Re: two instances of the same kernel version, different modules
Cornelius Kölbel wrote: Hello, I use one version of kernel sources but will compile different instances of the kernel and of the modules. So I will have a module directory /lib/modules/2.4.20 and a module-directory /lib/modules/2.4.20-b belonging to the second compiled kernel. How can I tell lilo that the second compiled kernel will find its modules in the directory /lib/modules/2.4.20-b. I want to be able to boot either kernel 2.4.20 2.4.20-b But also when I but the kernel 2.4.20-b the kernel looks for its modules in /lib/modules/2.4.20 You need to edit the Makefile in the kernel source and change the etraversion varible before you do a make clean dep bzImage modules modules_install install -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: GRUB question
Susan Champigny wrote: Grub is not a favorite of mine, I prefer lilo, so I am hoping someone out there can help w/ the following. I initially installed lilo, and was having boot problems, so I editted /grub/grub.conf. This particular system has and ide, and scsi disk, scsi being the disk w/ the OS, and ide disk w/ data. I believe the system was confusing the disks, thinking hda was my primary and bootable device, wrong. Reason why I went to grub. However, I can boot the system now, but when I boot I get the grub (prompt), and I have to type grub configfile /grub/grub.conf What do I need to run so I do not have to do this everytime I boot the system. I know lilo I run lilo -v -v. help ! With lilo you need to do something like below to force ordering of the disks: disk = /dev/sda bios = 0x80 disk = /dev/hda bios = 0x81 In grub I believe you will need to edit /boot/grub/device.map, and do something like this: (fd0) /dev/fd0 (hd0) /dev/sda (hd1) /dev/hda PS- You should make sure you have a TESTED means of booting the system from other media. Before you try any of this. Remember your redhat install cddom has a rescue mode. -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: Fedora
You'll have to purchase an Enterprise offering to get support from RedHat. But there will surely(?) be auto updating feature provided for Fedora. Fedora currently supports yum, and apt-get. I've been using fedora, and before that freshrpms with apt-get, and yum for 6 months. I've been very happy with them. -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: Fedora
Buck wrote: 3. Will the Fedora Project Releases be compatible to existing Red Hat? In other words, can I upgrade what I have to Fedora or will I have to start with Fedora from scratch? You can upgrade to the current Red Hat beta via apt-get or yum using fedora now. All I did was the following: 1)Download yum rpm from the fedora web site, and install it. 2)Download redhat-release rpm for the beta off of the fedora ftp site, and install it. 3)Edit /etc/yum to point to a closer mirror. 4)Do a yum upgrade 5)Do a yum install synaptic 6)Run synaptic, and decide if you want to install any additional packages. -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: Fedora
Gerry Doris wrote: At 12:23 9/22/2003, you wrote: Does this mean that I won't be downloading RH 10, but instead will be downloading Fedora 10 or something? Fedora Core 1 (Cambridge), apparently, which will contain everything you expected to see in Red Hat Linux 10 and more due to the contributions of the Fedora Project. and less since Redhat will insist on several apps being removed due to potential licensing issues. Yes, but you'll be able to just yum install something from another repository. The mplayer, mp3, and the other rpms aren't going to disappear. -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: REDHAT 9.0 PROMISE FASTRAK 100 TX2
Michael Mansour wrote: Placa base intel con bios actualizada Promise fastrak 100 tx2 con bios actualizada. Red hat 9.0 i drivers de promise ultima version. La instalación del red hat tambien la hace correctamente solo hay una pequeña duda, cuando entro en el discdruid veo 3 discos los dos que he pinchado en la contoladora raid y otro que me sale con el comentario promis raid que es de la misma capacidad que los otros, supongo que sera el mirror pero mi pregunta es, ¿no tendría que ver uno solo?. Linux kernel does not support FastTrak mirror. If you want to use it for mirror, you have 2 choices: 1) disable mirror in the fasttrak bios; then use linux software mirroring. Basically you'll use the fasttrak as an non-raid ata cotroller. 2) use binary drivers promise provides here http://www.promise.com/support/download/download2_eng.asp?productId=88category=Allos=100 This will allow you to use the mirror you set up in the fasttrak bios. Why don't these companies release sources to their drivers to be included in the kernel? I don't see this binary driver implementation very flexible and self-sustaining, puts more work on the supplier side to get this working for each distribution and each kernel release, on top of the inflexibility it can provide to the customer. These cards do raid in the driver. Thus they would need to open source their raid engine. In many cases the vendor has licensed their raid engine from some where else. This is why the cheap cards have binary only drivers, and the expensive card have open source drivers. The expensive cards are doing raid on the card so all they are giving away is how you talk to their card. -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: Intel E/1000 driver
Vinny Valdez wrote: MW Mike Weiner (5028) wrote: I am having some issues with a module I need to load for an intel e1000 interface using RedHat 7.3 linux-2.4.18-3. For some reason, the new imaged machines that I am trying to bring up will not load that module nor initialize the interface, so I am stuck with a single 10/100 interface. Interestingly enough I get the following in dmesg First, I would recommend using a newer kernel. Aside from numerous security fixes, the driver version of the e1000.o in that default kernel is 4.1.7. Compared to the e1000.o driver in the 2.4.20-20.7 kernel, which is 5.0.43-k1. eepro100.c:v1.09j-t 9/29/99 Donald Becker http://www.scyld.com/network/eepro100.html eepro100.c: $Revision: 1.36 $ 2000/11/17 Modified by Andrey V. Savochkin [EMAIL PROTECTED] and others eth0: OEM i82557/i82558 10/100 Ethernet, 00:30:48:28:8D:CB, IRQ 22. Board assembly 00-000, Physical connectors present: RJ45 Primary interface chip i82555 PHY #1. General self-test: passed. Serial sub-system self-test: passed. Internal registers self-test: passed. ROM checksum self-test: passed (0xd0a6c714). Intel(R) PRO/1000 Network Driver - version 4.1.7 Copyright (c) 1999-2002 Intel Corporation. This is your problem. You want to use e1000, but eepro100 is getting loaded. No he has a 10/100 and a GigE nic. The e100, or the eepro100 will only grab the 10/100 nic And nothing else really, but when I go to manually probe or insert the module I get the following: [EMAIL PROTECTED] root]# modprobe e1000 Warning: loading /lib/modules/2.4.18-3smp/kernel/drivers/addon/e1000/e1000.o will taint the kernel: non-GPL license - BSD with patent grant See http://www.tux.org/lkml/#export-tainted for information about tainted modules /lib/modules/2.4.18-3smp/kernel/drivers/addon/e1000/e1000.o: init_module: No such device The device is in use by eepro100. No see above. The older e1000 driver doesn't recognize the newer chipset. [EMAIL PROTECTED] root]# insmod e1000 Using /lib/modules/2.4.18-3smp/kernel/drivers/addon/e1000/e1000.o Warning: loading /lib/modules/2.4.18-3smp/kernel/drivers/addon/e1000/e1000.o will taint the kernel: non-GPL license - BSD with patent grant See http://www.tux.org/lkml/#export-tainted for information about tainted modules /lib/modules/2.4.18-3smp/kernel/drivers/addon/e1000/e1000.o: init_module: No such device Same thing. Nope [EMAIL PROTECTED] root]# lsmod Module Size Used byTainted: P autofs 12804 0 (autoclean) (unused) eepro100 20816 1 See, eepro100 is currently loaded. You need to remove it first. Does anyone have any quick solutions they might suggest to get this module to load correctly? Or any other suggestions? You will need to down your interface, change the module, and bring the interface up again. If you are doing this remotely, you can do it all in one command: [prompt]# ifdown eth0 rmmod eepro100 insmod e1000 ifup eth0 You may loose connectivity for a second, but it should come back. You may have to reconnect the remote session. If you are satisfied, then make the change in /etc/modules.conf from eepro100 to e1000. Again, use a newer kernel if you can. -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: help with partitioning
Chris W. Parker wrote: Samuel Flory mailto:[EMAIL PROTECTED] on Saturday, September 13, 2003 11:47 AM said: You need to create a partition that will be a part of the raid array on each disk. So the followiing is what I do: Yes, I finally figured this out. So far the computer is working excellently. It's much faster than our email gateway (my first linux box). :) Thanks for your help, Chris. p.s. I still have plenty of time to reinstall and repartition if necessary. I'd appreciate it if you would comment on the way I've setup the partitions. Here is the output of df -ah: FilesystemSize Used Avail Use% Mounted on /dev/md1 4.0G 136M 3.7G 4% / none 0 0 0 - /proc usbdevfs 0 0 0 - /proc/bus/usb /dev/md0 99M 9.2M 85M 10% /boot none 0 0 0 - /dev/pts /dev/md4 21G 33M 20G 1% /home none 188M 0 188M 0% /dev/shm /dev/md2 7.9G 746M 6.8G 10% /usr This doesn't look good for the type of use you are planning. This is a web server and logging 1) Is /usr going to grow that much? You've got less a 1G of used space now. 2) You have only 4G in / which also contains /var. /var is where Red Hat keeps logs, web pages, and the like. 3)You have 21G in /home. What is going to use that space? My advice for what it's worth. /boot ~100M / ~4G swap~2 x ram /vareverything else Now once you've finished installing link /home, /opt, and /usr/local to /var. These are the only places that are likely to grow in size over time. / and /usr will grow if you upgrade, but you still have 3G of space to grow into. mv /home /var/home mkdir /var/opt mkdir /var/usr mv /usr/local /var/usr cd / ln -sfi var/home home ln -sfi var/opt opt cd /usr ln -sfi ../var/usr/local local -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: Best HBA's for use with Red Hat
Rhugga wrote: We are building an Oracle RAC and I wanted to get some feedback on the best HBA's to use with Red Hat 9.x Raid or non raid. Fiber, scsi, or ide? -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: Adaptec ATA RAID 1200A under Linux
Michael Mansour wrote: Hi, I've just purchased a new AMD Ahtlon system, and got with it an Adaptec ATA RAID 1200A as it seemed to do what I wanted. Unfortunately, this card is not officially supported by Adaptec, after emailing their support they only tell me unfortunately it's not supported under Linux, yet they have all the Windows drivers. They told me the 2400A card (which is many times more expensive and has more features than what I need), is supported under Linux and they provide the drivers for it. They claim to support 2400A because they wrote an open source linux driver for it. The driver is now in the main line linux kernel. Thus it will just work in future releases of any distro. This is what good linux support means!!! Not Promise says Sure I'll release a binary driver, and maybe update every years or so. Highpoint is at least much better about providing partial source, and the ability to use their cards in a non raided state. Having googled for information on this card and Linux, I've discovered that the card can be used using the HPT370 (HighPoint Technology) driver, since Adaptec use that chipset on their card. I obtained the driver from: http://www.highpoint-tech.com/r100r_down.htm for Red Hat 8.0 and it works fine. I'm wondering why Adaptec don't officially supply this driver themselves??? and don't provide support for the card under Linux since it works perfectly well. All I need is mirroring facility on the IDE drives, so that's all I wanted in the card. The more expensive 2400A (which Adaptec officially support under Linux) has all the bells and whistles, has a processor on board and supports RAID 5, but I don't need those features so don't need that card. If it wasn't for the HPT driver, I would have got rid of the Adaptec and looked at something else, maybe a Fastrack 100 or similar, whatever is supported under Linux. I don't see the reasoning behind Adaptecs decision not to support the 1200A under Linux. My contacts with Adaptec lead me to the following conclusions. Adaptec wisely views binary drivers as unmaintainable under linux. As they can't give you good support they choose not to do so at all. This is true of all of their controllers that do driver based raid. Such controllers are only supported as in jbod mode under linux. Adapter is quick to point out linux has great software raid support. Don't bad mouth Adaptec for it's linux support. They said it didn't work under linux. You were the one that disagreed with them. They are one of the best controller card vendors in terms of linux support. They don't make promise they can't, or won't keep. Keep in mind they couldn't release the driver source for this card even if they wanted to. They could have easily done what Highpoint and Promise do, but they have higher support standards. Promise's linux support is incredibly bad they claim they support Linux, but they really only support a handful of kernels. They don't provide drivers for newer releases of Red Hat in timely manner. Generally you are stuck with a release 6 to 12 months old. Which has secuirity holes, and bugs. Heaven help promise's linux customers when Promise stops support the fasttrak line. Anyway, if anyone has experience with this card I do have one query. I'd like to upgrade my kernel from 2.4.18-14smp which I have on the fresh RH8 install (installing the HPT driver upon installation via the driver disk), to the newest 2.4.20-20.8smp, but HPT have only got a module for 2.4.20-8smp on their driver disk. Can I use this module on the newer 2.4.20-20.8smp by just copying the module into the kernel tree for 2.4.20-20.8smp ? No you can't simply use a module compiled for one kernel on another kernel. It might work, not work at all, or gradually corrupt your data. In theory you can find the partial source driver below. I don't remember which chip the adaptec card uses, and can't check any more. (I gave the eval cards my adptec rep gave me back.) http://www.highpoint-tech.com/372drivers_down.htm http://www.highpoint-tech.com/371drivers_down.htm http://www.highpoint-tech.com/374drivers_down.htm If not, how else can I upgrade the kernel to 2.4.20-20.8smp while still using the HPT module / driver. Note that I have my RAID 1 mirror as the boot drive for Linux. You best bet going forward is to stop using the highpoint binary driver!!! This problem is going to come up over and over. You got 2 choices. #1) Use the card in jbod mode and use the linux software raid driver. This would reqire a complete reinstall, but will just work for release after release of any linux distro, or kernel. #2) Try using the ataraid driver. Which is fairly complex and not support in the 2.6 kernel yet. https://listman.redhat.com/mailman/listinfo/ataraid-list http://www.murty.net/ataraid/main.html -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED]
Re: help with partitioning
Chris W. Parker wrote: Samuel Flory mailto:[EMAIL PROTECTED] on Friday, September 12, 2003 5:38 PM said: Ok I'm convinced, I'll use RAID. I found this page http://www.redhat.com/docs/manuals/linux/RHL-9-Manual/custom-guide/ch-so ftware-raid.html which you'd think would be the perfect set of instructions. BUT IT'S NOT!! The raid instruction are a little vague. I'm trying to follow Samuel's original instructions: -=-=-=- Use software raid 5 on each disk: raid 1 /boot (~100M) Raid 1 as you can't boot off of raid 5!!! raid 5 swap raid 5 / (~2G) raid 5 /var (Most of the rest of your space, for your logs and /var/www) -=-=-=- (Following the steps in the above document) Step 5 says For Allowable Drives, select the drive(s) on which RAID will be created. If you have multiple drives, all drives will be selected here and you must deselect those drives which will not have the RAID array on them. But if I select more than one drive I get the message Partitions of type 'software RAID' must be constrained to a single drive. This is done by selecting the drive in the 'Allowable Drives' checklist. Ok, So what's the deal with that? How can you have RAID if you're forced to only use one disk? I don't get it!! You need to create a partition that will be a part of the raid array on each disk. So the followiing is what I do: Create a partition of type raid on disk 1 ~100M via the new partition button Create a raid partition of type raid on disk 2 ~100M via the new partition button Create a raid partition of type raid on disk 3 ~100M via the new partition button Create the raid array using the above 3 disks using the raid buton Repeat for all arrays I've always thought Red Hat's method was a little strange;-) Why not let me create multiple partitions on each disk in one stoke. AAHH!! Chris. -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: Best HBA's for use with Red Hat
Rhugga wrote: Samuel Flory wrote: Rhugga wrote: We are building an Oracle RAC and I wanted to get some feedback on the best HBA's to use with Red Hat 9.x Raid or non raid. Fiber, scsi, or ide? Fibre, using them to connect 2 linux boxes to some external shared storage. I think IBM uses QLogic 2300's but would like some more input. The QLogic cards seem fairly popular, and they are the only ones I've used. -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: help with partitioning
Chris W. Parker wrote: Hey people. I've got a new computer with 3 scsi drives at 17gb each. This machine is going to be used as a web server. I've only done one other install (tried twice on same machine) and it had only one HD and a much much smaller one at that, so this seems to be a different ball game. How can I best take advantage of the three drives? Thanks, Chris. p.s. Installing RH9. Use software raid 5 on each disk: raid 1 /boot (~100M) Raid 1 as you can't boot off of raid 5!!! raid 5 swap raid 5 / (~2G) raid 5 /var (Most of the rest of your space, for your logs and /var/www) Be sure to use the lilo as grub will not be installed to all all of your drives, and if you lose the 1st drive you can't boot!!! -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: help with partitioning
Chris W. Parker wrote: Samuel Flory mailto:[EMAIL PROTECTED] on Friday, September 12, 2003 5:08 PM said: Use software raid 5 on each disk: Will this degrade the performance much? This will increase performance of reads a lot, and degrades write performance a little. As this is a web server you will be doing almost entirely reads. I've read that software RAID is quite a bit slower than hardware RAID. The reverse is true in every benchmark I've run. This is because the cpu in your system is easily a x1000 faster than that of a raid controller. In addition raid 5's cpu overhead occurs durning write not reads. (Unless you've lost a drive.) Also, do I have to use RAID at all? No, but you asked How can I best take advantage of the three drives. http://www.tldp.org/HOWTO/Software-RAID-HOWTO-2.html Can I not just use each drive separately? (Only asking because I don't know any better. ;) ) Sure put / on one drive, /var on another, and swap on the 3rd. Or split swap up evenly on all drives, and figure some complex scheme of partitions, and mount to give you use of the the space on -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: file system type for Super Talent USB key drive
John Mann wrote: so it looks like I have the USB drive plugged into /dev/sda3 (there are four USB ports on the computer), and the file system is OnTrack DM6 Aux3. Earlier today, I stored three test jpeg files on it, on a Windows 2000 system at work. Is there a filesystem I could format it with, that would work for Windows XP, Windows 2000, and Linux? And how would I do that? I appreciate the help very much. Format it as fat32. XP, or 2k format as fat32 when you select fat. -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: Segmentation fault
Devon Harding - GTHLA wrote: This happens when I run certain programs like 'wget', 'rpm' 'Segmentation fault' Can you show a strace of the program? strace command -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: Viruses W32/Sobig.F Worm
[EMAIL PROTECTED] wrote: Dear All, I just found the viruses W32/Sobig.F Worm attack our system ( RH 7.2 ), but there is no Anti-Viruses... Sobig is a windows virus. It won't effect a linux system. -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: kernel screw up .. upgrade to 2.4.20-20.8
Ian L wrote: At 04:39 PM 9/5/2003, you wrote: At 11:25 6/09/2003, you wrote: I recently downloaded the updated kernel rpm and instead of doing rpm -i, i did rpm -U and replaced the existing kernel. I havent rebooted the machine yet so its still running the old kernel. i have 2 questions: 1. is there some way i can get the old kernel back? I'm a little worried about rebooting in case the new kernel doesnt work for some reason. You should be able to re-download your old kernel and just rpm -ivh file don't forget to get the SMP one :-) Well, i cant seem to find the SMP version. I looked on the redhat ftp mirrors, and the closest i found were: kernel-2.4.18-14.src.rpm kernel-source-2.4.18-14.i386.rpm kernel-smp-2.4.18-14.i686.rpm Unless you don't have a PII or better the i686 is what you need. Red Hat doesn't compile a i386 smp kernel rpms any more. (Mainly as almost no one has an smp 486, or 386.) kernel-BOOT-2.4.18-14.i386.rpm kernel-2.4.18-14.i586.rpm I'm guessing anything other then i386 isnt going to work, but i didnt see an i386 smp version. Nope, those two files are gone, i assume replaced when i upgraded. I was assuming lilo would update itself when i rebooted. I just dont want to reboot until i can get the old kernel back, and get the smp version of the new kernel. I've looked on the redhat update ftp site and i did not see any smp version of the kernel. Although there is an smp version for i586 and i686. Maybe i'm getting a little screwed up with my cpu architecture. Are xeon cpu's i686? because thats what in the machine, 2 xeon cpu's. Xeons aren't really i686's. The i686 means the kernel was compiled with the CONFIG_M686 option set. This turns on the Pentium Pro optimatizations. These will work and make things faster on Pentium Pros, P2, P3, P4, and as well as the Xeon P2, P3, and P4s. (In theroy you might want to compile your own kernel with CONFIG_MPENTIUM4, but I've never noticed any performance increases.) -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: kernel screw up .. upgrade to 2.4.20-20.8
Ian L wrote: Thanks, i think i got it now. I just installed kernel-smp-2.4.18-14.i686.rpm. One question though, my lilo file is called lilo.conf.anaconda and when i run /sbin/lilo it complains it cant find lilo.conf That mean you are using grub. (Red Hat's installer creates a lilo.conf.anaconda in case grub doesn't work.) Take a look and the /etc/grub.conf file. -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: yum vs rpmfind vs autorpm
Herculano de Lima Einloft Neto wrote: Hello all, I'm trying to figure out which one of these tools I should spend my time configuring and learning for now. It seems like yum is the only one still under development. Can anyone give me some insight out of personal experience? You want either yum, or apt-get. Take a look at fedora or freshrpms they support both methods. http://www.fedora.us/ http://freshrpms.net -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: /boot won't mount after upgrade to hardware RAID
Thomas Fortner wrote: Hi folks: I added an IDE RAID controller and a second hard drive to my Red Hat 9 system, which moved the hard drive from hda to hdg. I found a couple of places where I needed to change the (HD0,0) definition, such as /boot/grub/device.map and /etc/sysconfig/grub. However, I must have missed one somewhere. During the boot process, /boot fails to remount in rw mode after unmounting, reporting that it is busy. If I bring the system up in single user mode, mount /boot manually, and change to init 5, everything works properly. In fact, I am writing this on the system having this problem, playing The Two Towers DVD in mplayer and running Netscape and evolution with two terminals in the background. The system is solid after this initial problem has been bypassed. So, can anyone tell me where the configuration file is that I missed that is causing this error? Looking /etc/fstab, and /etc/grub.conf (or lilo.conf if your using lilo) for references to /dev/hda. PS- Or you could try adding ide=reverse to the kernel options you boot with. That should make hde into hda;-) -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: formatting a disk in a USB floppy drive
Lorenzo Prince wrote: Hi. I just bought a USB floppy drive to go in my laptop. It mounts the disks I put into it with no trouble and reads very fast. The problem is that the drive is using /dev/sda. Neither fdformat or floppy like that. What can I do to format a floppy disk in this drive? Thanks. Try mkdosfs /dev/sda. If that doesn't work you can do what I do with a camera memory I use for data transfer. cat /dev/sda flop mkdosfs flop cat flop /dev/sda -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: formatting a disk in a USB floppy drive
Lorenzo Prince wrote: Sorry for starting a new thread in your mail client that supports threading. Unfortunately I deleted the previous message regarding mkfs before I went to the laptop and tried it. The mkfs command returns a warning that states that /dev/sda is an entire device instead of a single partition and asks if I want to proceed. If I press y it seems to work fine, so I don't know what it is doing. It almost looks like it's seeing the floppy as a hard drive. Yes the usb mas storage driver goes thru the scsi subsystem. So every thing is a scsi disc, cdrom, or a scsi generic device. I recommend putting something in your /etc/fstab like this: /dev/sda/mnt/usbautonoauto,owner 0 0 Now mount /mnt/usb should work. However when I first plugged the drive into the computer and booted it, it mounted with the following command: mount /mnt/floppy Am I missing some special property of these drives or something? Thanks. Possibly you have legacy usb support enabled in your motherboard bios. When you boot the usb might be being emulated as a normal floppy. I know this works with usb keyboards, and mice. Certainly this works under dos when I boot from usb floppy. (I've never seen this under linux, but I never thought to check.) If this is true then the key is booting with the usb floppy in the system. -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: formatting a disk in a USB floppy drive
Lorenzo Prince wrote: Samuel Flory staggered into view and mumbled: Try mkdosfs /dev/sda. I get a strange error stating mkdosfs will not try to make filesystem on /dev/sda. This seems to be a bug in mkdosfs. I've seen this as well, but didn't really look at it to closely as I've only got the one 16M memory stick that came with my camera. (16M just isn't big enough for my photo needs.) I suspect that the device is failing some sort of test. I've not looked at the mkdosfs source. with a camera memory I use for data transfer. cat /dev/sda flop mkdosfs flop cat flop /dev/sda This seems to work, and the files I add afterword are readable after unmounting and remounting /mnt/floppy. But will this work on a previously unformatted disk? It seems work on the memory stick which had a partition table and partitions on it. As far as working on an unformated floppy. All the floppies I get are already dos formated;-) -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: Noticeable Improvement with 2.4.22 (FWIW)
David Hart wrote: Usually kernel changes go without much difference but 2.4.22 (just released) seems to provide a real performance boost. YMMV. There was a fair amount of work that went in to fix interactivity issues. -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: Anyone grabbed and compiled a 2.6 kernel yet?
Douglas Phillipson wrote: I would like to demo a 2.6 kernel for our LUG. I don't do this kind of thing often and was wondering if there were any new things I need to know about compiling a 2.6 kernel as opposed to a 2.4 kernel. Is it the same make config, make dep, make bzImage method? Or has it changed. Anyone have a good kernel reference for compiling 2.6? It's pretty easy just be sure to compile in support for virtual consoles, and keyboards. Or grab RH's kernel rpms. http://people.redhat.com/arjanv/2.5/RPMS.kernel/ -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: acpi
David Hart wrote: On Tue, 2003-08-19 at 17:57, Brian wrote: How do i install acpi on Redhat 9? You need to custom compile the kernel. It works quite well BTW but cannot coexist with APM (whichever first occurs gets loaded). FWIW, I had better results with 2.4.21. If you want APCI support you should really look into the severn beta. Also the 2.4.22-pre, and 2.6.0-test have vastly better APCI suppport than 2.4.21. -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: creating an ext3 partition
Jason Dixon wrote: On Fri, 2003-08-15 at 09:26, Earl wrote: I'm using RH9 with ext3 partitions. I want to add another ext3 partition. How can I do this? Parted, for instance can edit ext3 partitions, but not create new ones. If you have free space on the drive use fdisk. Remember to reboot after running fdisk (need to update the partition table if you are already using the disk). Mke2fs -j the new partition, and add an entry in your fstab. Also mkpartfs, and mkpart in parted should work, but I've never used parted. Ext3 is simply ext2 with a journal. Create your ext2 filesystem, then your journal with tunefs -j /dev/hdXX. This is not what he is asking. And you can also do mke2fs -j. -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: Kernel Vicious Circle
David Hart wrote: In my infinite wisdom I seem to have compiled my kernel without loopback support. I tried to re-compile with some changes and MkInitdrd fails - all your loopback devices are in use. The only fix seems to be to recompile the kernel with loopback support which I can't do because it is already compiled without it. This precludes me, of course, from installing a kernel via rpm as well. Any ideas? Do you need an initrd? Did you compile ext3, and your scsi driver (if any) into the kernel? I never use initrd for my own kernels. 1)Install the kernel-BOOT from the RH cdrom. (Yes mkinitrd will fail) 2)Boot the RH cdrom in rescue mode. 3)Now you should be able to mkinitrd. -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: Linux Memory Behavioir
Jonathan Bartlett wrote: I was under the impression that only 2GB was mapped to userspace, and the other 2GB was mapped for kernel data, although I could be wrong. I'm fairly sure redhat's kernel configures a 3G/1G split. So the max you could ever get would be 3G, also it will depend on which version of glibc you are using. You can push this to 3.5 by hacking the kernel a bit. Take a look at Andrea's kernel: ftp://ftp.kernel.org/pub/linux/kernel/people/andrea/kernels/v2.4/2.4.22pre7aa1/00_3.5G-address-space-5 Jon On Tue, 5 Aug 2003, Robert Vaughn wrote: During some tests we have observed some odd memory behavior in Linux. It appears that our Linux server with 10 GB of RAM will only allocate a maximum of 2.8 GB per process. When we try to exceed 2.8 GB per process the process dies. We are interested in finding why and how to fix the behavior. Our questions are... 1) What is limiting the amount of memory that we can access? I understand that we should not be able to access above 4GB per process on a 32 bit system. However, being able to access only 2.8GB is not very good. 2) What can we do about the rather low memory limit? Following is some information about what we have been doing. I can provide additional details as requested. The program... The program that we're using to perform the tests is a Perl script that consumes a specified amount of memory through a loop of stuffing characters into an array. I wrote the Perl script. We are going to create a C++ version of the Perl script. However, we do not think that Perl is the problem. Previously I have used the same script to consume about 3.8 GB of memory before the script gets killed. When the script runs we observe no swap behavior. We can run multiple scripts of say 2GB memory consumption and eat up the entire free memory space and then start eating into swap. Background... Intel Xeon 8X 10 GB ram Redhat As 2.1, 2.4.9-e.3enterprise Memory Info with No Load on Server... Total: 10303272 KB Used: 315512 KB Free: 9987760 KB -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: Command Line for Searching for Files Containing Text?
Lorenzo Prince wrote: Hmm. Try the following if you want to search within a specific directory: for f in /directory/*; do grep text $f; done or you can omit /directory/ if you want to search the current directory, so the command would look something like this: for f in *; do grep text $f; done I think this will do what you need it to do. HTH. That's a little complex why not grep some text /directory/* Or grep -r some text /directory if you want to search all of the files in any directorys within a directory. -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: RH 10
mark wrote: Well, I just read about RH opening up the development process to outsiders, over on ZDNet. *Then* I read the system requirements for 10: 200MHz for *non-graphical*, 400MHz for graphical...*minimum*. Does RedHat think they're the next M$? Now, up until a month or so ago, I was running 7.3 w/ KDE, w/ kernel lib updates, etc. 7.3 came out, what, a year and a half ago?...and I'm running an AMD K-6 233 (ok, mine's overclocked to 250 g, for the SDRAM...). No problem. I put IceWM in place of KDE, and it runs like a champ. Now I've just upgraded to 9, same deal (though the once or twice I tried KDE, it ran slower than Lose95 on a '486). Non-graphical is jes' fine. To me, one of my arguments is that you don't *have* to upgrade your hardware until it physically gives up the ghost. It's M$ that makes you have to upgrade hardware, every time you get a new release. Especially in the middle of the most major recessino since the Depression, when companies are running tight, and home users are strapped for cash, most can't afford to buy new hardware. And RedHat's answer is...? You aren't our target market. That said I think it may be easier to run redhat on older systems than before. I'm hoping that we'll manage to get them to add a light wm like blackbox. So you should be okay as long as you don't run OO, or mozilla. -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: Solaris-RH8 Backup
System Administrator wrote: On Monday 21 July 2003 18:25, Samuel Flory wrote: System Administrator wrote: I need to backup a Solaris 8 system (actually parts therein) to a RH8 box. I am currently doing that using a simple scp (secure copy) script. The RH8 filesystem is ext3. Question 1) When I copy the files over (all normal files), they occupy 2-3 times the space. Why? Question 2) is there a cleaner way to do this? I read recently where someone suggested rsync. I am investigating whether that is an option on the Solaris side. Use tar. tar czvpf - (some files, or directory) |ssh [EMAIL PROTECTED] cat my.tarfile.tgz Well It sounded greatunfortunately all it does is create the loccal tar file. Forgot the quote to prevent the shell from stealing the . tar czpf - (some files) |ssh [EMAIL PROTECTED] 'cat foo.tgz' -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: No Wine on RH9.0 ?
Joseph A Nagy Jr wrote: On Tuesday 22 July 2003 08:59, Kent Borg wrote this in an attempt to be witty and informative: On Tue, Jul 22, 2003 at 12:25:33AM -0500, Joseph A Nagy Jr wrote: Wine is in 9.0. Are you sure? I don't see it in my everything install. -kb It was installed by default for me and I didn't do an everything install. Wine is not in RH 9. Read the release notes folks. I believe that it was taken out due to issues with the new threading model. http://sourceforge.net/project/showfiles.php?group_id=6241 -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: Opteron Processor
Christian Fredrickson wrote: I would like to create a database server using dual Opteron processors. I read that Redhat planned to support the Opteron, but I cannot find any information on what versions of Redhat will support the Opteron. Does anyone have any information or suggestions? There is a preview of the opteron on their ftp site or any mirrors that mirror the redhat/linux/preview/ directory. -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: Root is GONE
Mr. L.R. Adrian wrote: Thanks for the quick response. I dont claim to be great with linux {probly borderline mediocre :O)] But I tried to get to the boot prompt to enter linux single but could not do it. as an ordinary user I cant seem to accomplish anything once in. Can i not run single user mode because of GRUB??? You should be able to get into single user mode via grub. -At the grub menu hit e -chose the line starting with kernel, add single, and hit enter -hit b to boot If you can't do this because you have a grub password you can't remember, or something. Use the redhat install cdrom and type rescue at the initial prompt before the cdrom boots. thanks Les - Original Message - From: Jonathan Bartlett [EMAIL PROTECTED] To: [EMAIL PROTECTED] Sent: Monday, July 21, 2003 2:57 PM Subject: Re: Root is GONE IF you have been hacked, you should just backup your data and reinstall. Are you sure theres no other way you could have lost your /etc/passwd file? That's basically the problem - /etc/passwd is either missing or corrupted - and thus it can't find the root user. I don't think shadow has much to do with it. If you can copy /etc/passwd from a good RH machine of the same version, and then run the passwd command from single-user mode, you should be at least in a little better shape. Jon On Mon, 21 Jul 2003, Mr. L.R. Adrian wrote: Assistance is both urgent and appreciated. Obviously i have been compromised. I run 7.3 Valhalla. i went to login to my server as root today and recieved the message. Usr root does not exist upon research this is in fact the case. i boot from grub and as the boot sequence progresses it gives: getpwnam failed for [EMAIL PROTECTED] Swap Space [OK] I can log in as user only. I think it is my shadow thats been breached. boot disk does not even work. {possibly because of an earlier kernel config. This is my only server ..please help if you can. Les -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: Root is GONE
Mr. L.R. Adrian wrote: Also, In /etc/ there are three passwd files: passwd passwd- These are normally there. an passwd.OLD This is not normally on most systems. Maybe a someone was editing the password file by hand. Or possibly a broken script. Maybe from some from some script kiddy. the bottom two contain the root listing on the top line the passwd file this is deleted Tried of course to overwrite but no permissions. -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: Solaris-RH8 Backup
System Administrator wrote: I need to backup a Solaris 8 system (actually parts therein) to a RH8 box. I am currently doing that using a simple scp (secure copy) script. The RH8 filesystem is ext3. Question 1) When I copy the files over (all normal files), they occupy 2-3 times the space. Why? Question 2) is there a cleaner way to do this? I read recently where someone suggested rsync. I am investigating whether that is an option on the Solaris side. Use tar. tar czvpf - (some files, or directory) |ssh [EMAIL PROTECTED] cat my.tarfile.tgz -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: Alternatives to KDE/Gnome?
David Hart wrote: KDE seems to run better than Gnome w/RH9 (for me - YMMV). Nevertheless both environments seem to have gone way over the top. There MUST be alternatives that are less ponderous. I'm much more interested in efficiency and speed than pretty displays, animations or other visual effects that add nothing to usability. Suggestions? blackbox enlightenment sawfish http://www.plig.org/xwinman/ -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: Alternatives to KDE/Gnome?
Hal Burgiss wrote: On Mon, Jul 21, 2003 at 06:08:40PM -0400, David Hart wrote: There MUST be alternatives that are less ponderous. I'm much more interested in efficiency and speed than pretty displays, animations or other visual effects that add nothing to usability. You don't need either one. I run just plain Windowmaker, and it flies and does everything I need. Its fairly stable too: $ ps |grep wmaker hal 849 0.0 0.2 9724 2752 tty1 S Feb23 3:22 /usr/bin/wmaker GNOME/KDE are not essential, and in fact, are newish X add-ons. Windowmaker is not present in newer versions of RH. Even in older versions of RH it has not been maintained in RH for a long time. The menus are 80% filled with programs no longer in redhat. -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: Alternatives to KDE/Gnome?
Emmanuel Seyman wrote: On Mon, Jul 21, 2003 at 06:08:40PM -0400, David Hart wrote: Suggestions? Fvwm. http://www.fvwm.org/ Why not just tell him to do this: cat .xinitrc exec xterm ^c -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: [OT] Does this mean that IP was proven on the SCO case?
Edward Dekkers wrote: http://www.arnnet.com.au/index.php?id=1679444165eid=-100 Surely some clever programmers can just re-write the offending code so as not to breach IP? (from memory wasn't it the TCP/IP stack?) This will not save IBM, but surely then they would not be able to sue Linux users if their so called 'copied code' wasn't in the new versions? No one has proved anything legally or otherwise. A large (if not all) of the the code in question was written by IBM (or companies they bought)for various versions of Unix. Now SCO appears to be claiming they own this code. If this is true legally will depend on the contracts IBM (or companies they bought) signed. In any event this will take a long time to settle. I would suspect that IBM would just buy SCO if they really thought SCO had a case. PS- IANAL!!! -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: Kernel compile
David Mackay wrote: Thanks for your response. The problem with that is that 2.4.20 doesn't contain all the patches required for my chipset. I can do back to a RH kernel like 2.4.20-8 and have PCMCIA but no USB, or compile 2.4.21 and get USB but no PCMCIA! There must be a scripting/startup problem that i can work around with 2.4.21, because as i say PCMCIA works if i start manually. http://www.wsu.edu/~ice124/ http://videl.ics.hawaii.edu/mailman/listinfo/linuxpresario900 -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: RedHat 2.1 ES and Compaq DL 380 G3
SAQIB wrote: Hello, We are trying to load RedHAt 2.1 ES on a Compaq DL 380 G3. Compaq DL 380 G3 have a Gigabit NIC on it. Seems like RedHAt 2.1 ES does not install the driver upon initial installation, so we installed the driver seperately. The network works upon booting, but after approx 20 mins it dies, and the servers need to be restarted. So i installed RedHat 9.0, and seems like there are no issues with 9.0. Anyone has any ideas on how to fix the problem with RedHat 2.1 ES? -Make sure you are running the latest errate redhat kernel for AS. -Make sure you are using the latest e1000 driver. For example there is an e1000_4412k1 driver in the new revs of AS. -If none of this work grab the latest e1000 driver from intels site and compile it form your kernel. -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: Failures on rpm -Fv updates
Michael Schwendt wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Thu, 10 Jul 2003 17:10:32 -0700, Samuel Flory wrote: It's really bad when the mkinitrd fails, and you can't create a new as can't load the loop module. Is that still the case? rpm --query --scripts kernel The scripts can still fail to create a correct initrd if you've screwed up your modules.conf. -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: problem with raid
Nicholas wrote: Hi folks : Could anybody show me on how to guide me here in step by step configuring my system with raid controller here? Keep in mind the raid controller in question is not really a raid controller. It's an ide controller with a few bios features to build an array from the bios, and boot from a raid array. The actual raid occurs in the driver. 1st problem : I am using a intel 845wd1 motherboard with 2 maxtor 60 gig hard disk i plan to mirror the OS + 40 gig data from hd 1(hde) to hd (hdg), i have try few times, but i dont think i manage to do it. the experience : 1. i configured raid 1 PDC20267 with 2 cables, each set to master, means one of the controller will be master controller. 2. then i install linux, i can see the partition in the menu... it shows hde and hdg. because of i want to mirror hhe to hdg, so i did disturb or create any partition in hdg. 3 then i proceed with installation. 4 after i reboot the PC, it just cannot boot, it said grub error 22 please guide me on this or correct me on this. 2nd experience : 1. i just clean the entire hd and then i did a fresh installation. 2. when finish doing it, i rebuild the mirror which from the source to the destination. 3. it boot successfully, but the problem is, the hd just booted with read only mode,' help!! well, now i have a system with following the first experience, and i still cannot boot, i also lack of knowledge then just didnt rebuild the mirror. for booting, i created a CD to boot. please show me where i gone wrong. i am using a redhat 9.0 Use the linux software raid driver it's much better: 1)Create 2 arrays in the promise. Each array should consist of a single drive. This will give your bios boot devices. 2)Boot the installer and use the native linux software raid support. Remember that software under linux is per partition not per disk. 3)Be sure to use the lilo bootloader instead of grub. Grub really doesn't do software raid support. -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: RHAdvanced Server
Jacques Lederer wrote: Hello, I have the Red Hat Advanced Server, and there are 2 or 3 very specific problems which seem to be related to the distribution. I don't know if anybody here can help... 1) the e100 interface driver (This is a network card incorporated in a pentium IV motherboard) does not work. 2) the bonding doesn't work. This has to do with a kernel module, there is a bonding.c file and an if_bonding.h in the source tree, but something is wrong with it. 3) the kickstart installation doesn't work. I checked in various ways. I also checked it with a simple installation and a kickstart file generated by kickstart configurator. Each time it comes to an error message about /usr/bin/anaconda, or something like that. The Advanced Server runs on a 2.4.9-e3 kernel, and it seems hard to get updates and stuff for that sort of kernel. Don't you have access to the rhn? The current kernel is 2.4.9-e.25. The rhn is slow, but you can get updates thru it. I have also been unable to report these. I tried bugzilla, but it just would never accept my submission, each time I just get back a blank page. What is the procedure to get any support on these? Jack If you have a valid AS entitlement go RHN, login, and find the instant iso section. Grab the latest of the quarterly updates for AS 2.1. These have driver updates that may fix your issue. rhel-21-as-u2-i386-disc1.iso rhel-21-as-u2-i386-disc2.iso -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: Failures on rpm -Fv updates
Mitchell K. Smith wrote: You can't (shouldn't) do it that way. Put all your update files into /var/spool/up2date This is the way I do the updates manually. Maybe someone else has a better way but this has worked for me for several years. 1. determine which kernel rpm you need and get it from the appropriate directory. i386, i686, etc. 2. from the /var/spool/up2date directory, run rpm -Fvh kern*.rpm Make sure you have ONLY the required version of the kenel in there and not both i386 i686 versions. 3. After the kernel is updated, reboot the machine. 4. go back to the /var/spool/up2date directory and remove the kernel files. 5. update everything else with rpm -Fvh *.rpm An alternative method is to just dump all the update files into the /var/spool/up2date and then run the up2date utility. Red Hat Network from the Gnome menu. Then it will only download the header files which are small. Up2date will see that you already have the rpm's in the up2date directory and it won't try to DL them again. I hope this helps. Mitch Smith Or rpm -Uvh apt*.rpm apt-get update apt-get upgrade apt-get dist-upgrade http://freshrpms.net/ -Original Message- From: Greg Bell [mailto:[EMAIL PROTECTED] Sent: Thursday, July 10, 2003 3:38 PM To: [EMAIL PROTECTED] Subject: Failures on rpm -Fv updates Hi Gang, I grabbed the RH9 updates directory, and am doing a rpm -Fv * There are a few disturbing errors, most notably this one: glibc-2.3.2-27.9 error: %post(glibc-2.3.2-27.9) scriptlet failed, exit status 121 This is actually my second go at it - last time caused my system to be unbootable (init disabled all the runlevels because they were spawning too fast... possibly because none of the executables worked anymore?) Has anybody looked into this? Thanks, -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: Failures on rpm -Fv updates
Ed Wilts wrote: On Thu, Jul 10, 2003 at 12:38:08PM -0700, Greg Bell wrote: I grabbed the RH9 updates directory, and am doing a rpm -Fv * There are a few disturbing errors, most notably this one: glibc-2.3.2-27.9 error: %post(glibc-2.3.2-27.9) scriptlet failed, exit status 121 It sounds like you grabbed the i386 update directory when you don't have an i386 package installed. That's a good way to shoot yourself in the foot. Actually the fun way to shoot yourself in the foot is the reverse on an old pentium system;-) -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: Failures on rpm -Fv updates
Michael Schwendt wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On Thu, 10 Jul 2003 16:43:22 -0400, Mitchell K. Smith wrote: So what happens if you update the kernel with rpm -Fvh ? It uninstalls the currently installed kernel package and removes the kernel module files for the currently running kernel. In case the still running kernel needed to load a module, it would fail. And imagine the new kernel doesn't work at all or some of its drivers don't work, e.g. your network driver. With rpm -ivh you keep the previous kernel packages installed as backup kernels. It's really bad when the mkinitrd fails, and you can't create a new as can't load the loop module. Or you find that the kernel doesn't work on your hardware. -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: Failures on rpm -Fv updates
Mitchell K. Smith wrote: I've never used apt before. Is this really safe? It's worked well for me for over 6 months. As I live/work on the West Coast in SV the rhn is insanely slow compared to apt-get or a rh mirror. Also I can easily cache the the rpms for either. traceroute to mirrors.kernel.org (204.152.189.120), 30 hops max, 38 byte packets stuff 6 66.250.7.201 (66.250.7.201) 10.008 ms 9.741 ms 9.678 ms 7 g1-0.core02.sfo01.atlas.cogentco.com (66.28.6.5) 9.927 ms 10.567 ms 12.391 ms 8 g50.ba01.b003070-2.sfo01.atlas.cogentco.com (66.28.66.142) 11.659 ms 61.378 ms 74.705 ms 9 isc-2.demarc.cogentco.com (66.250.6.54) 146.461 ms 163.465 ms 70.194 ms 10 mirrors.kernel.org (204.152.189.120) 92.924 ms 16.070 ms 10.464 ms For beginners especially I would say to stick with the updates that up2date grabs or that you DL from the Red Hat updates ftp server. Keep in mind unless you start installing new package that are freshmeat only you are getting the exact same updates. But to each his own Mitch Or rpm -Uvh apt*.rpm apt-get update apt-get upgrade apt-get dist-upgrade http://freshrpms.net/ -Original Message- From: Greg Bell [mailto:[EMAIL PROTECTED] Sent: Thursday, July 10, 2003 3:38 PM To: [EMAIL PROTECTED] Subject: Failures on rpm -Fv updates Hi Gang, I grabbed the RH9 updates directory, and am doing a rpm -Fv * There are a few disturbing errors, most notably this one: glibc-2.3.2-27.9 error: %post(glibc-2.3.2-27.9) scriptlet failed, exit status 121 This is actually my second go at it - last time caused my system to be unbootable (init disabled all the runlevels because they were spawning too fast... possibly because none of the executables worked anymore?) Has anybody looked into this? Thanks, -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: How to find RPM from command
Sambit Nanda wrote: is there any way to find out the RPM name from the command name, mean if i want to know this command belongs to which RPM ? can i do that ?? example how i will come to know named belongs to which RPM ? whichrpm: #!/usr/bin/perl -w foreach $file (@ARGV){ $path =`which $file`; system rpm -qf $path; } Example: [EMAIL PROTECTED] tests]$ whichrpm ls top apt-get coreutils-4.5.3-19 procps-2.0.11-6 apt-0.5.5cnc6-fr1 -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: Suspending running job
Jeff Kinz wrote: On Tue, Jul 08, 2003 at 08:11:16PM -0700, Samuel Flory wrote: Use the man pages folks man kill man 7 signal What really kills me here is that I looked at those man pages just before I posted, assuming that the feature had to be in Linux by now. That's pretty much why I looked it up. I was in a bad mood yesterday. But signal 19 is the STOP signal, not the Pause signal, so I assumed (Yeah, I know..) that no pause/resume existed. However, a rose by any other name would smell as sweet. But the man pages are always so clear;-) And it's obvious that the correct command to send arbitrary singal to porccess would be kill. kill -s 19 pid killall -s 19 command name Thank you for the correct information Sam. -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: Antivirus for Linux
Benjamin J. Weiss wrote: Have you tried ClamAV its in the very list a virus scanner ... (http://www.clamav.org) I was under the impression that F-prot only scanned for M$ virii. Does clamav scan for M$ virii, *nix virii, or both? Linux virii? Name one? Yes there are a number of worms, but no real virii. -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: Recreating stock kernel
Steven W. Orr wrote: On Tuesday, Jul 8th 2003 at 09:52 -0700, quoth Greg Bell: = =Hi List Folks, = =First, let me point out that I am fairly seasoned at kernel recompiling... =I have done dozens successfully, so please don't let the naive question =below cause you to send me to the Howto :) = =With my new RH9 install, I figured it'd be best to recreate the stock =kernel, as a starting point. I thought I should be able to: = =1) cp /boot/config-2.4.20-8 .config =2) make dep etc etc = =However, this results in a gcc error and an aborted compile. Shouldn't I =be able to use that config? = =Second, if I run make xconfig, and don't do anything but save and exit, =it drastically changes the .config I copied from /boot. Not just the =order... new options there, etc. = =What critical piece of RedHat knowledge am I missing? = =Thanks, I went through the whole thing myself and noone was really helpful. So I shall look down you mere mortals and give you the answer ;-) There are a number of 'config' targets: config, menuconfig, xconfig and oldconfig. If you want to use an old config file with no changes, you should run oldconfig to make sure that all the stuff that's in there is good and that all the stuff that isn't gets added. But the point is that all of the config targets do more than just set up the .config file. There are other mysterious things that happen as well. The real proper way to do it is to unpack your src (whatever feels best) and then run make mrproper. (BTW, mrproper is apparently from the German version of Mr. Clean in this country.) Then you must run *one* of the config targets to end up with the proper pre-build setup. Note that if you have a good .config and you run mrproper, it *will* delete your .config file. Zat help? No and I know what you are talking about;-) The short answer. You need to run make oldconfig, make menuconfig, make xconfig to setup a number of things up. The following should nearly always compile a kernel: 0)Edit EXTRAVERSION value in Makefile so you don't call this kernel the same name as one of your other kernels. 1)Backup .config if you still want it 2)make mrproper (clean everything up) 3)copy a config file to .config 4)make oldconfig (Asks you any new questions if your config is out of date. Handy if you are using a config from older kernel, and want to know whats new. Don't do this you only if you didn't copy a config in step 1.) 5)make menuconfig Note- make xconfig should work, but kernel hackers rarely test that xconfig works. Linus is known to release kernels with only make oldconfig working;-) 6)make dep 7)make bzImage 8)make modules 9)make modules_install 10)make install (This may break if you have misconfigured lilo/grub, or modules.conf, but in will install your kernel, and System map in /boot.) In theroy only steps 2, one of either 4 or 5, 6, and 10 are required. Thus a more typical install would be: 0)Edit EXTRAVERSION value in Makefile. 1)Backup .config if you still want it 2)make mrproper (clean everything up) 3)copy a config file to .config 4)make oldconfig/menuconfig/xconfig 5)make menuconfig 6)make dep install -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: Disabling grub boot-selection screen
-{ Rene Brehmer }- wrote: Hi gang How do I disable the grub boot-selection screen? ... In a way so that I can reactivate it once I get going on my custom kernel ... Right now it serves no purpose, as the machine has only 1 kernel to load anyways. I've changed the boot-timeout in grub.conf to 0, but it still spends time loading the graphical screen, and as soon as it's displayed it moves on ... I just want it to skip the selection screen completely, but still allow to run it once I need it for custom kernel testing ... TIA Rene Remove the splash screen. splashimage=(hd0,0)/grub/splash.xpm.gz -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: Suspending running job
Jeff Kinz wrote: On Tue, Jul 08, 2003 at 05:57:01PM -0600, Ashley M. Kirchner wrote: Okay, assume a user started a task on the server, and let it run. Assume I have root access to the server. Is there some way that I can pause the task, and resume it later without screwing up permissions? Although this is a very reasonable thing to want to be able to do and it seems quite doable in terms of what would be needed in the kernel there is currently no way to do it in Linux. Time to add a feature suggestion to the kernel list. I think that this feature makes a lot of sense in both data center and desktop environments. (when used with appropriate caution) Use the man pages folks man kill man 7 signal kill -s 19 pid killall -s 19 command name -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: 2.40-20-20-1.2013 Compilation Problem
David Hart wrote: This seemed to compile, boot and load modules just fine. However, it won't load the X server. Can anyone suggest what I screwed up? This was my first attempt at this endeavor. BTW, what does the nptl designator mean? /usr/share/doc/redhat-release-9/RELEASE-NOTES-i386: Red Hat Linux 9 includes the Native POSIX Thread Library (NPTL), a new implementation of POSIX threads for Linux. This library provides performance improvements and increased scalability for i686 or better processors. aka the new thread library that breaks things;-) -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: 2.40-20-20-1.2013 Compilation Problem
David Hart wrote: This seemed to compile, boot and load modules just fine. However, it won't load the X server. Can anyone suggest what I screwed up? This was my first attempt at this endeavor. Whats the output of startx output BTW, what does the nptl designator mean? -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: grub and raid1
Ian Mortimer wrote: And what occurs if hda dies. You can't boot!!! Either: install grub in the MBR of each disk (before) make a boot floppy (or two) (before) boot off the CD or a rescue disk and install grub (after) You make it sound simple. Have you even tried it from the installed system. If so please tell us how it can be done? -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: Binary Compatibility
Haley Crowe wrote: Hello. I'm looking at upgrading to one of the newer versions of Redhat from 7.x. Can anyone tell me how much binary-compatibility there is from 7.x to 8 or 9? In the old days Red Hat maintained binary compatiblity between version that shared a major number. Now they seem to be okay with breaking compatibility every release. You still may be able to install older binaries if you install the compat- rpms. For RH9 be sure to read the release notes about NPTL. -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: grub and raid1
Ian Mortimer wrote: [EMAIL PROTECTED] /]# /sbin/grub-install /dev/md2 /dev/md2 does not have any corresponding BIOS drive. [EMAIL PROTECTED] /]# /sbin/grub-install --recheck /dev/md2 Probing devices to guess BIOS drives. This may take a long time. /dev/md2 does not have any corresponding BIOS drive. Grub gets installed in the MBR of a physical drive (there are other ways but that's the most common). You can't install it in a raid partition. Do: /sbin/grub-install /dev/hda (or whatever is your primary hard disk - the one the BIOS boots from). And what occurs if hda dies. You can't boot!!! This is why I always use lilo for raid 1 systems. Lilo actually knows what to do if you tell it to use an md device. Grub is cool, but lilo works;-) -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: Muliti-processor supports...?
Win Toe wrote: I would like to know the no of multi-processors that Red Hat supports Thanks in advance In theroy the 2.4 kernel supports a rather large number of cpu. (In the range of 32 or 64. I've forgotten as I've never used more than 8.) Scaling is another matter. Depending on what you are doing linux tops out around 4-8 cpus. Of course this all changes with numa under 2.6 kernels. -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: RHN servers are overloaded...
Ricky Boone wrote: On Thu, 2003-06-26 at 17:48, Panos Tsapralis wrote: ...so, what other options do I have to download the latest patches without using the up2date mechanism? For example, can I download them via ftp into a local directory on my system and install them as regular RPM packages? Find a mirror close to you. http://www.redhat.com/download/mirror.html Do you have Basic (or Enterprise) entitlements? You shouldn't be receiving messages like that unless you're running on Demo entitlements... at least that's been my experience with the service. If you're doing that, I recommend getting even a Basic license. RH has lowered the price considerably. I don't know I was getting 16K for an AS entitlement last week. I get x4 that from a nearby by RH mirror!! An alternative would be Red-Carpet, which is provided by Ximian, though you may run into the same problem of servers being overloaded. -- Once you have their hardware. Never give it back. (The First Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: [LISTADMIN] Great - just another spam block...
Joseph A Nagy Jr wrote: snip One could always get an email client with spam filtering capabilities built in (like Mozilla) It just takes a while teach mozilla that all of your mailing lists aren't spam;-) -- There is no such thing as obsolete hardware. Merely hardware that other people don't want. (The Second Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: Optimize Redhat9
Josep M. wrote: Hello! I have Redhat9 in a laptop 128mbRAM 10GB hb,and runs a little slow,too much swapping,I optimized kde settings and is running a little better,but too much swapping...I would like ask about steps that can I do for optimize performance. 128M isn't enough to run kde and anything else. Don't run kde or gnome. Try blackbox, sawfish, or the like. Be sure to install and use less memory hungry apps. Galeon instead of mozilla, abiword instead of open office. -- There is no such thing as obsolete hardware. Merely hardware that other people don't want. (The Second Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: What distribution works well with old computers?
Marcos de Souza Trazzini wrote: WOW !!! or you're too patient, or you really optimize your Linux BOX running on a Pentium 120... Congratulations On Wed, 2003-06-04 at 14:13, Patrick Nelson wrote: On Wed, 2003-06-04 at 09:49, Marcos de Souza Trazzini wrote: Well... this is a RedHat mailing list... hehe I'm using a RedHat 7.2 box with a recompiled kernel in a Pentium 233MMX and it was working well. But i had to recompile the kernel with many customized options for best performance with this machine. If you don't want to recompile the kernel, try to use a previous version of the RedHat, like the 6.2 (kernel 2.2). On Thu, 2003-06-05 at 00:42, Paul Sutcliffe wrote: What distribution works well with old computers? like a pentiumMMX 233MHz Have RH9 installed on a P120 with 198MB RAM and a 8MB video. It's not hard I've got a laptop P /w 64M of memory running RH9. I even run X on it sometimes. My .xinitrc spawns an xterm, and I run either abiword or galeon. The install works just fine in text mode. If you really want to run X on an older system try running blackbox, or evilwm. -- There is no such thing as obsolete hardware. Merely hardware that other people don't want. (The Second Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: ReiserFS good idea on RH9?
Stephen Kuhn wrote: On any given day, I'm dumping customer's HD contents to a ReiserFS partition on my server - anywhere from a few gigs upward of 50-60gb - along with the daily bits of moving around MP3's, AVI's, MPG's and SVCD's - when I was using ext3, I had a much slower go of it, but since using ReiserFS, it's definitely been smoother. This is a network file server and all the user homes reside on ReiserFS, IMAP lives on ReiserFS; I had been a proponent of ext3 for quite some time until I tested ReiserFS for that reason - now I can't/won't go back. Keep in mind that the default ext3 journaling mode is slow as hell. Mounting a partition with data=writeback greatly improves performance. -- There is no such thing as obsolete hardware. Merely hardware that other people don't want. (The Second Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: hibernate
João Borsoi Soares wrote: I'm using RH 7.3, kernel 2.4.18-26.7.x. But it seems not to suport full hibernate: from man apmd I got: suspend [ system | user ] Invoked when the APM driver reports that system suspension has been initiated. The second parame ter indicates whether the BIOS or a user action (such as closing a laptop) initiated suspension. The BIOS suspend mode aggressively conserves power, and normally involves shutting off power to all devices except the CPU core and memory, which is kept in a very low power mode. Most laptops can stay suspended, using battery power alone, for sev eral days. (Hibernation is a kind of super-sus pend, where all that state is written to disk and the machine uses even less power bcause it can turn off that CPU core, using no battery power at all. here== At this writing, Linux does not support hiberna tion.) PCMCIA devices should be manually suspended using cardctl(8), and some modular drivers may need to be unloaded. But this writing is from Jun 1999. So, now it may work. But how do I invoke it from bash? The real question is it a hardware suspend, or a software suspend? A hardware suspend will restore from the bios, and will have a suspend partition. A software suspend will start booting the OS before resuming, and will not have a suspend partition. In the case of a hardware suspend it should just work. You setup bios is setup to suspend after x amount time in suspend mode. My old sony viao works like this, and inaddition has a hotkey to do this. My newer sony viao is a pure software suspend box, and doesn't work under linux. The only option is the software suspend in the 2.5 kernel or the laptop kernel patch. You results may vary. http://savannah.nongnu.org/projects/laptopkernel/ http://sourceforge.net/projects/swsusp http://fchabaud.free.fr/English/Tricks/Laptop/Swsusp/Doc/Software-suspend.html -- There is no such thing as obsolete hardware. Merely hardware that other people don't want. (The Second Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: ReiserFS good idea on RH9?
Christopher Henderson wrote: I had never used it with RedHat, but I have used it since it first appeared in Mandrake at least two years back or more. Stable? I've never had a problem, never lost data. I wonder why RedHat doesn't offer the customer the choice? Label Ext3 as prefered or recommended but still allow us to choose for ourselves. You can actually install it by typing linux reiserfs at the initial prompt. Then it sound up as an option. -- There is no such thing as obsolete hardware. Merely hardware that other people don't want. (The Second Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: Finding Software and Support for RedHat Linux 8.0
[EMAIL PROTECTED] wrote: Hi gang... I just purchased RedHat Linux 8.0 last year (and I'm slowly realizing I probably should have waited for Linux 9.0), and I'm having the hardest time finding software that supports it. Case in point, I also purchased Kylix for Linux last year but didn't realize that 8.0 doesn't really support it. Now I have a $300+ piece of software that I can't use on my system. Currently I'm trying to find a version of Forte for Java that will work on my system and I've found that IT only supports RedHat 7.2. Is there ANY IDE (or ANY SOFTWARE ) that supports RedHat 8.0, out there. Or did I just shoot myself in the foot with the purchase of RedHat 8.0? Andre--- What feature are you looking for exactly. The Red Hat 8.0 cdroms include Kdevelop, and Xemacs which aren't generally insalled. There are a number that should compile under RH8.0: http://www.eclipse.org/ http://www.newplanetsoftware.com/jcc/ Or check freshmeat: http://freshmeat.net/browse/65/?topic_id=65 -- There is no such thing as obsolete hardware. Merely hardware that other people don't want. (The Second Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: Issue installing XINE
Mathieu Masse wrote: Another issue when installing RPMs. I am trying to install xine-0.9.21-fr1.i386.rpm and xine-lib-1.0.0-fr0.beta12.1.i386.rpm. WHen I install xine-lib it asks for the following library: glut-3.7-9.i386.rpm, which is no problem, I got it; but when I install that one it says I need OpenGL. From the OpenGL.org website I understand that I should already have this with X-Free. What gives? How come I get the dependencie. Also where can I get it. Get apt from freshrpms, and apt-get it. http://freshrpms.net/ [EMAIL PROTECTED] cujo]# apt-get install xine xine-skins Reading Package Lists... Done Building Dependency Tree... Done The following NEW packages will be installed: xine xine-skins 0 packages upgraded, 2 newly installed, 0 removed and 0 not upgraded. Need to get 0B/3135kB of archives. After unpacking 4371kB of additional disk space will be used. Executing RPM (-Uvh)... warning: /var/cache/apt/archives/xine_0.9.20-fr2_i386.rpm: V3 DSA signature: NOKEY, key ID e42d547b Preparing...### [100%] 1:xine ### [ 50%] 2:xine-skins ### [100%] [EMAIL PROTECTED] cujo]# No issues here;-) -- There is no such thing as obsolete hardware. Merely hardware that other people don't want. (The Second Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: Adaptec 21960 on RH 7.3, 8 or 9
John Lang wrote: I've had no luck loading RH on any of my servers with running Adaptec 29160. Install starts to load the driver aic 7xxx, then immediately reboots. What is the hardware involved? What is going wrong with the installation? (Check the debugging screens on ctl-alt-F3, ctl-alt-F4, ctl-alt-F5.) Any suggestions? Try a driver's disk from Justin. If that doesn't work send a detail bug report to Justin Gibbs. He's adaptec's linux/bsd driver maintainer. http://people.freebsd.org/~gibbs/linux/DUD/aic7xxx/ Also try using the expert mode, and see if there is a aic7xxx_old to try. -- There is no such thing as obsolete hardware. Merely hardware that other people don't want. (The Second Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: how to change display
Julie Xu wrote: Hi, I have installed RadHat 9 on my Toshiba laptop without my Philips 170B2 screen. After the installation, I can not see anything using the screen. How can I reconfigure the monitor/display after the installation. Many comments will be appreciated You can try the following: 1) at the grub prompt -hit e -high light the line starting with kernel -hit enter -add init=3 to the end -hit enter -hit b (or if you used lilo just hit control-x and type linux init=3 at the lilo prompt) 2)login and run redhat-config-xfree86 Or try 1)stick in your Red Hat install cdrom 2)type linux rescue at the intial prompt 3)at the prompt type chroot /mnt/sysimage 4)run redhat-config-xfree86 5) reboot -- There is no such thing as obsolete hardware. Merely hardware that other people don't want. (The Second Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://www.redhat.com/mailman/listinfo/redhat-list
Re: SiI 3112 SATA Raid Controller
Thomas Tolborg wrote: Hello I would like to know how to install redhat 9 on a system with SiI 3112 Raid controller. I have created my raid array in the SiI 3112 configuration. But it seems that Redhat 9 don't have drivers for this controller. Redhat 9 just treats my disks as two seperate disks. I can see on http://www.siimage.com homepage that linux is actually a supported os, but I can't seem to find drivers or anything that will help me. The SiI 3112 controller is onboard on an ASUS P4G8X motherboard. ASUS don't have drivers for the controller either. This controller isn't really a raid controller. It's a sata controller with a few bios hooks to allow you to boot off the raid arrays. All of the raid is done in the driver. You are really better off treating the controller as an ide controller and using linux software raid. That said in theory if you load the ataraid, and silraid modules it will see the arrays. This is not supported in the install however. -- There is no such thing as obsolete hardware. Merely hardware that other people don't want. (The Second Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://listman.redhat.com/mailman/listinfo/redhat-list
Re: how to format hda1
lito lampitoc wrote: Hi All, I have the following partition: /dev/hda1 FAT (Win) /dev/hda2 swap /dev/hda3 Linux I want to convert FAT partition (hda1) to a Linux partition but everytime I do an fdisk it shows me the following error. WARNING: Re-reading the partition table failed with error 16: Device or resource busy The partition table hasn't been updated. This is due to the fact that you have partitions in use on hda. All you need is a reboot to update the partition table. These are the steps I did: 1. unmount /dev/hda1 2. delete /dev/hda1 3. create a new partition for /dev/hda1 4. while writing it shows the above error other method: 1. boot on linux rescue 2. do steps 2 - 4 above. But still no success. Any help appreciated. Thanks All you really need to do is mke2fs the partition. `mke2fs -j /dev/hda1` PS- You should also toggle the partition to linux. (t, 1, 83) -- There is no such thing as obsolete hardware. Merely hardware that other people don't want. (The Second Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://listman.redhat.com/mailman/listinfo/redhat-list
Re: Uploading from digital cameras and movie cameras
Art Ross wrote: I'm planning to build a box for Linux. I want to be able to download pictures from a digital camera and movies from a digital movie camera. What types of hardware will I need for these capabilities? I'll be running a dual boot between RedHat and maybe Lindows. Thanks in advance, There are 2 ways this is done: 1)Many cameras like my sony are seen as usb mass storage devices. 2)Others need a user space program to download the images. Generally gphoto2, or a gui frontend like gtkam. I'd just try gtkam, and if that doesn't work type dmesg on the command line. If dmesg indicates the kernel dtected a usb mass storage device. Try mounting the device it detected. (example mount /dev/sda1 /mnt/camera) -- There is no such thing as obsolete hardware. Merely hardware that other people don't want. (The Second Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://listman.redhat.com/mailman/listinfo/redhat-list
Re: Full Duplex or Half Duplex Again?????
Rodrigo Nascimento wrote: Thanks Douglas and Rebecca... But I think which I didn't know bade me... I know which my ethernet card works with Full Duplex, but I need to know if the eth0 is running in Full or Half Duplex...ok... Thanks again... Rodrigo Nascimento Try |mii-tool eth0, or ethtool eth0. -- There is no such thing as obsolete hardware. Merely hardware that other people don't want. (The Second Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://listman.redhat.com/mailman/listinfo/redhat-list
Re: Multiple identical NFS mounts under RH8
Michael Mansour wrote: Hi, I think I've found a bug in RH8's NFS. I have 3 Red Hat machines, 2 of the RH8 and 1 RH6.2 On one of the RH8 machines, I export a filesystem, while the other RH8 and RH6.2 machines mount that exported filesystem. On the RH8 client, I can mount the filesystem many times as shown (df -k output): testserv:/data01 8744304 6903844 1840460 79% /mnt/testserv testserv:/data01 8744304 6903844 1840460 79% /mnt/testserv testserv:/data01 8744304 6903844 1840460 79% /mnt/testserv testserv:/data01 8744304 6903844 1840460 79% /mnt/testserv testserv:/data01 8744304 6903844 1840460 79% /mnt/testserv and a cat of my mtab: testserv:/data01 /mnt/testserv nfs rw,addr=203.14.211.9 0 0 testserv:/data01 /mnt/testserv nfs rw,addr=203.14.211.9 0 0 testserv:/data01 /mnt/testserv nfs rw,addr=203.14.211.9 0 0 testserv:/data01 /mnt/testserv nfs rw,addr=203.14.211.9 0 0 testserv:/data01 /mnt/testserv nfs rw,addr=203.14.211.9 0 0 while on my RH6.2 machine, when I try to mount the /mnt/testserv directory: mount: testserv:/data01 already mounted or /mnt/testserv busy mount: according to mtab, testserv:/data01 is already mounted on /mnt/testserv Has anyone else experienced this issue before? BTW, I can dismount each and every RH8 mount point one at a time also. Weird. This is normal. -- There is no such thing as obsolete hardware. Merely hardware that other people don't want. (The Second Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://listman.redhat.com/mailman/listinfo/redhat-list
Re: how to access Linux pc from another Windows PC
Limb wrote: Sorry for making you have to see that bad word.. but anyways... How can i conenct to my linux box from another pc and compile/edit C++ stuff on it? You should use ssh. There are a number of windows clients. I prefer putty or ssh from cgywin. Of course with cygwin you can just install gcc, and compile things from windows;-) http://www.chiark.greenend.org.uk/~sgtatham/putty/ http://www.cygwin.com/ Also how do i get my network assigned IP? i can't find it from any network/ethernet config editors. Try `/sbin/ifconfig` from a command prompt. I'm sure that there is some neat gui, but ifconfig always works. -- There is no such thing as obsolete hardware. Merely hardware that other people don't want. (The Second Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://listman.redhat.com/mailman/listinfo/redhat-list
Re: ReiserFS Support?
Robert Adkins II wrote: Will, With regards to ext3fs, I am aware of the journaling properties and the superior recovery it has in comparison to ext2fs. However, ext3fs has considerable overhead for writing data to the disk. ReiserFS is supposedly far swifter at disk writes, which I need to quell the minor revolt a few of my users are beginning to threaten. One thing to try is mounting the ext3 partition in writeback mode. This is faster, but not quite as fast reiserfs. It is possible to lose newly written data with writeback. Of course reiserfs, like most jfs, has the same issues. mount -o data=writeback /dev/foo /bar http://www.redhat.com/support/wpapers/redhat/ext3/tuning.html#journaling -- There is no such thing as obsolete hardware. Merely hardware that other people don't want. (The Second Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://listman.redhat.com/mailman/listinfo/redhat-list
Re: Out of Disk PSace to Insatll Kernel Update
Robert Love wrote: Thank you! Ben and Irwin I used RPM to remove some of the previous kernals then was able to run up2date -u to get the latest kernal :-) Did not want to remove anything to do with the kernal, if I did not know it was going to be safe. In general it's safe to uninstall all kernels that you aren't currently booted with. You should always keep the current kernel for 2 reasons. 1)That new kernel might not boot;-) 2)You might want to load a new module for your current kernel before you reboot. For example might want to mount a cdrom which requires the iso filesystem module. -- There is no such thing as obsolete hardware. Merely hardware that other people don't want. (The Second Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://listman.redhat.com/mailman/listinfo/redhat-list
Re: Ethernet speed
Ed Wilts wrote: On Tue, Mar 18, 2003 at 08:14:17PM -0300, Jefferson Granatto wrote: Sorry, I don't speak english well... I don't know you write better than I do some days ;-) How can I reduce the speed of ethernet to 10mbps on Red Hat 7.3? Your English is perfect - better than some of the requests from native English members! Check for /sbin/mii-tool. It's part of the net-tools package and allows you to force the NIC to 10mbs at either full- or half-duplex. .../Ed Mii-tool works for most ethernet cards. If it does not try ethtool -- There is no such thing as obsolete hardware. Merely hardware that other people don't want. (The Second Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://listman.redhat.com/mailman/listinfo/redhat-list
Re: Recover Raid 5 of Red Hat 7.3
Rus Foster wrote: On Thu, 13 Mar 2003, Hong Tian wrote: If one disk of Raid 5 is found bad, Could I just replace the bad disk and recover the system and data? Or Should I re-install the Linux system from scratch again and recover the data from the backup? The idea of RAID-5 is to survive this. Assuming you've only lost one disk then you should be able to plug in the spare and they system should rebuilt the data from the paratiy bits on the other disks. Well that might occur on a scsi raid controller with the drives on a saf-te back plane. With software raid you will need to replace the failed drive. Create new md partitions on the drive, and do a raidhotadd to the md partitions. Example I've got disk sda-sde, all disks are the same size and have the same partitioning scheme, disk sdd has failed, and been replaced. I have 3 md partitions md0/sd*1, md1/sd*2, and md2/sd*3 -fdisk -l /dev/sda (figure how the disk should be partitioned) -fdisk /dev/sdd (configure the new drive as the other drive) -cat /proc/mdstat (Check out what drive goes where.) -raidhotadd /dev/md0 /dev/sdd1 -raidhotadd /dev/md1 /dev/sdd2 -raidhotadd /dev/md2 /dev/sdd3 -- There is no such thing as obsolete hardware. Merely hardware that other people don't want. (The Second Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://listman.redhat.com/mailman/listinfo/redhat-list
Re: galeon faster?
Mike Vanecek wrote: On Sat, 8 Mar 2003 00:30:26 +0100, Emmanuel Seyman wrote On Fri, Mar 07, 2003 at 05:47:00PM -0500, Anthony E. Greene wrote: It feel faster on my machine too, probably because it uses Gtk instead of XUL. Gnome, actually. The Gtk-only browser is skipstone. As a long time GNOME user, I tried Galeon early on, and have never looked back. It's my favorite browser, hands down. Amen. The first thing I've done when installing a new RHL is replace the mozilla icon on the taskbar with a galeon one. I do not have it installed on my RH 8 system. To use it, would I uninstall Mozilla 1.3a and install Galeon? Are the prefs compatible or would I lose all of them. Do RH do a better job in keeping it current (as opposed to Mozilla which is rather behind)? The galeon rpm that comes with 8.0 can be installed without uninstalling mozilla. -- There is no such thing as obsolete hardware. Merely hardware that other people don't want. (The Second Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://listman.redhat.com/mailman/listinfo/redhat-list
Re: Kernel versions
Cannon, Andrew wrote: Hi all, I'm a newbie, and I was wondering what advantages there would be to upgrading the kernel on my RH8 boxes from the i386 version to the most current i686 and Athlon kernels? (I've got a couple of boxes in work that are PII and P4 and an Athlon at home). Are there any advantages or should I just stick with waht I've got at the moment? Redhat should be installing the correct version of the kernel on your system. On the other hand I always recommend keeping up to date with redhat eratta. [EMAIL PROTECTED] root]# echo `rpm -q kernel --qf %{ARCH}` i686 [EMAIL PROTECTED] root]# -- There is no such thing as obsolete hardware. Merely hardware that other people don't want. (The Second Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED] https://listman.redhat.com/mailman/listinfo/redhat-list
Re: ftp'ing directories
Ed Wilts wrote: On Tue, Feb 11, 2003 at 03:17:55PM -0800, Samuel Flory wrote: I believe that new rev of redhat installs lftp by default. (Which pisses off me as a ncftp user.) ncftp is still there on your CDs and should therefore also be available via up2date. Probably takes you 2 minutes to install it if you have a decent network pipe. Yes, but when you're dealing with a lot of systems installed by someone else. Many of which you lack root access to. It's a real annoyance. -- There is no such thing as obsolete hardware. Merely hardware that other people don't want. (The Second Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED]?subject=unsubscribe https://listman.redhat.com/mailman/listinfo/redhat-list
Re: ftp'ing directories
gabriel wrote: i highly reccomend installing ncftp. it's a commandline ftp program that supports stuff like this and it's far easier to use. it even comes with two other little programs that you can use to automate ftp puts and gets. There is also ncftp-get, ftpcopy, and lftp. I believe that new rev of redhat installs lftp by default. (Which pisses off me as a ncftp user.) -- There is no such thing as obsolete hardware. Merely hardware that other people don't want. (The Second Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED]?subject=unsubscribe https://listman.redhat.com/mailman/listinfo/redhat-list
Re: Linux redhat 8.2 and XP dual boot questions
Payam Fard wrote: Hi all, I have a Pentium 4 with 256 Meg of RAM. It is running Windows XP. Yesterday, I added another hard drive for linux installation. So, one hard drive has Windows XP and the other has Linux redhat 8.2. I am trying to make my system a dual boot system. Now here are my questions: Red Hat 8.2? The last release was 8.0. 1) I am not sure whether I have used Lilo or Grub during installation. How can I know which one I am using? When I boot to Linux, it shows a man with a redhat in the left half of the screen, but no indication of Lilo or Grub. If it says something about control-x for command line it's lilo. 2) How can I switch between Lilo and Grub? Generally redhat installs grub, and lilo config files. Install one into your mbr is as simple as typing lilo, or grub-install. 3) The choices that I get during the reboot are: Linux, Dos and Linux bak. I am not sure what Linux bak is for. When I choose Linux, redhat comes up fine and everything is cool. When I choose Dos, nothing happens. I tried to modify lilo.conf and grub.conf to add Windows XP to the list, but for some reason it does not take into effect. Do I need to run anything after modyfing these two config files, or reboot should pick it up? When I reboot, nothing is changed!!! Grub reads it's config file at boot from the file system. Lilo need to be rerun after evey change. 4) What should I add to my lilo.conf or grub.conf in order to boot into windows XP? What device is your windows XP partition? -- There is no such thing as obsolete hardware. Merely hardware that other people don't want. (The Second Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED]?subject=unsubscribe https://listman.redhat.com/mailman/listinfo/redhat-list
Re: Deleting a file
Ted Gervais wrote: Was wondering something here. Now and than I end up with a file that has '#' in front and back of it. ie: #filename#. No doubt that is caused by using MC and while MC can delete these files how does one do it from the command line?I have often tried to try a few approaches to removing them 'rm #* or rm *# and can't seem to delete them. How is this done from the commmand line without resorting to using MC.?? Tab completion in bash is your friend. rm #tab -- There is no such thing as obsolete hardware. Merely hardware that other people don't want. (The Second Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED]?subject=unsubscribe https://listman.redhat.com/mailman/listinfo/redhat-list
Re: Gnome 2.2
Caleb Groom wrote: Based on prior releases, is Red Hat likely to release an RPM update for Gnome 2.2 on RH8? without upgrading to RH8.1? I'm sure I'll upgrade without hesitation, just wondering if Red Hat will more than likely say if you want Gnome 2.2, you should run RH8.1. Pretty much zero chance of a 8.0 update. -- There is no such thing as obsolete hardware. Merely hardware that other people don't want. (The Second Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED]?subject=unsubscribe https://listman.redhat.com/mailman/listinfo/redhat-list
Re: ext3 writeback option difficulties
John Joseph Roets wrote: Hello guys. Has anyone successfully used the 'data=writeback' option on an ext3 partition [on Redhat 8.0]? I've perused the newsgroups and google, yet have only found a few references to this problem, with no answers/resolutions. When I try to mount the partition with the data=writeback option... # mount -o remount,noatime,defaults,data=writeback /dev/hda2 It returns 'mount: / not mounted already, or bad option'. I can successfully mount with the noatime option [and all other given options of course (already within fstab)]. These also have the same effect. # mount -o remount,noatime,defaults,data=writeback / # mount -t ext3 -o remount,noatime,defaults,data=writeback /dev/hda2 / I also want of course to get this into my /etc/fstab. Using: RH 8.0 linux-2.4.18-19.8.0 These changes are primarily for optimization for use with MythTV [www.mythtv.org]. TIA. You can't remount partition with a different journal type. What you need to do is create a new fstab and initrd based on that fstab that uses the right options. You should be able to edit your fstab and put the right options there and then create an initrd using mkinitrd. Or you could recompile your kernel with ext3 built into the kernel. (Why RH leaves it a module is beyound me.) Then all you need to do is add rootflags=data=writeback to the kernel command line. -- There is no such thing as obsolete hardware. Merely hardware that other people don't want. (The Second Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED]?subject=unsubscribe https://listman.redhat.com/mailman/listinfo/redhat-list
Re: ext3 writeback option difficulties
John Joseph Roets wrote: I am not remounting the partition with a different journal type. It is already ext3. There are 3 journal types/modes ordered (the default), writeback (fastest), and journal (don't use). I need only to be able to add the data=writeback option to the fstab. Alternatively, I would even remount the partition in rc.local if need be (if module load order is a problem), if only it would accept it. I'll repeat it one more time. What you want to is NOT possible. Once a partition is mounted as one journal mode you can't remount it as another. You can umount it and mount it, but this can't be done on /. (You can't umount /.) You need to mount / in writeback mode at boot. The only way to do this if you use an initrd is to make sure linuxrc in your initrd is using the right options. If you are using a kernel with ext3 compiled in you need to give the rooflags option. -- There is no such thing as obsolete hardware. Merely hardware that other people don't want. (The Second Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED]?subject=unsubscribe https://listman.redhat.com/mailman/listinfo/redhat-list
Re: ext3 writeback option difficulties
Michael Schwendt wrote: I need only to be able to add the data=writeback option to the fstab. Feel free to add it there, then reboot because you cannot change the journaling mode with remounting. Actually as I remember just adding the writeback option is really bad. I seem to remember that the remount of / as rw fails. With 8.0 (maybe earlier) mkinitrd is smart enough to read the fstab for the options you use to mount /. -- There is no such thing as obsolete hardware. Merely hardware that other people don't want. (The Second Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED]?subject=unsubscribe https://listman.redhat.com/mailman/listinfo/redhat-list
Re: NFS question
Thomas E. Dukes wrote: After I sent that I thought I may have used the wrong terminology. I am using disk and network files systems under webmin to mount a directory through the network. I think its using samba. I can mount a shared directory, just can't figure out how to do the whole drive -- other than by individual directories. I can see the shared drive (in webmin), it just won't mount. You can have a samba mount (or nfs mount) only for a directory, and any dirs under it. I guess you could export /. You may need special options to span filesystems and follow links. -- There is no such thing as obsolete hardware. Merely hardware that other people don't want. (The Second Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED]?subject=unsubscribe https://listman.redhat.com/mailman/listinfo/redhat-list
Re: XFS with RedHat?
Alexander Skwar wrote: Hi! As I'm thinking about switching to RedHat, I'd also like to know, if RedHat supports XFS (the filesystem)? I'd need this, because I'd like to have Samba shares with ACL and XFS is the only FS supporting ACLs isn't it? There is a iso for 7.3, and there will not be a 8.0. ftp://oss.sgi.com/projects/xfs/download/latest/installer/installer/i386/ Keep in mind Red Hat doesn't support xfs, jfs, or anything other than ext3. -- There is no such thing as obsolete hardware. Merely hardware that other people don't want. (The Second Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED]?subject=unsubscribe https://listman.redhat.com/mailman/listinfo/redhat-list
Re: Duplicating a drive
Gerry Doris wrote: I have a single 30GB drive that is getting full. I want to put in a second larger hard drive and duplicate the first one, then remove the first drive and replace it with the new larger drive. I'm worried that this first drive contains the operating system. Can it be duplicated while it is mounted and active? Can do this with dd? What should be the correct command? You can use dd, but... 1)It's not a good idea on a live filesystem. 2)It's only really usefull on a live filesystem 3)You may end up coping a bunch of empty sectors. Try this: -install new drive -create new partitions on the disk. (via fdisk or parted) -create swap and filesystem on the new drive (via mke2fs, and mkswap)* -mount the new filesystem on /mnt/cdrom (mount /dev/??? /mnt/cdrom) -copy the files from your current installation to the new drive cd / tar -czvpf - . --exclude /proc --exclude /mnt/cdrom |tar -xzpf - -C /mnt/cdrom -make sure you have a useable emergency floppy (mkbootdisk) -remove your old drive and boot from floppy -rerun your installer *Be sure to create label on the file systems as redhat uses them instead of absolute device names. You can do this as part of your mke2fs command, or with e2label. -- There is no such thing as obsolete hardware. Merely hardware that other people don't want. (The Second Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED]?subject=unsubscribe https://listman.redhat.com/mailman/listinfo/redhat-list
Re: Memory Leak
Robert Adkins wrote: Kevin, I am far from an expert in this manner. I was seeing a similar thing happening with my own server. Using the 'top' utility, I didn't see any process gobbling up memory, even over time. The largest process that I had was and still is Squid, which sits at a nice 17M of usage. The rest of the processes come and go as users logon and off the PDC this server also acts as. My baseline memory usage is just about 1GB of memory. It stays consistent at that for at least 69 days. (My last uptime, taken down for hardware maintenance not any other reason.) The most of the memory being used is being used as a buffer roughly 208MB and as Cache roughly 600 MB. Which leaves virtually all of the memory being used on the server accounted for. Okay folks memory used for buffer, and cache is free memory!!! If you need to use it the kernel will discard the buffer. example: root@sflory tmp]# free total usedfree shared buffers cached Mem:514836 508620 6216 0 112976 218312 actual free -/+ buffers/cache: 177332 337504 ^^ free not counting buffer, and cache Swap: 1726940 1583001568640 Open up a bunch of xemacs, open office, kword [root@sflory tmp]# free total used freeshared buffers cached Mem:514836 510444 4392 0 81000 184124 -/+ buffers/cache: 245320 269516 Swap: 1726940 1537921573148 close the programs [root@sflory tmp]# free total used free shared buffers cached Mem:514836 486992 27844 0 80752 220692 -/+ buffers/cache: 185548 329288 Swap: 1726940 1537761573164 [root@sflory tmp]# cat /pub/iso/enigma-i386-disc* /dev/null let it run for a while and hit control-c [root@sflory tmp]# free totalused free shared buffers cached Mem:514836 511492 3344 0 8312 357668 -/+ buffers/cache: 145512 369324 Swap: 1726940 1584081568532 [root@sflory tmp]# You are probably experiencing a similar usage of memory. I also typically show no swap file usage, although it does creep up to 12MB and I am unsure why it has done so, since there is always 12MB of RAM shown as free in 'top'. Does anyone have more information to add here? -- There is no such thing as obsolete hardware. Merely hardware that other people don't want. (The Second Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED]?subject=unsubscribe https://listman.redhat.com/mailman/listinfo/redhat-list
Re: Advanced Server 2.1 update SRPMS - where?
Dmitry Melekhov wrote: I disagree with you too :-) Original post says that one want to buy subscription to one server and then install updates from sources to several servers. This is impossible according to Service Agreement. You must pay for service for all servers. If you can't read this is not my problem ;-) Have you read the gpl? http://www.fsf.org/licenses/gpl.html * **4.* You may not copy, modify, sublicense, or distribute the Program except as expressly provided under this License. Any attempt otherwise to copy, modify, sublicense or distribute the Program is void, and will automatically terminate your rights under this License. However, parties who have received copies, or rights, from you under this License will not have their licenses terminated so long as such parties remain in full compliance. *6.* Each time you redistribute the Program (or any work based on the Program), the recipient automatically receives a license from the original licensor to copy, distribute or modify the Program subject to these terms and conditions. You may not impose any further restrictions on the recipients' exercise of the rights granted herein. You are not responsible for enforcing compliance by third parties to this License. According to the gpl license if they distribute gpl binaries. They MUST give the recipients' the full rights to the programs. If they don't they lose the right to distribute it. -- There is no such thing as obsolete hardware. Merely hardware that other people don't want. (The Second Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED]?subject=unsubscribe https://listman.redhat.com/mailman/listinfo/redhat-list
Re: Advanced Server 2.1 update SRPMS - where?
Ed Wilts wrote: On Tue, Jan 14, 2003 at 12:20:34PM +0400, Dmitry Melekhov wrote: I disagree with you too :-) Original post says that one want to buy subscription to one server and then install updates from sources to several servers. This is impossible according to Service Agreement. You must pay for service for all servers. If you can't read this is not my problem ;-) If the software update is distributed under the GPL (or similar), then you can legally distribute those updates to your other systems or to the world if you so chose. Distributing outside your organization, however, mandates that you also be able to provide the source, but since your re-distributing SRPMS, that shouldn't be an issue. Yes, but if you have a binary. You have a right to Morally, it's another story. You should by a subscription for all your systems, but you don't have to legally do it as long as the packages you redistribute are GPL'd. I'm missing me the moral point here. The orginal creators of most of the programs in question wanted their program to be used freely. Otherwise why use the gpl? For the most part RH is simply packaging these programs in an rpm. Besides the whole point of AS is the support, and certifications. -- There is no such thing as obsolete hardware. Merely hardware that other people don't want. (The Second Rule of Hardware Acquisition) Sam Flory [EMAIL PROTECTED] -- redhat-list mailing list unsubscribe mailto:[EMAIL PROTECTED]?subject=unsubscribe https://listman.redhat.com/mailman/listinfo/redhat-list