17 2003 08:18 pm, David E. Fox wrote: I highly recommend lame as in lame --decode mostexcellent.mp3. Only problem is that it doesn't handle converting more than one at a time. Rob There's always shell loops :). Or, try the diskwriter plugin for xmms. Select the plugin via ctrl-v, then select your files, and press play. It does all of them in one fell swoop (tm). That's my preferred converter. Plus if you highlight all the files in the playlist, xmms gives the total time down on the bottom of the window. Very useful for jugglin 'em around to fit on an 80 minute CDr. -- Tom Brinkman Corpus Christi, Texas Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [expert] Cups web interface files missing
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Larry: That's the odd thing. I reinstalled the cups rpms several times and still the files does not show. It happens with the cups rpm in the Mandrake 9.1 as well as the last one (cups-1.1.19-1.1mdk) downloaded from http://distro.ibiblio.org/pub/linux/distributions/mandrake/Mandrake/updates/9.1/RPMS;. After I manually copied the /usr/share/doc/cups files from my workstation, I decided to retry reinstalling all cups packages again (included ghostscript, requiered by cups-drivers). The problem still persists, but one thing I noted is that while rpm -Uhv cups-1.1.19-1.1mdk does not install these files, rpm -e cups does not delete them either. I had to delete the /usr/share/doc/cups directory manually after removing the packages and before installing them again. Normally, rpm -e cups remove the complete directory automatically. To me that means rpm knows it is not installing these files, so it does not need to remove them. The problem is why it is deciding not to install them? El Sunday 17 August 2003 13:44, Larry Sword escribió: Carlos: It is odd that these files did not install as they are part of the cups-1.1.19-1.1mdk. You might try re-installing this rpm to ensure that the rpm database is correct for any future updates. Larry -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Larry: Thanks for your replay. These are the very same programs I have installed: [EMAIL PROTECTED] root]# rpm -qa | grep cups cups-1.1.19-1.1mdk libcups1-1.1.19-1.1mdk cups-common-1.1.19-1.1mdk cups-drivers-1.1-104.2mdk The weird thing is that in my workstation I have exactly the same programs and here I have the /usr/share/doc/cups files. I have resolved the problem (at least temporally), copying all the /usr/share/doc/cups directory from my workstation to the print server and now everything seems to be OK. While this solved the problem now, I still have some concerns on what is going to happen when I have to update any future cups release. I think the problem resides in the cups rpm configuration. Maybe it requieres another package to be installed in order to install these files. Anyway urpmi has never reports any error. El Saturday 16 August 2003 19:02, Larry Sword escribió: Carlos A. Siso wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Greetings: I am trying to setup a Cups print server using Mandrake 9.1. I installed 9.1 from CD's without any additional packages and installed Cups using the urpmi command urpmi cups. While the cups server is running, the problem is that the Cups web interface files at /usr/share/doc/cups are not installed, so there is no access to the print server using http://server:631;. I have reinstalled cups a few times (rpm -e cups; urpmi cups) and still there are no files at /usr/share/doc/cups. The strange thing is this: [EMAIL PROTECTED] cups]# rpm -qf /usr/share/doc/cups/index.html cups-1.1.19-1.1mdk [EMAIL PROTECTED] cups]# ls -la /usr/share/doc/cups/ total 16 drwxr-xr-x4 root root 4096 Aug 16 14:29 ./ drwxr-xr-x 17 root root 4096 Aug 16 14:29 ../ drwxr-xr-x2 root root 4096 May 28 16:15 de/ drwxr-xr-x2 root root 4096 May 28 16:15 fr/ What am I missing here? Any help woul be appreciated. TIA. Runnig fine here. These are the programs I have installed. # rpm -qa | grep cups libcups1-1.1.19-1.1mdk cups-common-1.1.19-1.1mdk cups-drivers-1.1-104.2mdk cups-1.1.19-1.1mdk Larry -- Abit VP-6 Dual Pentium III 1GHz Mandrake 9.1 Kernel 2.4.21-0.25mdksmp - -- Carlos A. Siso Hyper Sistemas s.a. GPG Key: E762AD4D http://www.hypersistemas.com/keys/casiso.asc -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.2 (GNU/Linux) iD8DBQE/P7ll+tvDo+dirU0RAl27AJwProezPZreNCmIbZfe1Mr2LcFxwQCgkrxm KhUgSWbsh/DnOMjIZKbHgPA= =cZ7C -END PGP SIGNATURE- Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com -- Abit VP-6 Dual Pentium III 1GHz Mandrake 9.1 Kernel 2.4.21-0.25mdksmp - -- Carlos A. Siso Hyper Sistemas s.a. GPG Key: E762AD4D http://www.hypersistemas.com/keys/casiso.asc -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.2 (GNU/Linux) iD8DBQE/QN1f+tvDo+dirU0RAiMWAKCgIJC8U1+WS8XZOI5dfgdlZhEz+QCfXXDY Hr5e8plwYeSaVlp/k7jzfLQ= =AWhk -END PGP SIGNATURE- Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [expert] OT: router reboots
On Monday 18 Aug 2003 9:03 pm, Lyvim Xaphir wrote: On Thu, 2003-08-14 at 07:52, Anne Wilson wrote: SMC Barricade 7401BRA adsl router On average, about once a week I have to reboot my router. No apparent reason - just that no Internet connection works any more. After a reboot, all is well, both on this box and all others on the net. I am not able to find any common denominator leading up to the problem. Any ideas, anyone? Anne Does it have firmware that you can update. I'd check on that first. --LX Hi, LX. Franki raised that one, but my reply to him was one of the ones that went missing, never appeared on the list. Here is what I said to him: Hi, Franki. A little while ago I downloaded the latest flash for the router and the installation instructions, which were to flash it from within the web page. 'Mine not to reason why'. I thought, and tried it. It didn't work. I contacted their tech support, who said they would email alternative instructions. They did, and it didn't work. It did screw up everything that had been there before, though, so I had to go in an reconfigure it. There was a complication in that I couldn't get a connection, which my isp traced to a line fault, which BT denied, but a short time after it was working again. Then I rang SMC's tech support again. This time a guy asked me why I was trying to flash it. He rather rudely said it was pointless - it was not necessary with the revision I had got. Between that experience and the fact that it is impossible to release a range of ports for a session (ideally I would like it to be software triggered, but I'll do it manually if necessary) I feel that when I replace it, it will not be with an SMC product. If I may, I will also copy in my replies to Ronald and Miark, which also went missing. Ronald: Hmm, I wonder if this has anything to do with lease time. My Dlink 604 uses a web page for configuration, and at one point you can adjust the lease time from a single hour up to a week. I have it set for a week, but I've never went longer than that to see if there is a lease issue. When booting up from any of my other 2 comps on the same LAN (and going thru the router), I get a boot up message to the effect that infinite lease time is used. Can any one else bring light to the murkiness here? :-) Lease time to me implies dhcp, which I don't think is the issue. I have a static ip both externally and internally - all our lan uses static. Miark: Instead of re-starting the router, try re-starting the NIC. If that works, you can set up a cronjob to check and/or fix the problem on a regular basis. Miark, this affects all boxes on the lan, so it's not the local nic. The router is a black box. Do you mean the nic on the router? We have tried an ftp session on it, and it seems to be running some sort of windows system. I suppose it would be possible to work out a way of doing it. I'll have a think about that. It's a bit mind-boggling for so early in the morning g to use a linux box, ftp to a windows box, and remember what command the windows box would understand bg I also sent a 'what's happening' mail, but sympa bounced it. I complained to sympa, but haven't had anything back :-( but that's another thread g Anne Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
[expert] Multi-page tifs
Another mail, sent last Friday, that I haven't seen on the lists: I occasionally need to view a multi-page tif - a fax to email document - and embarrassingly have to take it to the awful windows imaging program to view it. Surely there is something in linux that can handle this? Anne Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [expert] OT: router reboots
On Thu, 2003-08-14 at 07:52, Anne Wilson wrote: SMC Barricade 7401BRA adsl router On average, about once a week I have to reboot my router. No apparent reason - just that no Internet connection works any more. After a reboot, all is well, both on this box and all others on the net. I am not able to find any common denominator leading up to the problem. Any ideas, anyone? Anne Does it have firmware that you can update. I'd check on that first. --LX -- °°° Linux Mandrake 9.1 Kernel 2.4.21-0.13mdk *Catch Star Trek Enterprise, Wednesdays on UPN* Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [expert] Multi-page tifs
The GIMP seems to handle everything under the sun. Any luck there? Miark On Mon, 18 Aug 2003 21:12:25 +0100, Anne Wilson [EMAIL PROTECTED] wrote: Another mail, sent last Friday, that I haven't seen on the lists: I occasionally need to view a multi-page tif - a fax to email document - and embarrassingly have to take it to the awful windows imaging program to view it. Surely there is something in linux that can handle this? Anne Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [expert] Multi-page tifs
display seems to work fine - right mouse click and choose next - and it shows the next page. Cheers, R.Fox On Mon, 2003-08-18 at 21:04, Miark wrote: The GIMP seems to handle everything under the sun. Any luck there? Miark On Mon, 18 Aug 2003 21:12:25 +0100, Anne Wilson [EMAIL PROTECTED] wrote: Another mail, sent last Friday, that I haven't seen on the lists: I occasionally need to view a multi-page tif - a fax to email document - and embarrassingly have to take it to the awful windows imaging program to view it. Surely there is something in linux that can handle this? Anne __ Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com -- Robert Fox [EMAIL PROTECTED] Fox Consulting Services Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [expert] Multi-page tifs
On Mon, 2003-08-18 at 13:12, Anne Wilson wrote: Another mail, sent last Friday, that I haven't seen on the lists: I occasionally need to view a multi-page tif - a fax to email document - and embarrassingly have to take it to the awful windows imaging program to view it. Surely there is something in linux that can handle this? Anne archives, I asked this March 31 :-) From man tiff2ps: The following generates POSTSCRIPT Level II for all pages of a facsim- ile: tiff2ps -a2 fax.tif | lpr So multipage TIFF = PS should work. this works great -- program is in the libtiff3-progs RPM. -- Jack Coates Monkeynoodle: A Scientific Venture... Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
[expert] OT RedHat questions
About as OT as it comes, sadly... I'm needing to use an RH system, and I find that urpmi has me massively spoiled. Is there anything like it or a port of it for RH7.3? The old rpmfind/google dependencies hunt for every library and program under the sun is taking hours out of my day, and Ximian's red-carpet doesn't touch urpmi for completeness. I'm actually getting worried that the packages for common Perl stuff simply don't exist, as half the time I can't find them and end up compiling from source any way. What do other RH admins use, the apt-get port from Conectiva? thanks for humoring me, -- Jack Coates Monkeynoodle: A Scientific Venture... Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [expert] Laptop dies unexpectedly!
*** Jack Coates 14 Aug 2003 06:57:35 -0700 : APM isn't supposed to load if ACPI is possible, though I haven't tried that on a laptop which supports both. Looking at that bug-e page, it sounds to me like it is an ACPI laptop but that this chip is too new for the current release of ACPI to handle. It is an ACPI laptop and I played around a bit. I still get that error message about ACPI but now I have a working battery applet which shows me the status. I get the warning signals at the defined states of power and the laptop does a suspend to disk when power runs down to 5%. Do I need more? More Power! (Thanks, Tim and Al!) wobo Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
[expert] Postfix UCE Controls and Fetchmail
Not sure if this is the right list for this type of question but I will plunge in anyway. I am running Postfix 2.04 on my Linux system and using Fetchmail to pull down mail from my ISP and inject it into Postfix. I have local aliases setup for all of the ISP accounts so the mail comes in from the ISP mail address and is then routed to the local address. What I am trying to do is to get Postfix to use RBL checks to bounce mail when it comes from a DNSBL listed address. I am currently using Procmail and Spamassassin for filtering but the DNSBL filters in Spamassassin don't seem to be working at all since mail is never tagged when it comes from a DNSBL listed spammer. I suspect that it is seeing the headers from my ISP as the originating source due to Fetchmail and is therefore deciding that blocklisted spam is actually from my ISP and thus okay. I also suspect that the Postfix RBL client restrictions are failing for much the same reason, it reads the ISP headers and decides that the mail is okay. Has anyone on the list implemented a solution that enables DNSBL filtering from Postfix when mail is injected with Fetchmail, or even some method of directly checking DNSBL from Procmail and rejecting the mail there? -- Bryan Phinney Software Test Engineer Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
[expert] Mdk9.1: Apache 2.0 PHP Script exec's a segmentation fault.
Using: Mandrake 9.1: uname -a: Linux red.somewhere.com 2.4.21-0.13mdk #1 Fri Mar 14 15:08:06 EST 2003 i686 unknown unknown GNU/Linux Greetings! This is my first post to this list, but I've used Mandrake for awhile. I recently did a fresh 9.1 install onto a brand new disk as a way of force upgrading an older Mandrake based system (8.1, if my old install disks are right). I'm not sure if my problem is related specifically to Mandrake or Apache 2.0, but I think it must be the Linux side. I'm trying to exec a command in a PHP script that works fine when invoked from an interactive command line, but seg faults when launched from within the PHP script and I'm wondering if someone can tell me what relevant configurations have changed (I had been using an Apache 1.3.x series before) or why a PHP exec() might cause a Seg Fault for a command that I can launch by hand on the command line. I've also encountered the Signal 11 caught by ps (procps version 3.1.6) error that appears on various mailing lists as being fixed awhile ago. I thought perhaps this was caused by linux-secure, but switching to a plain kernel (see uname above) didn't help. I've had that both in scripts and on the command line, and am not sure if it's related, but it feels like it might be. I've found lots of postings to various mailing lists that talk about various issues and howto's (eg, Mandrake 9.1 + Apache 2.0 + PHP, Signal 11 caught by ps, etc.), but after two days of fighting with this, I don't feel like I'm any closer to an answer. Can anyone share any clues that might help me figure this one out? Thank you! Sincerely, Joe Pezzillo Boulder, Colorado [EMAIL PROTECTED] Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [expert] Postfix UCE Controls and Fetchmail
Bryan Phinney grabbed a keyboard and wrote: Not sure if this is the right list for this type of question but I will plunge in anyway. I am running Postfix 2.04 on my Linux system and using Fetchmail to pull down mail from my ISP and inject it into Postfix. I have local aliases setup for all of the ISP accounts so the mail comes in from the ISP mail address and is then routed to the local address. What I am trying to do is to get Postfix to use RBL checks to bounce mail when it comes from a DNSBL listed address. I am currently using Procmail and Spamassassin for filtering but the DNSBL filters in Spamassassin don't seem to be working at all since mail is never tagged when it comes from a DNSBL listed spammer. I suspect that it is seeing the headers from my ISP as the originating source due to Fetchmail and is therefore deciding that blocklisted spam is actually from my ISP and thus okay. I also suspect that the Postfix RBL client restrictions are failing for much the same reason, it reads the ISP headers and decides that the mail is okay. Has anyone on the list implemented a solution that enables DNSBL filtering from Postfix when mail is injected with Fetchmail, or even some method of directly checking DNSBL from Procmail and rejecting the mail there? I don't know if there's anything that will do what you want, but your assumption here is correct - a DNSBL lookup is going to take place against the IP address of the site connecting to you to send you mail. Anything other than that, it doesn't know. When fetchmail is running, it connects to the localhost SMTP port to let your MTA handle the final delivery - so the only connection that is happening at that point is from your own address. Hopefully, someone out there will know of something that can figure out what Received: line to check in order to do a lookup. --Dave Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [expert] Postfix UCE Controls and Fetchmail
I think there is something in fetchmail to handle this, but try asking on the [EMAIL PROTECTED] list -- there's some smart mail folks there. Jack On Mon, 2003-08-18 at 16:02, Bryan Phinney wrote: Not sure if this is the right list for this type of question but I will plunge in anyway. I am running Postfix 2.04 on my Linux system and using Fetchmail to pull down mail from my ISP and inject it into Postfix. I have local aliases setup for all of the ISP accounts so the mail comes in from the ISP mail address and is then routed to the local address. What I am trying to do is to get Postfix to use RBL checks to bounce mail when it comes from a DNSBL listed address. I am currently using Procmail and Spamassassin for filtering but the DNSBL filters in Spamassassin don't seem to be working at all since mail is never tagged when it comes from a DNSBL listed spammer. I suspect that it is seeing the headers from my ISP as the originating source due to Fetchmail and is therefore deciding that blocklisted spam is actually from my ISP and thus okay. I also suspect that the Postfix RBL client restrictions are failing for much the same reason, it reads the ISP headers and decides that the mail is okay. Has anyone on the list implemented a solution that enables DNSBL filtering from Postfix when mail is injected with Fetchmail, or even some method of directly checking DNSBL from Procmail and rejecting the mail there? -- Jack Coates Monkeynoodle: A Scientific Venture... Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [expert] OT RedHat questions
On Mon, 2003-08-18 at 15:07, Jack Coates wrote: About as OT as it comes, sadly... I'm needing to use an RH system, and I find that urpmi has me massively spoiled. Is there anything like it or a port of it for RH7.3? The old rpmfind/google dependencies hunt for every library and program under the sun is taking hours out of my day, and Ximian's red-carpet doesn't touch urpmi for completeness. I'm actually getting worried that the packages for common Perl stuff simply don't exist, as half the time I can't find them and end up compiling from source any way. What do other RH admins use, the apt-get port from Conectiva? thanks for humoring me, Jack, I've found extreme cursing a swearing to be helpful when dealing with Red Hat. Even Suse is ahead of RH in this area (they are about where MDK was around 7.2) A lot of RH users I've talked to, don't believe urpmi is possible. (then I show it to them, but they have a million excuses why it's no good.) However since urpmi is based on perl-URPM and since the hdlist is done by the command genhdlist (available on RH) it would seem to me that a port could be done. But you'd really have to work hard on setting up a sync for the updates. They seem to run about 3 to 1 vs MDK and SuSE. No problem just that they have more of them. James Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [expert] OT RedHat questions
About as OT as it comes, sadly... I'm needing to use an RH system, and I find that urpmi has me massively spoiled. Is there anything like it or a port of it for RH7.3? The old rpmfind/google dependencies hunt for every library and program under the sun is taking hours out of my day, and Ximian's red-carpet doesn't touch urpmi for completeness. I'm actually getting worried that the packages for common Perl stuff simply don't exist, as half the time I can't find them and end up compiling from source any way. What do other RH admins use, the apt-get port from Conectiva? The closest thing to urpmi on RedHat systems is 'up2date'. You can, for example, install a package with 'up2date httpd'. I'm not certain if it exists for 7.3, but I *think* it is. This is not free though, since it requires a $60/year subscription to the RedHat Network. Well, you can get a demo license for one computer but some updates may fail when the servers are very busy. Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
[expert] Dell Inspiron 3200 for 9.1?
I am being offered this laptop, and I am wondering if anyone has any thoughts about running 9.1 on it. It is a '98 3200 Inspiron, Pentium2-266, 512 mb Ram, with a 4G hard drive. It seems a little slow to me, as my current computer is almost 10x as fast. But with a lightweight window manager I think it might be pretty good for general computing. Rob -- Linux: For the people, by the people. Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [expert] k3b CD writer
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Ronald J. Hall wanted us to know: k3b is not very good as far as I'm concerned. gcombust does everything I want and more. Yea! Another Gcombust advocate! :-) Gcombust also invokes some kernel PFM. On one older kernel, my 24X burner only burns at 16X. With one of the newer kernels (can't recall which one), it burns at 30X. I have no explanation why it's doing that. According to the time it _does_ appear to be burning that fast, but I still have my doubts. - -- Blue skies... Todd | Get a bigger hammer! | Two things prevent you from having| | http://www.mrball.net | sex with women: sunlight sobriety | | http://faq.mrball.net | --Tom Leykis | Linux kernel 2.4.21-0.25mdk 6 users, load average: 1.00, 0.92, 0.53 -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.2 (GNU/Linux) Comment: http://www.mrball.net/todd.asc iD8DBQE/QZz3IBT1264ScBURAkejAJ0cr40LbbDT/UnewU651+5fXkrSHQCfY9Cp tvVQsf1xg4VNSBu/lyV4PH4= =DoWU -END PGP SIGNATURE- Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [expert] Dell Inspiron 3200 for 9.1?
I am being offered this laptop, and I am wondering if anyone has any thoughts about running 9.1 on it. It is a '98 3200 Inspiron, Pentium2-266, 512 mb Ram, with a 4G hard drive. It seems a little slow to me, as my current computer is almost 10x as fast. But with a lightweight window manager I think it might be pretty good for general computing. Hmmm.. I'd check the specs on that Inspiron. As far as I know, the maximum memory on the 3200 is nowhere near 512M. I have a couple Inspiron 3500's though and they work fine. I couldn't get the NeoMagic NM256 sound chip to work at all in any recent post 2.2 kernel however. With the Fluxbox window manager, a HD upgrade the machine works great. Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [expert] Dell Inspiron 3200 for 9.1?
On Mon, 2003-08-18 at 20:27, Rob Blomquist wrote: I am being offered this laptop, and I am wondering if anyone has any thoughts about running 9.1 on it. It is a '98 3200 Inspiron, Pentium2-266, 512 mb Ram, with a 4G hard drive. It seems a little slow to me, as my current computer is almost 10x as fast. But with a lightweight window manager I think it might be pretty good for general computing. Rob totally -- I'd snap it up and use it as a thin terminal. Replace the battery and stick a wireless card in it. -- Jack Coates Monkeynoodle: A Scientific Venture... Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [expert] Cups web interface files missing
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Carlos A. Siso wanted us to know: requiered by cups-drivers). The problem still persists, but one thing I noted is that while rpm -Uhv cups-1.1.19-1.1mdk does not install these files, rpm -e cups does not delete them either. I had to delete the /usr/share/doc/cups directory manually after removing the packages and before installing them again. Normally, rpm -e cups remove the complete directory automatically. To me that means rpm knows it is not installing these files, so it does not need to remove them. The problem is why it is deciding not to install them? If there is a file in that directory that is not owned by the cups package, then it won't delete the directory. Run 'rpm -qf *' in that directory and see if it recognizes everything in that directory. - -- A: No Q: Should I quote at the end of the message? Linux kernel 2.4.21-0.25mdk 7 users, load average: 2.24, 1.77, 1.11 -BEGIN PGP SIGNATURE- Version: GnuPG v1.2.2 (GNU/Linux) Comment: http://www.mrball.net/todd.asc iD4DBQE/QZ/XIBT1264ScBURAt/7AJjx+nINdmD3s0j3XLSd8pLAUSIDAJ9uzx56 AohBTKCdP/CRyg5d/Ukkag== =gU+M -END PGP SIGNATURE- Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
Re: [expert] Installed cable -- now host = HWaddr of NIC
Rolf Pedersen wrote:
Just for reference, I have hooked the adsl modem, which uses dhcp also,
back up, there is no problem with an altered $HOSTNAME and kprinter
works immediately when asked to. I did not even re-run drakconnect or
change any files. Seems to be how the cable modem/comcast dhcp works.
In /var/lib/dhcp/dhclient-eth0.leases, the stored lease info from
comcast includes:
option host-name x1-6-00-03-47-95-32-57;
option domain-name comcast.net;
There are no corresponding options for the adsl lease. Google groups
has quite a few threads about this cable phenomenon, as a search on
'$hostname + $dhcp_hostname' has shown. Maybe Comcast support has an
answer. Maybe another wild goose ;p
Thanks to Narfi Stefansson, I am now enjoying better behavior from my
cable account.
Apparently, certain dhcp servers, such as Comcast's, issue a host name
that does not resolve to the ip that is issued, or anything, for that
matter. This causes lags when programs try to resolve the host name,
which I noticed. This invalid host name is evidenced in the snippet of
the lease I quoted above and is incorporated into my machine by
Mandrake's network scripts (AFAICT). There is a bug report for this
here: http://qa.mandrakesoft.com/show_bug.cgi?id=4227
Narfi has written a script, which he helped me install as
/sbin/ifup-local, which is invoked by Mandrake scripts each time the
interface is brought up. This script ensures that hostname resolves to
the ip issued by the dhcp server. As it turns out, I use guarddog and
guarddog links /sbin/ifup-local to /etc/rc.firewall, where the iptables
rules appear to be written (I know nothing about networking). So, we
mved that link to back it up and Narfi added one line to the end of his
script to ensure that /etc/rc.firewall continues to be read when it should:
#!/bin/bash
outfile=/tmp/report.txt
echo ifup-local has been invoked $outfile
dhcpfile=/var/lib/dhcp/dhclient-eth0.leases
ip=`grep fixed-address ${dhcpfile} | tail -1 | sed 's/[^0-9.]*//g'`
address=`host $ip 2/dev/null| grep domain name| sed 's/.* // ;s/.$//'`
echo IP is ${ip} $outfile
echo Address is ${address} $outfile
echo Current hostname is `hostname` $outfile
hostname ${address} $outfile
echo Corrected hostname is `hostname` $outfile
/etc/rc.firewall $*
This has solved the lagginess I was seeing when loading the desktop,
starting konsole, etc. BTW, I booted connected to my dsl modem and the
unproblematic, familiar behavior of that account (with no hostname
option given by the dhcp lease) was unchanged by the script.
The problem with printing turned out to be a separate consequence of my
switch to the cable account. An strace of kprinter turned up a
reference to 172.16.1.10 just as it stalled, which was an ip from my dsl
account, and grep -r 172.16.1.10 /etc revealed an occurrence at the
bottom of /etc/cups/cupsd.conf:
ServerName 172.16.1.10
That explains why there wasn't a problem when I was hooked up to the
adsl account. Commenting that line restored printing to its usual,
snappy self. Thanks, again, to Narfi for helping me ferret that one out
and thanks to all who responded.
Rolf
Want to buy your Pack or Services from MandrakeSoft?
Go to http://www.mandrakestore.com
Re: [expert] OT RedHat questions
apt-get for rpm. Check out www.freshrpms.net. I've used it before, and it's actually better than urpmi (the servers are more reliable and it takes less time to update sources). Jeremy Gregorio Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
[expert] Any way to chroot jail a user?
I have a user login name that is used to run a game server process (Neverwinter Nights, if it matters :). I don't know if it's possible for a remote user to crash the game process in a way which would leave them sitting in a shell, but since I don't know that the chances are 0%, I'm thinking that having the login name chroot jailed to its home directory would limit the damage that someone could do if they *did* somehow manage to end up in a shell via a server process crash. Is there a way to to this? --Dave Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
