Re: [gentoo-user] OT: Mbox vs maildir
On Mon, 7 Mar 2005 22:17:37 -0500 (EST), Brett I. Holcomb [EMAIL PROTECTED] wrote: [ snips ] To me, if all we asked about on this list was purely gentoo related this would be a dead list as X, bootsplash, sound, network, editors, terminal emulators, etc. all have nothing to do with gentoo per se so we'd all be elsewhere. If you look at it that way there isn't much left to discuss expect the stage 1-3 installs, and the live CDs. As the original poster I asked here for two reasons. First and foremost I'm running Gentoo and want feedback on this issue from others who use Gentoo. Second, there's a lot of good knowledge on this list from people who know Gentoo and other distros as well. I agree 100%, Brett, but we are in the minority. [It seems that] most on the lists are purists. Why, you can't even ask a question about catalyst (which should be a perfectly legitimate Gentoo subject) without being told: get thee to the catalyst list. Sigh. -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] OT: Mbox vs maildir
On Tue, 08 Mar 2005 17:49:23 +1300, Nick Rout [EMAIL PROTECTED] wrote: Look it was me who raised the OT-ness of this thread and if you read back I kind of withdrew my complaint and apologised. There is a line somewhere, and the related problem in this case is how easy it is to get info on the pros and cons of mbox vs maildir by a very simple google search. Sorry to be a bugbear. I'm on a lot of other lists, and seldom do I read on those lists, go and search somewhere else, we're not interested, although I understand that is standard practice on Debian lists. I, for my part, am interested in all things Gentoo and by extension all things Linux, and as Brett stated, there's always the chance that someone on the list has a particular Gentoo insight into the supposedly OT question. Sometimes a Google search on common things such as mbox and maildir can throw out a lot of trivia, perhaps not even including a simple answer as to why one or the other without a tremendous amount of digging, whereas a real live body may have the desired simple answer, and we all (at least the portion of all who doesn't know the answer in advance) can benefit from the shortcut. In conclusion, my OT filter is not as finely tuned as yours. Later, -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] new wireless IP address on my LAN
On Sat, 5 Mar 2005 15:27:00 -0600, Chris Cox [EMAIL PROTECTED] wrote: On Saturday 05 March 2005 11:49 am, Chris Cox wrote: On Saturday 05 March 2005 11:34 am, Ralph Slooten wrote: Yes Chris, as well as MAC address filtering (not bulletproof, but helps). Also you should not advertise your SSID (turn it off). Ok I disabled SSID Broadcast and enabled WEP. I'm not sure how to setup MAC address filtering but I'll look into it. I guess I just never expected anyone to connect to my wireless network besides me. MAC address filtering is also enabled. Does that mean nobody can come in my Wireless network now? You're safer now, but it has been reported that sniffers can decode WEP if they scarf up enough data (it seems like a few weeks is enough). Probably a good idea (tm) not to leave your wireless powered on 24x7. -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] linux-2.6.11 is out
On Fri, 04 Mar 2005 19:57:24 +0800, William Kenworthy [EMAIL PROTECTED] wrote: Exactly - have you have been around gentoo long enough to remember the elegantly simple -u and -U mechanism we used to have to control upgrade/downgrade. Yes, most of us have been around long enough to remember that the elegantly simple mechanism you refer to that seemed to work was inherently broken and left your portage database in an anomalous state. -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] grub-install failure
On Thu, 3 Mar 2005 18:40:56 -0800 (PST), maxim wexler [EMAIL PROTECTED] wrote: Had the same problem. Have you forgotten to mount /proc before chroot:ing? (That peice is hidden a little earlier in the handbook). I mounted it. I followed the instructions in the FAQ re a kernel recompile in a broken system. Only I did a grub-install instead. OK, review time per the handbook. 1) You've booted from an emergency system 2) you've mounted your gentoo root partition 3) you've mounted your gentoo boot partion 4) you've mounted /proc 5) you've done chroot, env-update, source /etc/profile 6) you've determined that /mnt/gentoo/boot/grub does contain the grub stage files and a valid grub.conf 7) grub root (hdx,y) - specs for your boot partition (should get feedback on partition type) setup (hd0) - the first disk (should get feedback about finding all the files and success) quit What are the results of this operation for you? Specific messages please. -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] PDF document 1.4 created with Acrobat Distiller 6.0
On Fri, 25 Feb 2005 13:43:35 -0700, Joseph [EMAIL PROTECTED] wrote: When I try to print it or print the files is being spooled to the printer, the printer light goes ON for a second and out again with-out advancing any paper or anything. When I print that pdf file to ps file and try to open the ps file it appears for a second on a display and the page goes blank. My $.02. You will not be able to solve this problem until Adobe chooses to release an updated version of acroread. They released a beta but then withdrew it. $DEITY knows what and when they'll do. -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Reply-To: header seems broken
On Wed, 23 Feb 2005 17:52:08 +, Mike Williams [EMAIL PROTECTED] wrote: On Wednesday 23 February 2005 17:21, John Myers wrote: Reply to list? That would be a great thing, if I/Thunderbird had that. Which mail client has that? KMail does. If you press 'L' on a list message, it does reply-to-list. KMail also seems to get the reply to address right anyway. (I don't have anything setup to tell it gentoo-user is in this folder, except a list-id filter to put messages here) There are a few things to be learned from this (and several other) thread(s) whcih have been quite animated, to say the least. These truths seem obvious to me, but not so obvious to those who maintain the gentoo lists and others who have jumped into the fray. 1. It's NEVER (yes I'm shouting) a good idea to change existing list behavior without notifying the user base and (apparently) without testing the changes in a limited environment. The result of doing this blindly is almost always a lot of flailing around instead of an orderly process. If it ain't broke, don't fix it also comes to mind. 2. Advance notification of changes (and not just a we're making some changes, you get to figure out what they are) is always a very good idea (tm). 3. It seems totally impossible to root out the attitude Gentoo is for gurus/admins who know every trick in the book, and the rest of you Lusers can just lump it. Unfortunately, not every subscriber to the list is a master of the email headers, the ways of manipulating these headers, and the myriad ways that various email clients handle the headers. Some Lusers just want to read their mail and to be able to reply to the list without a lot of headaches. I don't find suggestions that the Lusers should just go find another list to be very helpful. Sigh. -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: A perfect example
On Wed, 23 Feb 2005 22:28:45 +, Neil Bothwick [EMAIL PROTECTED] wrote: On Wed, 23 Feb 2005 13:58:07 -0600, Boyd Stephen Smith Jr. wrote: Although I've just sent a similar email, I want to jump in here with a me too. Andrea, there are some users out here that realise you were doing the right thing and we applaud you for it. It's unfortunate that this list-at-large doesn't seem to want to fix things that are broken. Whether Reply-To munging is right or wrong, and this argument is likely to be resolved soon after the Vi vs. Emacs debate, changing the way the list works without letting people know only caused confusion and bad feeling. A simple post to the list informing subscribers that Reply-To had been removed, preferably with a link to the page explaining the reasons, would have avoided such an outcome. Yes, amen brother. Exactly what I said in an earlier post. Advanced warning is the polite approach, and it avoids lots of furor. -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: A perfect example
On Thu, 24 Feb 2005 00:37:53 +0100, Holly Bostick [EMAIL PROTECTED] wrote: I accept that header-munging is incorrect behaviour, however it solves some real usage issues, albeit in an imperfect fashion. Maybe instead of arguing about to mung or not to mung, we should be trying to find or create an alternative that solves the issues in a way more satisfying to everyone-- in other words, fix what's actually 'broken' (whether that be fixing one of the solutions, educating the users, or doing something completely new), rather than argue over which imperfect solution is less imperfect than the other. Or maybe, with 100% hindsight, make it a policy to follow normal business practices: 1. Decide the correct approach. 2. Test the effects. 3. Perhaps, gasp, even discuss this with others. 4. Notify the user-base in advance. 5. Provide some suggestions for hardship cases. 6. Stick by your guns. Of course, if you really prefer lots of whining, foist the change on the users with no advance warning. After all, we only want gurus on the list, so everyone will know what happened and what to do. -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] robin.gentoo.org
On Mon, 21 Feb 2005 14:02:48 -0600, Andrew Gaffney [EMAIL PROTECTED] wrote: A. Khattri wrote: Looks like one of the gentoo.org machines is not quite configured correctly - looks like email addresses are not being rewritten properly. The result is that posts from that machine are coming from [EMAIL PROTECTED] instead of [EMAIL PROTECTED] My server-side filtering of Gentoo mailing lists is not working for these posts. Can someone fix this please? Do you actually read the mail to this list? There's been a thread going on for a few hours about the same thing with the subject Mailing list address - who's robin? Whatever the threading, Ajai has it just right: fix the gentoo server(s). I read earlier in the thread(s) opinions stating that the user is the problem. Since not every mailer has the same filtering capability, I can't accept that opinion. Solve the problem at the source (the server), and this thread can die. -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] can't update kde because it needs a file that has been updated
On Sat, 19 Feb 2005 18:51:39 +0100, [EMAIL PROTECTED] [EMAIL PROTECTED] wrote: Hello, emerge -u world wants to update my kdegraphics, but fails because of the following: == START === /bin/sed: can't read /usr/lib/gcc-lib/i686-pc-linux-gnu/3.3.4/libstdc++.la: No s uch file or directory libtool: link: `/usr/lib/gcc-lib/i686-pc-linux-gnu/3.3.4/libstdc++.la' is not a valid libtool archive make[4]: *** [kspell_aspell.la] Error 1 make[4]: Leaving directory `/var/tmp/portage/kdelibs-3.3.2-r2/work/kdelibs-3.3.2 /kspell2/plugins/aspell' make[3]: *** [all-recursive] Error 1 make[3]: Leaving directory `/var/tmp/portage/kdelibs-3.3.2-r2/work/kdelibs-3.3.2 /kspell2/plugins' make[2]: *** [all-recursive] Error 1 make[2]: Leaving directory `/var/tmp/portage/kdelibs-3.3.2-r2/work/kdelibs-3.3.2 /kspell2' make[1]: *** [all-recursive] Error 1 make[1]: Leaving directory `/var/tmp/portage/kdelibs-3.3.2-r2/work/kdelibs-3.3.2 ' make: *** [all] Error 2 !!! ERROR: kde-base/kdelibs-3.3.2-r2 failed. !!! Function kde_src_compile, Line 166, Exitcode 2 !!! died running emake, kde_src_compile:make !!! If you need support, post the topmost build error, NOT this status message. === END === The above message is for kdelibs, but for kdegraphics it' something simular. I have the missing file, but it is in my 3.3.5 and 3.4.2 dirs. How do i convince my gentoo to look for the file in the right location? Other programs merge without problems so far. Henk, Check the archives and forum for information about fixing lib-f###ing-tool, i.e. libtool. This is a frequently occurring problem. -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Xorg and Gnome or KDE on Gentoo
On Sat, 19 Feb 2005 10:28:50 -0800, ME [EMAIL PROTECTED] wrote: Maybe I am wrong on that, but your emerge problem could be due to the libtool config. There is a manual fix to run after updating gcc, from a previous message: Probably upgraded gcc recently? Try fix_libtool_files.sh 3.3.4 as root. If that doesn't work try emerging libtool again, perhaps with emerge -1 libtool. This is sort of a recurring emerge problem that you see many people having in the mailing list. It happened to me with a similar No such file errors in the gcc-lib, the fix_libtool worked, but I am no wiz, somebody can comment on that? Just an aside. I've always been curious what really causes this problem. I've been running gentoo for 4+ years, and I've never encountered this. I wonder if it's because I reun 100% ~x86? -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] OT: Running something complicated from Gnome's browser
On Sun, 13 Feb 2005 10:22:36 -0800, Mark Knecht [EMAIL PROTECTED] wrote: Within Gnome is there a way that I could double click on a file in their browser and then cause the system to start executing a program that asks me questions about what to do? I don't know what you mean by their browser. In Firfox and all the mozilla/netscape derivatives, you can establish Mime relationships that cause a program or script to be invoked when you click (single click is enough) on a file with the registered file type. From the description below, the registered program would need to be a moderately complex script. I have some files that are used in a specialized audio program. Each individual file contain multiple instruments. I'd like the script to: 1) Check the file for available instruments. I know how to do this (sort of) at the command line using something like gigdump filename.gig | grep Instrument. That almost works well enough for this task but it does produce one extra line I'll need to get rid of. In a bash script you can easily pipe the output of your `gigdump filename.gig | grep Instrument` (note the backticks rather than quotes) commands into variables and iterate through these. 2) Pop up a window and tell me about the instrument list. 3) Ask me which on I want to use 4) Ask me a couple more questions that I answer All of this is doable in a script. 5) Take the set of answers and send it over the network using some command like echo command 1 | nc localhost echo command 2 | nc localhost Also not a problem in a script. I know nothing of doing stuff like this but it would be very helpful. The windows program that uses these files of course has a fancy file browser that allows you to do this with a couple of click drag operations. The Linux program has nothing so I'm doing this by hand and it's very tedious. Yes, Linux does not have a builtin script to do this, but you can roll your own. The question is: are you ready to learn bash or perl or python or php or ??? and to do a little research to build your knowledge of the browser interface and the use of these tools? A little knowledge can be a dangerous thing. -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] OT: Running something complicated from Gnome's browser
On Sun, 13 Feb 2005 15:48:02 -0600, Boyd Stephen Smith Jr. [EMAIL PROTECTED] wrote: On Sunday 13 February 2005 03:31 pm, Mark Knecht [EMAIL PROTECTED] wrote: 2) Pop up a window and tell me about the instrument list While bash has no support for this, you may be able to use GTK binding for some scipting language Perl/Python to do this. Your first stop for this step is the GTK documentation/tutorial. Gnome allows me to run my bash script in a terminal so now the terminal pops up, runs gigdump, greps out the right lines and saves them to a temp file. I then cat that file to the same terminal and see the results. Good so far. 3) Ask me which on I want to use 4) Ask me a couple more questions that I answer If you want a simple pseudo-gui textmode system to let you display simple menus and make choices in bash, google for information about 'dialog'. You can create a simple structure to do what you want. It looks very much like the panels in the Slackware installer, if you've ever seen that. -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] gmail question
On Sat, 12 Feb 2005 06:10:11 -0800, Mark Knecht [EMAIL PROTECTED] wrote: On Sat, 12 Feb 2005 14:56:21 +0100, Holly Bostick [EMAIL PROTECTED] wrote: So what does I don`t recive my mail-message back mean? Do you mean that 1) gmail does not save a copy of the mail you sent to the list in gmail's Sent Mail folder? or 2) You get the posts from the list, but your mail is not among them (you receive replies to what you posted, but you do not see the original post you made, which should be at the head of the thread)? Hi Holly, I use Gmail and I've experienced what Makurin sees. What Gmail does is puts everything ito a single thread. That's not unique. What is a bit different is that the post I make to the list is the copy it seems to keep in the thread. You do not end up with the copy you sent and the copy you received back. You see only one. Makurin is receiving his posts but GMail is hiding them from him. He can prove this by sending a message to this list but then going to the sent folder and deleting the sent copy to trash and emptying the trash. I find when I do this that when I receive my copy back from the list a few minutes later I do see it. It's just a strange aspect of GMail. I'm afraid I don't understand this at all. I've been using gmail for a long time, and I haven't ever had any problem with replying to the list, except that gmail almost always generates a duplicate reply - to the list and then to the sender, although somethimes this fails to happen. I any case, I see my message to the list at the right place in the chain, and thus I could care less which copy of my message has been saved. -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: [gentoo-alpha] gentoo KDE problem
Avanti System Machine Check Through Vector 0660 logout frame address 0x6048 code 0x10205 IPRs: EXC_ADD:00087D3A ICCSR: 46F80004 HIER: 0080 HIRR: 0040 MM_CSR: 5801 DC_STAT:0007 DC_ADDR:0007 BIU_STAT:0041 BIU_ADD:0003881001F8 FILL_SY: FILL_ADD:00011270 VA: 61D0 EXC_SUM:942E BC_TAG: 000A8057 EDSR (Com ...etc -- G a b r i e l M . B e d d i n g f i e l d -- gentoo-user@gentoo.org mailing list -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Hello
On Sat, 12 Feb 2005 21:25:41 -0500 (EST), Brett I. Holcomb [EMAIL PROTECTED] wrote: Haven't heard from you for a while. I hope all is well with you. All is well here, but I don't think you meant to send this to the list, or ??? -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] FireFox Search Extentions don't work
On Sat, 12 Feb 2005 22:55:22 -0500 (EST), Brett I. Holcomb [EMAIL PROTECTED] wrote: I have firefox 1.0 installed. When I go to the search bar and tell it to install other extensions it says it did it succesfully but they never show up. Any ideas. I've been using Firefox a long time, but I'm not sure what the search bar is? OTOH, I've never had much luck installing extensions from a normal user. If I'm going to install extensions, I startx as root, then do my extension work, then log out. Any other way, file permissions get in the way. -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] FireFox Search Extentions don't work
On Sun, 13 Feb 2005 01:21:16 -0500, Petter Häggholm [EMAIL PROTECTED] wrote: Collins Richey wrote: On Sat, 12 Feb 2005 22:55:22 -0500 (EST), Brett I. Holcomb [EMAIL PROTECTED] wrote: I have firefox 1.0 installed. When I go to the search bar and tell it to install other extensions it says it did it succesfully but they never show up. Any ideas. I've been using Firefox a long time, but I'm not sure what the search bar is? OTOH, I've never had much luck installing extensions from a normal user. If I'm going to install extensions, I startx as root, then do my extension work, then log out. Any other way, file permissions get in the way. Installing regular extensions should work fine as a user (does for me) since they're stored in users' home directories; Sure, but I prefer to have extensions installed for all users. search plugins in particular sadly won't work off the shelf, so to speak, since only root has access to /usr/lib/MozillaFirefox/searchplugins, which must be modified. Personally, I just change the permissions of the directory so I can install search plugins as a user; A good idea. alternatively, you could fire up Firefox as root (no need to startx as root!). Yeah, come to think of it, that's what I've done. At least I learned what the search bar is g. -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Gentoo Linux and Windows XP on the same hard disk on the same computer
On Mon, 7 Feb 2005 09:13:20 +0100, Mariusz Pkala [EMAIL PROTECTED] wrote: On 2005-02-06 20:19:58 + (Sun, Feb), Martin Scharrer wrote: On Sunday 06 February 2005 17:53, Collins Richey wrote: [...] WinXP in its great wisdom (NOT) puts the Windows Swap file right in the middle of the partition, and that file is marked as unmovable, so with all the tools I'm aware of, you will only be able to get half of the space (starting right after the swap file) free for use as new partitions for Linux. Because I don't had any problems to resize a 40GB Windows XP Home partition down to 8GB on my Laptop. I managed to shrink 20GB into 7GB for XP on my laptop and I remember no troubles with that. I did it using only GNUish tools: sys-fs/ntfsprogs , no PartitionMagic. As for the swapfile - you can always temporarily disable virtual memory in WinXP, reboot, test that WinXP is alive, repartition, reboot to WinXp, enable virtual memory and live happily ever after. YMMV, I tried that, but WinXP still left an unmoveable file in the middle of the disk even after turning off virtual memory. I'm not a windows expert, so I don't have a clue what the file was. So, I was left with the convenience of getting half the space freed quickly and the inconvenience of some unused space in the WinXP partition. Good luck. -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Gentoo Linux and Windows XP on the same hard disk on the same computer
On Mon, 7 Feb 2005 18:15:05 +, Neil Bothwick [EMAIL PROTECTED] wrote: On Mon, 7 Feb 2005 10:44:58 -0700, Collins Richey wrote: YMMV, I tried that, but WinXP still left an unmoveable file in the middle of the disk even after turning off virtual memory. I'm not a windows expert, so I don't have a clue what the file was. So, I was left with the convenience of getting half the space freed quickly and the inconvenience of some unused space in the WinXP partition. I dealt with that by booting back into XP and running the defragmenter again, which put the swap file back in the middle of the new partition. then I could halve its size again. Now that's the answer. I'll keep this little gem for when I have to do that again! Fortunately I've got an extra disk ready to install, so I'll have a little more room to play with the disk that has WinXP on it. Learning w-a-y more than I ever wanted to know about WinXP grin. -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] migrating installation from one disk to another
I knew there was a reason I put that note of caution there! Obviously I didn't test it out... yes, you are correct that would indeed delete everything from the old root partition. It should be this instead: # cd /mnt/usr On Saturday 05 February 2005 12:25 pm, Nick Rout wrote: On Sat, 2005-02-05 at 12:10 -0800, Adam Collins wrote: 8) Delete the data from the original /usr directory: # mount --bind / /mnt # cd /mnt # rm -rf * -- careful with this! # cd / # umount /mnt From here, you can run df -h to show your filesystem usage. That's it. whoop pull up, isn't this going to delete the whole root dir, therefore the whole filesystem? you bind mount / to /mnt then rm'ing whats in /mnt -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Gentoo Linux and Windows XP on the same hard disk on the same computer
, hda3, and hda5 as above (the first partition carved out of the extended partition is 5). Now decide how you want to carve up the remaining space. Gentoo recommends a /boot partition and a / (root) partition as minimum. Most people like to have a /home partition as well. I don't use /boot, so you won't see it in my example. b. If you want /boot (per the Gentoo recommendations), command 'n', then select logical, then hit enter to accept the starting cylinder location, then enter '+20M' (or your choice). You now have hda1, hda2, hda5, hda6 (for /boot). You don't need to reset the partiion type any more, since type '83' is the default c. Next create the / (root) partition. Command 'n', then select logical, then hit enter to accept the starting cylinder location, then enter '+10240M' (or your choice) to create a 10G partition. An extremely large Gentoo install my run to 10G. I've used 13G for my / partition, and I have currently 9G used. You can get by with only 3-5G, but I would recommend starting big. Now you have hda1, hda2, hda3, hda5, hda6, and hda7. d. Next create a /home partion, if you want it. I use 4G for mine, but you may want to keep scads of pictures and/or MP3 files, so make it as large as you want. Next command 'n', then select logical, then hit enter to accept the starting cylinder location, then enter '+10240M' (or pick a size or hit enter if you want to use all the remaining space). Now you have hda1, hda2, hda3. hda5, hda6, hda7, and hda8. If you didn't use all the remaining space for hda8, you can continue to create additional partitions. To see what is left, command 'n', then select logical, then hit enter twice. When you issue 'p', you will see that hda9 has the remining space on your disk. Check your work. If you don't like the results, you can delete what you have done working backwards from the list of partitions. You can remove anything except hda1!!! Then work through the procedure again until you get it right. e. When you have an acceptible layout, issue 'p' and record your work so that you have a written record of which partion is which. Then command 'w' to write your new partition table to the disk. 5. Note that you have created (if you followed my example) an unformatted FAT32 partition, so when you next boot WinXP, format that disk as FAT32 (don't accept the NTFS default!!!). 6. Now your linux partitions are ready for use. Just follow the handbook. Your partitions of concern are as follows (or whatever you chose to do). hda1 - WinXP hda2 - extra FAT32 hda5 - swap hda6 - /boot hda7 - / hda8 - /home HTH, -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: firefox random lockups
On Sat, 5 Feb 2005 14:01:30 +0100, Nuno Alexandre [EMAIL PROTECTED] wrote: On ons, 02 feb 2005, Collins Richey wrote: On Wed, 2 Feb 2005 15:39:26 -0500, Thomas Kirchner [EMAIL PROTECTED] wrote: * On Feb 2 5:07, comsatcat ([EMAIL PROTECTED]) thusly jotted: As of two days ago, firefox started to randomly lockup. Things will work fine for awhile, then when I go to a random site, it will just lock solid. The only way I've been able to get it to not lock up is to rm -rf /tmp/*. Does anyone know which files may be causing the lockup or does anyone have a solution for this? I can't say I've had it lock up, but it's definitely crashed a lot recently. It always occurs when trying to open a new site, but will not reoccur when I revisit the site later. I can't say I've noticed anything strange in /tmp though... Tom p.s. to be fair, a lot means more than usual. Nothing really crashes often. The only crash activity I've seen is: randomly when selecting the back key while viewing a PDF document. Hi, are you sure the problem is Firefox ? I doubt! and you are using Xorg, right ? Well look here. http://bugs.gentoo.org/show_bug.cgi?id=80642 Thanks for the info. Maybe this will help others, but this is not my problem. My problem is 1. Not a freeze or hang condition; browser always dies, but nothing else is affected. 2. Seldom reproducible, very random. 3. Always associated with use of acroread, and usually after printing from acroread. -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: firefox random lockups
On Sat, 05 Feb 2005 18:08:54 +0100, Holly Bostick [EMAIL PROTECTED] wrote: Collins Richey wrote: Thanks for the info. Maybe this will help others, but this is not my problem. My problem is snip 3. Always associated with use of acroread, and usually after printing from acroread. Well, that's a horse of a different color. Not to put too fine a point on it, the acroread browser extension coughsucks/cough, and has a strong tendency to crash any browser (Win or Lin). [ other helpful data snipped ] Not too surprising since the vendor refuses to do much for unix/linux users. Fortunately, I only need PDF documents once in a blue moon. -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] migrating installation from one disk to another
I'm not sure what you mean by system disk, but it is fairly easy to move /usr to a new partition, if that's what you mean. If you want to move your entire / (root directory) to a new partition (or partitions), it's a bit more complicated but is fairly similar to the steps below. For my example, /dev/hda6 is the new partition and I'm going to use reiserfs. This won't mess anything up, if done correctly. 0) Make the new partition with fdisk or cfdisk. 1) REBOOT! -- Just do it. 2) Make a filesystem on the new partition with mkreiserfs or whatever: # mkreiserfs /dev/hda6 3) Move the /usr data to the new partition: # mkdir /mnt/usr # mount /dev/hda6 /mnt/usr # cp -a /usr /mnt/usr 4) Cleanup: # umount /mnt/usr # rmdir /mnt/usr 5) Modify /etc/fstab. Add a line like this: /dev/hda6 /usr reiserfs noatime 0 1 6) Reboot. 7) Verify that the new partition is actually being used: # mount | grep /usr should show something like: /dev/hda6 on /usr type reiserfs (rw,noatime) If mount | grep /usr doesn't show anything, don't continue to the next step! 8) Delete the data from the original /usr directory: # mount --bind / /mnt # cd /mnt # rm -rf * -- careful with this! # cd / # umount /mnt From here, you can run df -h to show your filesystem usage. That's it. -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: firefox random lockups
On Wed, 2 Feb 2005 15:39:26 -0500, Thomas Kirchner [EMAIL PROTECTED] wrote: * On Feb 2 5:07, comsatcat ([EMAIL PROTECTED]) thusly jotted: As of two days ago, firefox started to randomly lockup. Things will work fine for awhile, then when I go to a random site, it will just lock solid. The only way I've been able to get it to not lock up is to rm -rf /tmp/*. Does anyone know which files may be causing the lockup or does anyone have a solution for this? I can't say I've had it lock up, but it's definitely crashed a lot recently. It always occurs when trying to open a new site, but will not reoccur when I revisit the site later. I can't say I've noticed anything strange in /tmp though... Tom p.s. to be fair, a lot means more than usual. Nothing really crashes often. The only crash activity I've seen is: randomly when selecting the back key while viewing a PDF document. -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] sudo xcdroast
On Sun, 30 Jan 2005 16:43:27 +, Dave S [EMAIL PROTECTED] wrote: Any suggestions for a lightweight cd burner, 3kb is a bit more than I want ... ;-) Sorry, I always use cdrecord as root and never (at least since I fixed my /dev/hdc to deactivate DMA) have any problems burning. I use a simple script that just inserts the name I want to burn. Also, I don't burn anything but iso's, so I can't help you with music cd's, etc. -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] OK, who do I have to thank? (WARNING TO THOSE ABOUT TO EMERGE)
On Thu, 27 Jan 2005 11:41:13 -0500, Dave Nebinger [EMAIL PROTECTED] wrote: Who do I have to thank for the recent x11-base/xorg-x11-6.8.1.902-r1 ebuild and it's nested spawning of bash processes grinding my systems to a total halt? Off hand, given that the ebuild is hard masked - # Don't use this, work in progress KEYWORDS=-* This is the first xorg release, in fact first ebuild ever, that was so poorly constructed as to start spawning shell processes to the point of total system lockup. I've been running the -* ~x86 releases of xorg since I did the setup of the system and this has never been an issue before. Yes, I know the doco says that ~x86 could introduce a package that may break the system, but to doco never says that ~x86 will lock up your system just for trying to emerge it... Sorry, this is just whining. Which part of the consequences of running non-stable (and hard masked to boot) code on your inaccessible production server did you fail to understand? sarcasmYou probably like Russian roulette, too/sarcasm. Gentoo developers do a fantastic job, and this includes releasing experimental ebuilds that are hard masked for the adventuresome to try and to report the results. Get a life. -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Re: CDrecord on kernels 2.6.8
On Thu, 27 Jan 2005 13:33:56 +1100, gmane [EMAIL PROTECTED] wrote: it will. burn as user, do not use ide-scsi and cdrecord MUST NOT be suid-root. Burning will be fine, when you do not forget to make a chmod -s cdrecord I've been using /usr/bin/cdrecord suid root (with K3B) with 2.6.9-gentoo-rx and it's been working fine. Are there security/stability problems with doing this?? I forget where the cutoff point is, but when you move to a newer kernel this will stop working. -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] emerge vanished, what to do now
On Tue, 25 Jan 2005 16:44:21 +, Jim Hatfield [EMAIL PROTECTED] wrote: I don't know what I did, but I no longer have an emerge. whereis emerge gives nothing at all. Is there a tarball I can download to get it back and then do a normal update, if not how do I get out of this one? Follow the instructions at /usr/portage/sys-apps/portage/files/REAME.RESCUE. This will get yoyu a working emerge. Then reemerge python and portage. HTH, -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] emerge vanished, what to do now
On Tue, 25 Jan 2005 11:50:03 -0700, Collins Richey [EMAIL PROTECTED] wrote: On Tue, 25 Jan 2005 16:44:21 +, Jim Hatfield [EMAIL PROTECTED] wrote: I don't know what I did, but I no longer have an emerge. whereis emerge gives nothing at all. Is there a tarball I can download to get it back and then do a normal update, if not how do I get out of this one? Follow the instructions at /usr/portage/sys-apps/portage/files/REAME.RESCUE. This will get yoyu a working emerge. Then reemerge python and portage. oops, thats README.RESCUE. -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Install error at env-update
On Wed, 26 Jan 2005 03:54:36 +, Aaron Elbaz [EMAIL PROTECTED] wrote: This problem makes absolutely no sense to me... After mounting the proc and chrooting to /mnt/gentoo, I type in 'env-update' as per the docs. After doing that, it says: unable to handle kernel null pointer deference at virtual address 0011. It prints out maybe 15 lines of output including the stack and call trace, then some assembly machine code (I think thats what it is-letters and numbers in clusers of 2). Any ideas on what this means? Its happened three straight times and I can't get past this point. I did manage to a few times before, but both times I never really did anything important. Stage 1 2 or 3? If Stage 3, did you by chance download a Stage 3 tarball that is not correct for your CPU? -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Secret bugs?
On Mon, 24 Jan 2005 12:05:50 +, Ian Hastie [EMAIL PROTECTED] wrote: On Monday 24 January 2005 10:31, Xavier-Francois Roblot wrote: So it is for a security issue that we cannot access the bug. Supposedly. On the other hand, it is not difficult to have a look at the (very short) corresponding patch: /usr/portage/mail-client/evolution/files/evolution-2-CAN-2005-0102.patch from which one can deduce what the bug was ;o) As you imply it doesn't work very well does it. Looks like a buffer overflow of some kind. The patch looks kind of unpleasantly hackish to me. The only advantage I can think of is that it doesn't need to know the limitations of the hardware or compiler. It just detects the condition that allows the overflow to take place. Whatever. The practice of secreting fixes is still complete horse feathers, and it reeks to high heaven. -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Secret bugs?
On Mon, 24 Jan 2005 14:33:36 +, Ciaran McCreesh [EMAIL PROTECTED] wrote: On Mon, 24 Jan 2005 07:14:21 -0700 Collins Richey [EMAIL PROTECTED] wrote: | Whatever. The practice of secreting fixes is still complete horse | feathers, and it reeks to high heaven. Indeed it does. Take it up with VendorSec. We don't really have much choice but to play along (or we could chose to just not know about any kind of security fix until several months after RedHat and Debian do, but that's not really a very nice option). OK. Perhaps someone could publish this standard exception prominently on the gentoo site so as to avoid wasted time in threads like this one. -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] xfce4-4.2 and xfce-base-4.0.6
On Sun, 23 Jan 2005 23:16:19 + (UTC), heide [EMAIL PROTECTED] wrote: Hi all, unable to solve a problem with emerging xfce4, please forgive me for consulting this kind list. Last week an emerge -uvDa --newuse world upgraded xfce4 from xfce-4.0.6 to 4.2 and it worked fine. Now, another emerge -uvDa --newuse world gives me: [blocks B ] xfce-base/xfce4-base (from pkg xfce-base/xfce4-4.2.0) snip After this removal the emerge -uvDa --newuse world wanted to install xfce4-base-4.0.6 again and the block was the same as the original one above. Should I file a bug or is this behavior just because of me trying to do some silly things? My suggestion is: this is a bug. I encountered the same problem. I was too lazy to report, so I just did emerge -C on everything in xfce-base and xfce-extra. Then I restarted the emerge. The developers have screwed up something, so I'm sure they would appreciate a hint. -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Custom Bootable CD
On Sun, 23 Jan 2005 20:51:02 -0500, Stefhen Hovland [EMAIL PROTECTED] wrote: On Sun, 23 Jan 2005 15:49:08 -0500, Nick Smith [EMAIL PROTECTED] wrote: It's also very helpful if solving most of windblowz troubles ... dd if=/dev/zero of=/dev/hda is my favorite command. zero's out your hard drive. Yeah, but he's presuming that's where your Windows C:\ drive is, so goodbye to Windows problems. -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] laptop's getting hot, any hints?
On Mon, 24 Jan 2005 12:51:39 +0800, mel [EMAIL PROTECTED] wrote: Hi all, I'm running Gentoo on Compaq Presario 1500 for almost a year now, and everything is fine and dandy except for one thing: the CPU gets hot quickly whenever I emerge large packages (xorg, mysql, kernels). Currently I don't have any control apps (speedfreq, cpufreq), and I resort to distcc (which keeps the CPU temp somewhere between 65 deg C to 68 deg C - without it it goes as high as 74 deg C). Another thing that I do is Ctrl-S to suspend the compile, let the CPU cools down, and Ctrl-Q to continue. The fans are all working fine. Question: What other things I can do to cool the CPU while emerging? Will using cpufreq/speedfreq helps? My first thought would be a question: do you have a desktop computer that could do the compile work? If that's the case, you could setup to build binary packages on the desktop compiled for the laptop and then install those packages on the laptop. This, of course, would completely circumvent the cooling problem. $.02. -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] laptop's getting hot, any hints?
On Mon, 24 Jan 2005 13:04:43 +0800, mel [EMAIL PROTECTED] wrote: Collins Richey wrote: My first thought would be a question: do you have a desktop computer that could do the compile work? Yeah I do except that the the desktop computer is an Athlon-XP, while my notebook is Intel-based. Another thing is that my notebook has different use flags compared to the desktop. So, having my desktop compile the apps for me is out of the question since I like to have my notebook slim :) I'm not an expert on these matters; perhaps a developer or guru will step in. Nevertheless, I believe with a little work you could setup to build packages with the USE flags you prefer for the laptop and maintain these separate from the desktop configuration. You might search on forums for more information. As a final note, if your laptop does not have adequate cooling and running the fans constantly does not produce the desired results, you may well have to consider the approach of building packages on the desktop machine. -- Collins -- gentoo-user@gentoo.org mailing list
[gentoo-user] sxcreensaver and glxgears again!
Well, I'm stumped. I've been having trouble with xscreensaver and glxgears for months. They work for a while, and then they crap out the X server. Yesterday I had a repeat of this behavior after an upgrade that installed the following. 4993-glib-2.6.1.log 4993-shared-mime-info-0.14-r2.log 4995-gcc-config-1.3.9.log 4995-gtk+-2.6.1-r1.log 4996-gcc-3.4.3.20050110.log I tried recompiling gtk+, xorg-x11, pango, glib, etc., anything pointed to by ldd, but no improvement. Finally, I entered 'glx-update xorg-x11' and the problem was gone. wtf! Why would I ever need to run glx-update more than once, most especially since there's only one choice these days? -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Ulimate Gentoo Box
On Sat, 22 Jan 2005 19:34:16 -0600, Andrew Gaffney [EMAIL PROTECTED] wrote: Nick Smith wrote: i was reading up on the wiki and came across the 'Ultimate Gentoo Box' i want to give it a shot i just had a couple questions. 1. i think i want to give ~x86 a try, i find myself adding more and more keywords to my package.keywords file anyway. how to i go about getting everything ~x86? and if so, is it known to break things all the time? or are people running pure ~x86 with no problems? also how to i tell the system to replace all the packages with ~x86 ones? and finally what happens to the things that are in my package.keywords file now? a) ACCEPT_KEYWORDS=~x86 in /etc/make.conf b) I've been running ~x86 for 2 years with minimal breakage c) There will never be *no* problems d) emerge -uD world e) if you do a), the entries in package.keywords will just be redundant Ditto. I've been running ~x86 for a long time with fewer problems than with x86. Most of the problems have been with packages that failed to compile when first marked ~x86, but soon were fixed. My system might be considered bleeding edge, but there's no blood in sight. I don't believe you need to recompile anything (see also below); just let 'emerge -u world' upgrade what is needed. The one thing you should not do (IMO) is to run ~x86 for a while and then revert to x86 without recompiling everything. This has been reported to create problems. 2. the CFLAGS on their looked really interesting and i think i want to recompile my entire system with those flags, how do i go about doing that? will an 'emerge -euD world' work? or is there something else i have to do to get it to recompile everything? 'emerge -e world' should do the trick. 3. could someone please explain to me in plain english ( i have read about them online ) what the main difference is between -O3 and -O2? i know -O3 is more optimized, but which is better to make the system perform better? the wiki site im refering to states he uses -O2, i am currently using -O3 when i did my stage one install. what is going to make my system perform better? -O3 is more optimized than -O2, but those optimizations usually make the binaries larger which causes other slowdowns. The ideal system-wide optimization level is -O2 or -Os (-O2 with additional flags for small binaries). im running a p4 2.4Ghz laptop with 1gig ram ati mobility video, and if this works i have several other systems i want to do the same thing with. (except the server) -O2 and -O3 are quite similar, but -O3 inlines certain functions and thus the executables are somewhat larger. My personal opinion is switch to -O2, but I don't believe you need to recompile anything unless you want to, in which case fire off an 'emerge -e world'. IMO, avoid -Os. There are some packages (can't remember which) that don't work well with -Os, and the ebuilds usually filter out -Os in favor of -O2. -Os executables will be somewhat smaller, but your system has enough oomph to handle the -O2 optimized executables. Whatever you do, I doubt that you'll notice much difference between -O2 and -O3. -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] No module named fcntl
On Wed, 19 Jan 2005 23:56:38 +0100 (CET), Jürgen Schinker [EMAIL PROTECTED] wrote: On Thu, January 20, 2005 0:39, Axel Schmalowsky said: Jürgen Schinker wrote: !!! Failed to complete python imports. There are internal modules for !!! python and failure here indicates that you have a problem with python !!! itself and thus portage is no able to continue processing. !!! You might consider starting python with verbose flags to see what has !!! gone wrong. Here is the information we got for this exception: No module named fcntl got this error can somebody help thanks Jürgen -- gentoo-user@gentoo.org mailing list try to re-emerge python because it seems that your python-lib is incomplete. Axel how can i do this with emerge not working? Go to /usr/portage/sys-apps/portage/files/README.RESCUE. Follow the instructions to get a working emerge, then remerge python and portage. -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] No module named fcntl
On Thu, 20 Jan 2005 14:25:55 +, Axel Schmalowsky
[EMAIL PROTECTED] wrote:
Jürgen Schinker wrote:
You can download python from http://www.python.org Unzip it, run
./configure with the parameters portage is using (you'll find this in file
/usr/bin/ebuild.sh in function econf) and then run make. The parameters
portags is using for make install can you find as well in file
/usr/bin/einstall in function einstall. After that, your python should be
complete thus you should be able to emerge normally. what parameters should
i use? econf() { if [ -z ${ECONF_SOURCE} ]; then ECONF_SOURCE=. fi if [
-x ${ECONF_SOURCE}/configure ]; then if hasq autoconfig $FEATURES !
hasq autoconfig $RESTRICT; then if [ -e /usr/share/gnuconfig/ -a -x
/bin/basename ]; then local x for x in $(find ${S} -type f -name
config.guess -o -name config.sub) ; do einfo econf: updating $x with
/usr/share/gnuconfig/$(/bin/basename ${x}) cp
/usr/share/gnuconfig/$(/bin/basename ${x}) ${x} done fi fi if [ ! -z
${CBUILD} ]; then EXTRA_ECONF=--build=${CBUILD} ${EXTRA_ECONF} fi if [ !
-z ${CTARGET} ]; then EXTRA_ECONF=--target=${CTARGET} ${EXTRA_ECONF} fi
# if the profile defines a location to install libs to aside from default,
pass it on. # if the ebuild passes in --libdir, they're responsible for the
conf_libdir fun. if [ ! -z ${CONF_LIBDIR} ] [ ${*/--libdir} == $*
]; then if [ ${*/--prefix} == $* ]; then CONF_PREFIX=/usr else local
args=$(echo $*) local -a pref=($(echo ${args/*--prefix[= ]}))
CONF_PREFIX=${pref} fi export CONF_PREFIX
EXTRA_ECONF=--libdir=/${CONF_PREFIX}/${CONF_LIBDIR} ${EXTRA_ECONF} fi echo
${ECONF_SOURCE}/configure \ --prefix=/usr \ --host=${CHOST} \
--mandir=/usr/share/man \ --infodir=/usr/share/info \ --datadir=/usr/share \
--sysconfdir=/etc \ --localstatedir=/var/lib \ ${EXTRA_ECONF} \ $@ The
parameters above are the one portage is using.
Additionally, you can choose optional parameters which you'll get by
running ./configure --help
${ECONF_SOURCE}/configure \ --prefix=/usr \ --host=${CHOST} \
--mandir=/usr/share/man \ --infodir=/usr/share/info \ --
gentoo-user@gentoo.org mailing list
As I mentioned earlier, gentoo already has recovery procedures builtin
(README.RESCUE), so you don't need to download and built python from
external sources.
If you want to do it the hard way, fine and dandy, but it's not necessary.
--
Collins
--
gentoo-user@gentoo.org mailing list
Re: [gentoo-user] which packages provide virtual xyz
On Mon, 17 Jan 2005 21:12:32 +0100, Maik Musall [EMAIL PROTECTED] wrote: Hello, how do I find out which packages provide a certain virtual? Say, some dependency would install mail-mta/ssmtp, but I want another mta. ssmtp is recorded in /usr/portage/profiles/base/virtuals as default to virtuals/mta, but how do I find out what other packages would provide the same virtual? The brute force method would be 'grep -R PROVIDE= /usr/portage/* (optional pipe to less or another grep or output to a file. -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] which cvs frontend to use?
On Fri, 14 Jan 2005 18:29:14 -0500, Dave Nebinger [EMAIL PROTECTED] wrote: As far as gotchas, when you install a source version it will overwrite files gentoo has put in place. The next time a new version is released in gentoo and you emerge the world, your custom version will be blown to bits. Not necessarily true. Most packages have the ./config options set to install to /usr/local, or you can use /opt. In either case, there will be no over lap with gentoo packages. -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Binary Package Repository?
On Wed, 12 Jan 2005 18:21:22 -0500, Nicholas Pappas [EMAIL PROTECTED] wrote: Hello all. I'm having a crisis of conscious at the moment. I've been using Gentoo for a while and I really like it, but the simple truth is I don't need the granularity that compiling the source for each update provides. That, and I don't fool myself into pretending I know enough to tune my USE (and per package USE) variable to make the source version any better then your average prebuilt P4 package. Are there mirrors that offer binary versions for the P4 for each package? I might have a few features built into apps that I might not need, or maybe have to juggle a few source distributions to make sure I get what I need, but a mostly pre-built repository would just make my updates so much better. :) Just pick [ any | your favorite binary distro ]. After [ a few | many ] binary package upgrades you'll reach the desired (g) unstable stage which is why the binary distros typically issue a new release to keep everything working. Otherwise, just run your compiles overnight or when you're having fun browsing the net or , and worry less about electricity. -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] Upgrading perl
On Wed, 12 Jan 2005 17:03:01 -0800, Grant [EMAIL PROTECTED] wrote: I'm currently using perl 5.8.4-r1 on my server, and I'm having an apache2 seg faulting issue that is supposed to be related to perl signals. I'd like to upgrade to the latest perl to try and resolve it. Is there anything I should be aware of, or is it just emerge and you're done? Check the recent archives or forum. There is a procedure to rebuild other perl modules that are dependant on the installed version of perl. rant I can't imagine why the 'emerge perl' process doesn't do that automatically. /rant -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] how can i do a _real_ check of the rootfs?
On Thu, 13 Jan 2005 04:36:18 +0100, Sven Köhler [EMAIL PROTECTED] wrote: hi, it seems that the rootfs (a reiserfs) is damanged. The problem is, that i cannot remount the rootfs read-only after the system has been booted completly. - how can i tell gentoo to do a real complete check of the rootfs at startup (where the rootfs should be read-only mounted) - how can i do that via ssh on a running system - which paramters can i pass when booting gentoo, so that it boots into maintainance mode where the rootfs is still mounted read-only? Answer to both questions: you can't do this booted from your gentoo system (standard *nix problem - checking a fs requires read only mode, but root fs is r/w when booted.) What you need to do is get a LiveCD (almost any distro, Knoppix, etc). Boot from the CD, mount your gentoo root fs as read only, and run your filessytem check and/or recovery utilities. HTH, -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] before begining
On Wed, 12 Jan 2005 22:09:17 -0800, John Coder [EMAIL PROTECTED] wrote: I have been reading most of the emails sent to this list for the past couple of days because I plan to start my first attempt at installing my gentoo system. Is there any addvice from those who have completed this recently? It will be an amd 800 and I have been running suse for the past year with great success, and I have tried Feddora core 1 and previous versions of redhat. 1. Read the install doco thoroughly. 2. Don't think you can skip steps just because you have experience with other distros. Please don't be offended. All too many try to get clever with the installation steps. 3. With an AMD 800, compile times are going to be lengthy. You might want a Stage 3 install and/or the precompiled binaries. After getting a working system, then you can run the lengthy compiles while enjoying your system. 4. Welcome aboard. My $.02 -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] [OT] Filesystem Benchmarks
On Fri, 7 Jan 2005 02:33:17 +0100, Marc Ballarin [EMAIL PROTECTED] wrote: In my experience the need for recovery tools is much smaller with ReiserFS. Internal journal replay seems to be very reliable compared to ext3. This might depend on specific hardware, though. Regards BTW: On broken hardware (power regulation), three reiserfs filesystems - two of them heavily used - survived ~100 hard crashes without any unexpected loss of data or any inconsistencies. In every case journal replay worked flawlessly, reiserfschk was never needed. My experiences with ext3 - and especially XFS - are much worse. (To be honest, XFS suffered from an actual bug, not a design issue. Some pages *never* got flushed to disk before umount. This has been fixed a while ago.) That's encouraging to hear. My only 3 experiences with reiserfs (1x late 2.4, 2x on 2.6, never again in this lifetime) led to total fs corruption after a powerfail. The recovery tools failed as well. Maybe there's hope with reiserfs 4. OTOH, I've used ext3 almost since inception and never experienced this sort of corruption across maybe a dozen powerfails. I had one instance where a directory (/boot, IIRC) became slightly snafu'd, but copying the files out and recreating the directory cured the problem. I've been running xfs now for 6 months or more, and I've encountered no problems there. Obviously, ymmv. -- Collins -- gentoo-user@gentoo.org mailing list
Re: [gentoo-user] vmware
On Wed, 03 Mar 2004 09:42:53 -0500 Larry Schuler [EMAIL PROTECTED] wrote: You missed one other option, do without. That is what I am doing. Exactly. I would love to have one of those new Ferrari's. But, oh my, they're too expensive, and my poor little self can't afford one on my piddly salary. So, should I just go steal one Bruce E. Harris wrote: You missed one other option, do without. That is what I am doing. On Tuesday 02 March 2004 07:08 pm, Bruce E. Harris wrote: I know that 300$ is nothing in your world with people being payed thatmuch in a day (if say a doctor worked in your country he might get closeto 300$ a day), but few people here have a salary in that range overhere. So we have several options, 1. Buy the exorbitantly priced software. 2. Buy a pirated cd of the software (.5$) 3. Download a cracked vesion of the sofware. Grendel You guys are obviously too far down the intellectual ladder. If you have high enough IQ, you can see right through all this personal integrity and moral honesty crap; just take whatever you want. What's the old saying among gang bangers - What you see is what you get! Doing without is only for the poor schmucks who play by the rules. Makes me want to head right down to Best Buy and rip off a laptop or (more likely) to puke!!! -- Collins Richey - Denver area gentoo testing 2.6.3-rc2 nptl udev -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] unusual installation ?
On Thu, 26 Feb 2004 12:25:46 +0200 raptor [EMAIL PROTECTED] wrote: u mean over my existing distribution... i.e. owerwrite !? isnt there danger to break something ?! |On Thu, 2004-02-26 at 01:21, raptor wrote: | hi, | | I want to reinstall my gentoo, the problem is that i want to reistall it as I work on my current gentoo w/o | repartitioning... | | Can I install gentoo on a directory under root...i.e. | | | chroot /mynewgentoo /bin/bash | | then install as usual...what I should set in GRUB.. | What about of the possibility when everything runs OK.. | to move the whole installation into '/' | I misread your initial post. For best results, you should always keep a spare partition or two for experimentation. What you are proposing will not work. Even if you could get a new install done in a separate directory, there is no way to boot this oddball setup. Back to square one: 1. Why is it that you need to reinstall? Why can't you just upgrade what you have? 2. Harddrives are cheap. Just buy another and do your install on the second harddrive. Thne you will have space to do decent backups, for example. HTH, -- Collins Richey - Denver area gentoo testing 2.6.3 nptl udev -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Hotkeys / Omnibook on HP Pavilion with 2.6.3
On Thu, 26 Feb 2004 05:30:51 -0700 Scott Taylor [EMAIL PROTECTED] wrote: I've made my own tweaks to the omnibook, and been using this with great luck on my pavilion with 2.6.3: http://openpax.net/996-hpomnibook.patch which applies directly to the kernel, saving the hassle of using the original attempt at kernel integration which didn't work well with 2.6 Just idle curiosity (I don't have an omnibook). Why is it necessary to roll your own? Or is this patch something concocted by kernel developers and not yet released? Are the kernel developers even aware of this problem (they should be)? etc., etc. I guess the point is: I hope this isn't just a one-off, and I hope that a more standard approach will come from this!? My $.02. -- Collins Richey - Denver area gentoo testing 2.6.3 nptl udev -- [EMAIL PROTECTED] mailing list
[gentoo-user] using udev - mystery dependancy for devfsd
FYI,
I have just put up a new system with ~x86, 2.6.3_r2, nptl, and udev, and
all is working well except for alsa (that was broken before). There
were a couple of minor wrinkles.
Here's what I have done, in case that rings any bells. Running from a
fairly current gentoo 2.6 system.
1) Performed stage1 from the 20040204 tarball
make.conf has USE=-java MAKEOPTS=-j2
used bootstrap-2.6.sh
2) Performed stage2 as follows
make.conf updated to ACCEPT_KEYWORDS=~x86 USE=nptl
emerge linux-headers-2.6.1 (bummer this is masked, not just ~x86)
emerge system - failed in iputils (known error)
emerge autoconf openssl flex iputils
emerge system (again, successful)
3) Continued with lots of emerges from the chroot environment
make.conf updated to USE=X -kde -qt -arts -gnome tcltk jpeg tiff gif
nptl dvrd
emerge python xfce4 (this rebuilds python for tkinter and drags in
all the X related stuff)
emerge gentoo-dev-sources-2.6.3_rc2, built kernel, copied to /boot
(not a separate partition)
emerge grub wdm sysklogd hotplug
emerge -C devfsd
emerge udev
emerge grub aterm nfs-utils alsa-utils sudo aumix cdrutils hddtemp
emerge vixie-cron
tried to emerge nvidia-kernel and glx (but nvidia doesn't like the
chroot environment)
emerge gimp-2.0_pre3 (masked)
rc-update add {sysklogd hotplug nfs vixie-cron} default
rc-update add alsasound boot
fixed up remaining stuff for bootable system (using pre-existing grub
setup)
4) Total time for the above was about 13 hours (AthlonXP 1800+ 512MB)
5) boot from new system
bummer: something deleted /proc along the way. I checked to see that
it existed after untaring stage1!!!
recreated /proc and boot ok
emerge nvidia-kernel and nvidia-glx
copied over my existing XF86Config
did a wget for firefox-0.8 and set it up in /opt
startxfce4
tried to run an 'aterm' but failed immediately
firefox runs ok
research udev on another system.
learned that the default under udev is that users must be in group
tty in order to run a terminal session!!! (Bad idea ???). Fixed
my user. Success with aterm.
tried to emerge sylpheed-claws - failed trying metamail
discoverd fix in bugzilla, emerged metamail sylpheed-claws
copied over my setup for nfs, started nfs, rc-update add nfs default.
mounted my other system's / and copied over sylpheed directories
sylpheed runs and processes mail
copied over plugins for firefox from my other system
firefox works with plugins
gimp works
emerge openoffice-bin and copied over my documents from the other
system
openoffice works
5) now back to the mystery
emerge sync
emerge -puv world
[ebuild N] sys-fs/devfsd-1.3.25-r5 0 kB
[ebuild U ] app-shells/sash-3.7 [3.6] +readline 49 kB
[ebuild UD] sys-kernel/linux-headers-2.4.22 [2.6.1] -build
28,836 kB
ok try again
emerge -pUv world
[ebuild N] sys-fs/devfsd-1.3.25-r5 0 kB
[ebuild U ] app-shells/sash-3.7 [3.6] +readline 49 kB
Now that you have the background, let me restate the questions:
1) What is the best way to prevent world from trying to downgrade my
linux-headers? Inject linux-headers-2.4.22? A better way?
2) How do I find what is causing world to want to reintroduce devfsd?
I've tried some of the scripts posted in the past to search the IUSE
conditions, but I didn't find the answer?
Thanks for your patience.
--
Collins Richey - Denver area
gentoo 2004 2.6.3_r2 nptl udev
--
[EMAIL PROTECTED] mailing list
Re: [gentoo-user] xfree fails with kernel 2.6.3-rc2-mm1
On Sun, 15 Feb 2004 09:56:44 -0800 (PST) Jakob Schiøtz [EMAIL PROTECTED] wrote: Hi! I (finally) decided to try the 2.6 kernels, but I ran into a problem. X cannot start, I get the following messages in dmesg, and I only see the nVidia spash screen, then it is back in text-mode. atkbd.c: Unknown key released (translated set 2, code 0x7a on isa0060/serio0). atkbd.c: This is an XFree86 bug. It shouldn't access hardware directly. atkbd.c: Unknown key released (translated set 2, code 0x7a on isa0060/serio0). atkbd.c: This is an XFree86 bug. It shouldn't access hardware directly. These messages were added to the later 2.6 releases. The same condition existed silently in earlier releases. The messages are not critical on my system, but apparently some people have non-working keyboards related to this. You need to check your/var/log/XFree86.0.log to see why X will not start. -- Collins Richey - Denver Area gentoo 2004 2.6.3_r2 nptl udev -- [EMAIL PROTECTED] mailing list
[gentoo-user] experience with udev
FYI, My experience with udev-017 and 2.6.3_rc2 1) Everything appears to be working fine. 2) By default, udev requires users to be in group tty in order to run a terminal session (aterm,xterm, etc.) under X. Is this a udev standard, or is this something gentoo developers have cooked up? If the latter, are there plans to update the user skeleton(s) to accomplish this automatically wshen a new user is created? 3) In the article published at www.reactivated.net/udevrules.php, the author claims that external action must be taken to make udev work with nvidia's drivers. That is not the case for me. I have nvidia-kernel and nvidia-glx active, and I did not need to tweak udev at all. Enjoy, -- Collins Richey - Denver area gentoo 2004 2.6.3_rc2 nptl udev -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Artwiz fonts in (a|w)term
On Mon, 16 Feb 2004 02:26:18 +0530 Aniruddha Shankar [EMAIL PROTECTED] wrote: I really like the artwiz fonts and use aterm as my term ... for a really nice translucent wterm which changes translucency depending on whether or not it has focus, try aterm -tr -bg black -fg white -fade 50 -cr green -si -sr -sl 1000 -sh 70 (This works best if you have some cool wallpaper) I'm trying to get artwiz fonts in aterm and am having very little success. I installed artwiz-fonts, added /usr/share/fonts/artwiz to the Catalogue section of /etc/X11/fs/config and restarted the X font server. the rc-script told me it was scanning /usr/share/fonts/artwiz . I also try and run aterm with the -fn [fontname] option but I can't seem to get them working... another problem that I'm facing is that I don't get any artwiz font in fc-list. What am I doing wrong ? 1. Please turn off your 'receipt requested' setting. Very bad manners on mailing list. 2. Have you run 'fc-cache'? -- Collins Richey - Denver area gentoo 2004 2.6.3_rc2 nptl udev -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] using udev - mystery dependancy for devfsd
On Sun, 15 Feb 2004 15:10:46 -0600 Kevin Hanson [EMAIL PROTECTED] wrote: Daniel Drake wrote: Collins Richey wrote: 1) What is the best way to prevent world from trying to downgrade my linux-headers? Inject linux-headers-2.4.22? A better way? It is /etc/portage/package.mask...singular, not plural. Plus, you would use /etc/portage/package.unmask to UNmask packages. ok, thanks. I now have /etc/portage/package.mask =sys-kernel/linux-headers-2.6.1 =devfsd-1.3.24 /etc/portage/package.unmask =sys-kernel/linux-headers-2.6.1 Now I can issue emerge -puv world without an attempt to downgrade linux-headers, but I can find no combination of values that will prevent portage from trying to emerge devfsd. -- collins Richey - Denver area -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] experience with udev
On Sun, 15 Feb 2004 22:05:58 + Daniel Drake [EMAIL PROTECTED] wrote: Collins Richey wrote: 3) In the article published at www.reactivated.net/udevrules.php, the author claims that external action must be taken to make udev work with nvidia's drivers. That is not the case for me. I have nvidia-kernel and nvidia-glx active, and I did not need to tweak udev at all. Interesting. Are you autoloading the 'nvidia' module on bootup or anything? Or is it still unloaded at the point where you run 'startx' or equivalent? As I explained on the page, X tries to load the module if it isn't already loaded. I guess X loads modules slightly differently, as (for me, and another test subject) it just refuses to start without the /dev/nvidia* nodes already in place (or the module completely loaded). Were you using 2.6 and udev-017? I did absolutely nothing to get this to work other than emerging the nvidia ebuilds and appropriately coding XF86Config. Here's the extract from syslog Feb 15 14:58:32 richpc1 sudo: collins : TTY=vc/1 ; PWD=/home/collins ; USER=roo t ; COMMAND=/usr/bin/wdm [ the window manager starts X after I enter login data ] Feb 15 14:58:33 richpc1 kernel: nvidia: no version magic, tainting kernel. Feb 15 14:58:33 richpc1 kernel: nvidia: module license 'NVIDIA' taints kernel. Feb 15 14:58:33 richpc1 kernel: 0: nvidia: loading NVIDIA Linux x86 NVIDIA Kerne l Module 1.0-5336 Wed Jan 14 18:29:26 PST 2004 [ nvidia is being loaded ] Feb 15 14:58:36 richpc1 kernel: atkbd.c: Unknown key released (translated set 2, code 0x7a on isa0060/serio0). Feb 15 14:58:36 richpc1 kernel: atkbd.c: This is an XFree86 bug. It shouldn't ac cess hardware directly. [ the usual 2.6 garbage at startx time ] I have nothing in autoload for this (or any other module), nor have I altered anything in /etc/udev. With X running, I get the following; I haven't checked before starting the window manager. ls /dev/nv* /dev/nvidia0 /dev/nvidia2 /dev/nvidia4 /dev/nvidia6 /dev/nvidiactl /dev/nvidia1 /dev/nvidia3 /dev/nvidia5 /dev/nvidia7 HTH, -- Collins Richey - Denver area gentoo 2004 2.6.3_r2 nptl udev -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] xfree fails with kernel 2.6.3-rc2-mm1
On Mon, 16 Feb 2004 00:46:07 +0100 Hemmann, Volker Armin [EMAIL PROTECTED] wrote: On Sunday 15 February 2004 18:56, Jakob Schiøtz wrote: Hi! I (finally) decided to try the 2.6 kernels, but I ran into a problem. X cannot start, I get the following messages in dmesg, and I only see the nVidia spash screen, then it is back in text-mode. atkbd.c: Unknown key released (translated set 2, code 0x7a on isa0060/serio0). atkbd.c: This is an XFree86 bug. It shouldn't access hardware directly. atkbd.c: Unknown key released (translated set 2, code 0x7a on isa0060/serio0). atkbd.c: This is an XFree86 bug. It shouldn't access hardware directly. I get this message, if I hit one of my programmable keys and the wrong layer is active. This message does not and will not mean any critical error that could kill X. Have a look into /var/log/XF86.0.log there you'll find your problem. Apparently the kernel developers don't agree with you. You can google for their comments. I get these messages every time I start X. Apparently X code is making some sort of (big no no) direct access to the keyboard that should be handled via more standard methods. In some cases this has even led to unusable keyboards. In my case, it's just an annoyance, but who knows where this will lead. I don't know whether or not the kernel developers have communicated this fact to the xfree developers. According to what I read, the kernel used to silently ignore this problem, but they have decided to log the occurrences in the hope that this will be fixed. HTH, -- Collins Richey - Denver area gentoo 2004 2.6.3_r2 nptl udev -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] experience with udev
On Sun, 15 Feb 2004 17:33:14 -0700 Collins Richey [EMAIL PROTECTED] wrote: I have nothing in autoload for this (or any other module), nor have I altered anything in /etc/udev. I also altered nothing in any initscripts. I'm at baselayout-1.8.6.13 in case that makes any difference. This is an all ~x86 system. One change I just made. /etc/udev/udev.permissions has too restrictive (660) permissions for the /dev/nvid* devices, so I had to alter that to 666 to allow my screen saver to run. -- Collins Richey - Denver area gentoo 2004 2.6.3_r2 nptl udev -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] 2.6.2-r1
On Sun, 15 Feb 2004 18:52:53 -0600 Alec Berryman [EMAIL PROTECTED] wrote: On Sun, Feb 15, 2004 at 04:49:50PM -0800, Anthony Hoppe wrote: When moving from kernel 2.6.1-r1 to kernel 2.6.2-r1, can I use the same configuration file? Not straight up, but if you copy the config over and run a 'make oldconfig' it will prompt you for new questions that need to be answered. You won't need to run 'make config/menuconfig/xconfig' afterwards. The alternative I use is 'make menuconfig' then select load my existing .config before making any choices. -- Collins Richey - Denver area gentoo 2004 2.6.3_r2 nptl udev -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] bash-2.05b-r9 emerge fails
On Sun, 15 Feb 2004 22:47:04 -0500 Lincoln A. Baxter [EMAIL PROTECTED] wrote: portage wants to build build bash-2.05b-r9: current = -r7. Unfortunately this it fails to link: print_cmd.o(.text+0x1739): In function __builtin_va_start' print_cmd.o(.text+0x1927): In function __builtin_va_start' error.o(.text+0x108): In function __builtin_va_start' error.o(.text+0x1c3): In function __builtin_va_start' error.o(.text+0x243): In function __builtin_va_start' Don't know. I'm running a ~x86 system so I have gcc-3.3.2-r7 and glibc-2.3.3_pre2004020, and that level of bash built ok for my system. -- Collins Richey - Denver area gentoo 2004 2.6.3_r2 nptl udev -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] firefox compiles, won't start
On Mon, 16 Feb 2004 03:56:58 + [EMAIL PROTECTED] wrote: Sorry if this has already been covered. I have just now subscribed to this list and did not check the archinves. I am having problems starting firefox. This is the error message when attempting to start from the command line: bash-2.05b$ firefox /usr/bin/firefox: line 106: /firefox: No such file or directory did you emerge firefox (compiled) or firefox-bin (binary). If the latter, it will be in /opt/firefox.../firefox (don't know the exact name, since I download directly from mozilla land), and /opt/firefox... is most assuredly not on your $PATH. Either execute /full/path/to/firefox or create a /usr/local/bin/firefox symlink to /full/path/to/firefox to execute by the name firefox only. -- Collins Richey - Denver area gentoo 2004 2.6.3_r2 nptl udev -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Setting up 2.6 system from scratch
On Sat, 14 Feb 2004 15:33:02 +0100 Jonas Pedersen [EMAIL PROTECTED] wrote: Hi everybody I am going to install a fresh gentoo on my laptop in a week or so. I plan using kernel 2.6. I know gentoo is using devfs as default, but I can see that it is OBSOLOTE in 2.6 kernels. I can see devfs have been replaced by udev, which is a masked package in gentoo. How do I avoid gentoo installing the devfs stuff and using udev instead? Is there a way at all? Others may speak up, but from my readings, your don't want to touch udev with a fork - it's not ready to come out of the oven. devfs may be deprecated, but it still works just fine. Of course, if you want to be on the cutting edge, jump right in. You have been warned. Gentoo will no doubt offer udev when it's ready for public consumption. I can see there is a bootstrap-2.6.sh script. Is it a good idea to use this instead of the normal bootstrap.sh script? No experience with this. -- Collins - Denver Area - Gentoo stable kernel 2.6.3-rc2 -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Setting up 2.6 system from scratch
On Sat, 14 Feb 2004 11:09:48 -0600 Kathy Wills [EMAIL PROTECTED] wrote: Collins Richey wrote: Others may speak up, but from my readings, your don't want to touch udev with a fork - it's not ready to come out of the oven. devfs may be deprecated, but it still works just fine. Of course, if you want to be on the cutting edge, jump right in. You have been warned. Gentoo will no doubt offer udev when it's ready for public consumption. I have udev working just fine using the howtos found in gentoo forums. Cool. Does it do anything for you that devfs doesn't? -- Collins - Denver Area - Gentoo stable kernel 2.6.2-rc1 -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] CD burning fails to work
On Sun, 15 Feb 2004 00:23:54 + Stephen Liu [EMAIL PROTECTED] wrote: $ cdrecord --scanbus dev=ATAPI Cdrecord-Clone 2.01a25 (i586-pc-linux-gnu) Copyright (C) 1995-2004 Jörg Schilling scsidev: 'ATAPI' devname: 'ATAPI' scsibus: -2 target: -2 lun: -2 Warning: Using ATA Packet interface. Warning: The related libscg interface code is in pre alpha. Warning: There may be fatal problems. Using libscg version 'schily-0.8'. scsibus0: 0,0,0 0) 'MATSHITA' 'CD-ROM CR-588 ' 'LS15' Removable CD-ROM 0,1,0 1) * 0,2,0 2) * 0,3,0 3) * 0,4,0 4) * 0,5,0 5) * 0,6,0 6) * 0,7,0 7) * The CD-ROM is there but CD Writer get lost ! Stephen, I believe you are mixing apples and oranges. There are two possibilities: 1) Use ATAPI support - no SCSI emulation, no 'hdd=ide=scsi' in grub/lilo 2) Use SCSI emulation support - SCSI emulation, 'hdc=ide=scsi hdd=ide=scsi' By using the line 'hdd=ide=scsi', you have removed hdd from visibility as an ATAPI device, so of course it doesn't appear which you issue 'cdrecord -scanbus dev=ATAPI'. Whether or not you can use option 1) depends on how recent your 2.4 kernel is, or it's cool on 2.6, You appear to have an appropriate version of cdrecord. Apparently, even the latest versions of cdrdao with work with ATAPI support HTH, -- Collins - Denver Area - Gentoo stable kernel 2.6.2-rc1 -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Setting up 2.6 system from scratch
On Sat, 14 Feb 2004 18:06:46 + Daniel Drake [EMAIL PROTECTED] wrote: [ lots of helpful info snipped ] Thanks for the nice intro to udev! -- Collins - Denver Area - Gentoo stable kernel 2.6.2-rc1 -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Mail confirmation requests
On Sun, 15 Feb 2004 00:21:38 +0100 Matthias F. Brandstetter [EMAIL PROTECTED] wrote: -- quoting Peter Ruskin -- and you, grendel, have been added to mine. I think this list would be better without you and your nasty comments Is it wise to put someone into your killfile on a public mailing list? I mean, you would loose some information of one thread this way, wouldn't you? Some people, in life as well as on public mailing lists, are more ego driven than logic driven. To them dueling is more fun than thinking. It's best to ignore them. -- Collins - Denver Area - Gentoo stable kernel 2.6.2-rc2 -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] perfomance 2.6.3 = 2.4 more
On Thu, 12 Feb 2004 23:50:17 -0600 TriKster Abacus [EMAIL PROTECTED] wrote: I would be curious to know the memory size of the machines whose owners are singing the praises of, or conversely complaining about, 2.6. All my machines cept 1 are running on at least 512mb of ram ddr400 I think..? (dont usually watch exactly what I buy.. I just ask for the best atm). I have in my arsenal: Thanks for the rundown. I just can't remember: are you in the group reporting better performance, the same, or worse on 2.6? -- Collins - Denver Area - Gentoo stable kernel 2.6.2-rc1 -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] perfomance 2.6.3 = 2.4 - A-FRIGGIN-Men!
On Fri, 13 Feb 2004 18:24:12 +0600 (LKT) Grendel [EMAIL PROTECTED] wrote: ... DVD rips and encoding take long on 2.6 than 2.4 around 26 fps on 2.4 and ~ 16 fps on 2.6. It makes a large difference in time that extra 8fps so when I want to rip a dvd I switch to kernel 2.4. Yeah, but you have to rip a lot of DVDs to gain back the time you lose in rebooting. No one is claiming that 2.6 is faster across the board, but IBM has demonstrated that SMP performance is up by a factor of 5, and lots of users have demonstrated that desktop responsiveness has improved, so it's no big deal. Personally, I stopped running 2.4 6 months ago and haven't looked back. I was even able to get alsa marginally working on my sound card, and it didn't in 2.4. Enjoy, -- Collins - Denver Area - Gentoo stable kernel 2.6.2-rc1 -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Problem emerge scrollkeeper-0.3.14
On Fri, 13 Feb 2004 10:45:18 +0200 Gregory Staggel [EMAIL PROTECTED] wrote: But I can't find which ebuild packet contains XML::Parser dev-perl/XML-Parser -- Collins - Denver Area - Gentoo stable kernel 2.6.2-rc1 -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] unable to check root filesystem
On Fri, 13 Feb 2004 16:13:46 +0100 Christoph Gysin [EMAIL PROTECTED] wrote: my system doesn't boot correctly. it keeps hanging because checkroot complains not being able to check the root fs. [ rest snipped ] This won't help, but your experience only comfirms my decision never again to entrust data that I love to reiserfs. I know that almost everybody loves reiser, but I've experienced and read about too many failure scenarios. Good luck; maybe some of the reiser gurus can help you. -- Collins - Denver Area - Gentoo stable kernel 2.6.2-rc1 -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] perfomance 2.6.3 = 2.4 - A-FRIGGIN-Men!
On Fri, 13 Feb 2004 11:34:34 -0600 TriKster Abacus [EMAIL PROTECTED] wrote: [ lots of stuff that we don't need to repeat snipped ] TriKster, The essential flaw in your benchmarks is this: all of these items are single activities, whereas the changes in kernel 2.6 are not designed to improve the operation of single activities, but rather to improve the aggregate thruput of the system for l-o-t-s of activities, i.e. 1) more robust scalibility for 4 processors 2) better response for interactive jobs (i.e. no job should starve for service while a cpu hog is getting its cut off the top). This means, inevitably, that certain of your loved ones (DVD ripping, for example) may not be able to monopolize the CPU and thus may exhibit extended run times. If you want a fair picture of 2.6 vs. 2.4 operations, you must construct a very mixed, stressful combination of tasks to run opposite X sessions and measure not only the job throughput but the response time effect on the X sessions. Then you would repeat the study varying the number of cpus, memory, etc. I'm pretty sure you will find the same results that the IBM 8-way processor benchmarks did - individual component run times may lengthen, but the system is capable of accomplishing more concurrent work. Also, the IBM benchmarks demonstrated that 2.6 consumes a lot less kernel mode and a lot more user mode CPU time while getting more work through the box. I, personally, don't need to do this sort of benchmarking, because I remember full well the effect of running major compiles on 2.4 while trying to do something else useful. It wasn't a pretty picture unless you bumped the nice values for X, and it now works quite well without screwing with any settings, so that's as accurate a report as I need. YMMV. Enjoy, -- Collins - Denver Area - Gentoo stable kernel 2.6.2-rc1 -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] perfomance 2.6.3 = 2.4 - A-FRIGGIN-Men!
On Fri, 13 Feb 2004 12:57:09 -0600 TriKster Abacus [EMAIL PROTECTED] wrote: Collins Richey wrote: On Fri, 13 Feb 2004 11:34:34 -0600 TriKster Abacus [EMAIL PROTECTED] wrote: ... the changes in kernel 2.6 are not designed to improve the operation of single activities, but rather to improve the aggregate thruput of the system for l-o-t-s of activities, i.e. 1) more robust scalibility for 4 processors 2) better response for interactive jobs (i.e. no job should starve for service while a cpu hog is getting its cut off the top). Dude! OMG! are you not reading what the developers are saying about 2.6? They are saying that it has extreme performance gains for the desktop user. You might want to check this out: http://www.tinyminds.org/modules.php?op=modloadname=Newsfile=articlesid=845 I mean.. that is directly from Mr. RML (love-sources) himself! .. And he has quite a large paragraph speaking of the desktop qualities! start of excerpt Unfortunately, the interview confirms what I was saying. Here are relevant excerpts: TM: On to the actual questions, as Tinyminds.org is primarily a desktop user Linux site we were wondering if you would tell us a bit about the changes the desktop user can expect in the upcoming kernel 2.6? RML: Desktop users can expect a bit in 2.6, which is somewhat surprising as the goals of 2.5 were primarily to improve scalability and reduce bottlenecks in large machines. [ please read that again carefully and compare with my version! ] ... In short, 2.6 should be nicer on the desktop but for normal machines with normal workloads the improvement will not be huge since 2.4 was not that bad to begin with. But 2.6 will certainly be excellent. /end of excerpt Which part of this do you interpret as meaning that desktop users can expect fantastic gains in performance??? We're talking about modest improvements for n-o-r-m-a-l desktop users, not b***s to the wall gamesters or DVD ripping stations. I stick by my claims. 2.6 is better but the performance gains are subtle and most in evidence for interactive desktop users when there is a heavy load on the system. My kudos to the 2.6 developers. The system has been rock-solid stable on my system since late in 2.5.x days, and I could care less what is currently happening with 2.4 and its older (but still good) design. I'm a died-in-the-wool ext3 user. By the time a few more .releases have gone by, 2.6 should be rock solid for XFS and other fs as well. -- Collins - Denver Area - Gentoo stable kernel 2.6.2-rc1 -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] epiphany ebuild fails on configure
On Fri, 13 Feb 2004 21:57:31 -0600 Brendan Sullivan [EMAIL PROTECTED] wrote: started emerging epiphany and got this error. updated intltools to .30 like it asked what does this mean? please help! i love epiphany, and I (idiot as i am) cleaned out 1.0.6 before i compiled 1.0.7 configure: error: XML::Parser perl module is required for intltool !!! ERROR: net-www/epiphany-1.0.7 failed. !!! Function econf, Line 365, Exitcode 1 !!! econf failed 1. emerge XML-Parser epiphany 2. check on www.gentoo.org - bugs to see if this has been reported 3. If not, open a bugzilla account and report the bug The epiphany ebuild should be coded to detect the missing dependancy and emerge it automatically. -- Collins - Denver Area - Gentoo stable kernel 2.6.2-rc1 -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] bulloney 2.6.X benchmarks
On Thu, 12 Feb 2004 14:50:51 +0800 Jingtao Lv [EMAIL PROTECTED] wrote: On Thu, Feb 12, 2004 at 12:39:11PM +0600, Grendel wrote: On Thu, 12 Feb 2004, The awesome and feared Jingtao Lv commented thusly, I have no idea about this:( I am using a desktop:^) desktops also support ACPI, Just curious about one thing. If you run with 'ACPI=OFF' do you lose the ability to affect an actual power-off shutdown at halt time? -- Collins - Denver Area - Gentoo stable kernel 2.6.2-rc1 -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] New Install
On Thu, 12 Feb 2004 12:41:46 -0500 (EST) Ric Messier [EMAIL PROTECTED] wrote: On Thu, 12 Feb 2004, Rob Barnett wrote: I can not find the bzImage after I do a manual make. I mount /boot myself... How do I turn the kernel into a bzImage? bzImage is in /usr/src/linux/boot/arch/arch or something like that. You can also do a make install and have it put everything into place for you. To be sure, /usr/src/linux-/arch/i386/boot/ -- Collins - Denver Area - Gentoo stable kernel 2.6.2-rc1 -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] My Gentoo box revives (former: starting problem after 'emerge -u world')
On Fri, 13 Feb 2004 01:38:53 +0800 Stephen Liu [EMAIL PROTECTED] wrote: Edited as advised. But USER is not allowed to 'su -' on Konsole window Permission deny. Kindly advise how to fix it. Standard on linux (and a number of other distros), only users in group wheel may issue su to root. Add your user to group wheel. The cause of rendering the Gentoo box failing to reboot after emerge -u world' is still unknown. I only suspect 'upgrading betagenkernel' causing the box crashed. It proceeded automatically without asking a single question. I am in uncertainty of how to run 'emerge -u world' in future to avoid this mishap. I don't have a clue what happened. As a rule I only run 'emerge -p[uU]v world'; then I emerge packages that I like manually. If things like gcc, glibc, xfree, baselayout, etc. (i.e. very important packages), I usually wait a week or to to get comments from the list. HTH, -- Collins - Denver Area - Gentoo stable kernel 2.6.2-rc1 -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] perfomance 2.6.3 = 2.4
On Thu, 12 Feb 2004 23:07:36 +0100 Wazow [EMAIL PROTECTED] wrote: Guy Van Sanden wrote: I'm testing using kernel 2.6.3 (mm-sources rc2). On my system, everything is a little slower under 2.6 then it used to be in 2.4. measurable performance is only slightly worse (like dvdrip getting 2 fps less, 20 fps less for glxgears), but desktop responsiveness is noticably down under high loads. I tried with acpi on and off (no differnce) and I have preemtive enabled. So far, 2.6 is disappointing, specially since I heard it would be great performance-wise... Am I missing something? No clue. I do not know what are the measurements but GUI responsiveness wise I do have an opposite impression... From what I have read, the major performance gains are for MP machines. Check the recent IBM benchmarks that report a 5-1 gain in web pages served for a 24 hour test on an 8-processor machine! http://www-106.ibm.com/developerworks/linux/library/l-web26/index.html?ca=dgr-lnxw02KernCompare -- Collins - Denver Area - Gentoo stable kernel 2.6.2-rc1 -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Which kernel version to select?
On Fri, 13 Feb 2004 06:55:18 +0600 (LKT) Grendel [EMAIL PROTECTED] wrote: I managed to get gentoo running, the installation was the worst experience i had since I installed debian. You know some of the things I have to do are so common that why anyone cant write a intereactive install script to automate some aspects of the things is beyond me. Fortunately the harrowing aspefct of the installation is offsetted somewhat by the speed increase and good package management I see. I never understand why distros which have the best package management (gentoo, debian) have almost non existant installation support :( A couple of thoughts: 1) There are one or more projects underway to automate the install process. Search the archives or in forum. Most of us don't mind the manual process because normally you only need to do it once, then update forever. 2) I disagree. I've gotten better installation support from this mail group and from the gentoo forum users than any other distro I've worked with, bar none. 3) You will catch more bears with honey than vinegar. -- Collins - Denver Area - Gentoo stable kernel 2.6.3-rc2 -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Making cdrecord work
On Thu, 12 Feb 2004 21:08:36 +0600 (LKT) Grendel [EMAIL PROTECTED] wrote: No the cvs version of cdrdao works fine with ATAPI drives. cdrdao write --device /dev/hdd vice_city.toc works fine. I burned a lot of .bin files using it and had no problem. Good to know. I haven't used it. -- Collins - Denver Area - Gentoo stable kernel 2.6.3-rc2 -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] perfomance 2.6.3 = 2.4 B.S.!!!!!
On Thu, 12 Feb 2004 19:19:21 -0600 TriKster Abacus [EMAIL PROTECTED] wrote: From what I have read, the major performance gains are for MP machines. Ya, but the issue here is.. a desktop system, running basic programs.. I.E. X, xchat-2, mozilla, [your favorite email client here], fluxbox, kde whatever.. and a few Eterms/xterms There is not any noticeable improvement on speeds. Exactly my point. For the average desktop user there may be little or no improvement until the kernel developers have fine tuned the product a little more. Right now (only .3 release), they're still pretty much in cleanup mode. -- Collins - Denver Area - Gentoo stable kernel 2.6.2-rc1 -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Which kernel version to select?
On Fri, 13 Feb 2004 08:22:43 +0600 (LKT) Grendel [EMAIL PROTECTED] wrote: On Thu, 12 Feb 2004, The awesome and feared Collins Richey commented thusly, Yes, but I dont know if I am missing something here, but still this means that you need a lot of patience and a extremely high level of linux experience to install gentoo. Got it in one. Gentoo was not designed for the unexperienced, but a lot of unexperienced users have made the grade. Also there are annying things like something simple as emerge emu10k1 trying to download aumix and all its dependencies like xf86 :( This is (merely) a matter of understanding the USE variables. Always do an 'emerge -pv packagename' to understand what USE variables are recognized by the package. An example: xfce will drag in all of kde and gnome, if those USE variables are in effect! But all in all I like it, its just the installation process that sux. Currently I got the 2.6 kernel installed, but genkernel asks me to pass the options boot=/dev/ram0 real_boot=/dev/hda6, I tried but it dont work so I switched back to 2.4.20. Is there a way to get newer kernels ? Don't have a clue. genkernel has created more grief for more people than any other gentoo feature. I wouldn't touch it with a fork. -- Collins - Denver Area - Gentoo stable kernel 2.6.2-rc1 -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] perfomance 2.6.3 = 2.4 B.S.!!!!!
On Thu, 12 Feb 2004 20:10:50 -0600 TriKster Abacus [EMAIL PROTECTED] wrote: Gentoo Lists wrote: I don't want to feed this but what it really comes down to is if your not happy use the 2.4 kernels or make friendly constructive comments on the Kernel mailing lists Umm... that is complete bull, So what happens next? Are you going to have me banned from the list, because of my opinions? Get off your high horse and get a clue. [ lots of drivel deleted ] If I didn't know better, I would say it's full of the moon. Maybe both of you could ride a little lower on the horse. -- Collins - Denver Area - Gentoo stable kernel 2.6.2-rc1 -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] perfomance 2.6.3 = 2.4 - A-FRIGGIN-Men!
On Thu, 12 Feb 2004 18:18:16 -0800 Ian Truelsen [EMAIL PROTECTED] wrote: On Thu, 12 Feb 2004 19:05:09 -0600 TriKster Abacus [EMAIL PROTECTED] wrote: This just goes to prove that I am not the only one seeing this bullsh*ty ooohooohhhh ... 2.6.X is so great! crap Excuse me for interrupting, but does it make a lot of sense to be benchmarking what is essentially a development kernel? I imagine IBM did the benchmarks to get a feel for the improvements in the enterprise (i.e. multi-cpu server) support for 2.6. This is not to say that it won't imrove more after .10, etc. For the record, I have seen a slight downgrade in X performance with 2.6, but not much of one. However, I am not running 2.6 for speed, but to help in the development in whatever small way that I can. I am giving them a good 10 releases before I expect to see the same or better performance. I have noticed little difference in performance, and on my measly desktop, nothing but rock-solid stability for six months of 2.5/2.6. Be patient my son. -- Collins - Denver Area - Gentoo stable kernel 2.6.2-rc1 -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Safe? Multiple simultaneous emerge package-x
On Thu, 12 Feb 2004 21:24:12 -0500 Kevin [EMAIL PROTECTED] wrote: Is it safe to do multiple emerge commands (emerge package1 and emerge package2) simultaneously in different xterms or virtual consoles? Any exclusive locks on files or anything? In general, it's ok; just be sure that you don't try to emerge the same package in parallel! Do an 'emerge -pv ...' first. If there are no duplicates, fire away. Of course, you may slow everything down unless you have a very fast cpu. -- Collins - Denver Area - Gentoo stable kernel 2.6.2-rc1 -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Mail confirmation requests
On Fri, 13 Feb 2004 09:41:37 +0600 (LKT) Grendel [EMAIL PROTECTED] wrote: On Thu, 12 Feb 2004, The awesome and feared Glenn Johnson commented thusly, So the killfilter which i use is a bayesian filter, ie it has been trained to recognise people in email lists who have a annoying potential, Unfortunately, Grendel, many of your recent responses (not your technical queries) are prime candidates for your own killfilter, namely supremely annoying. -- Collins - Denver Area - Gentoo stable kernel 2.6.3-rc2 -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Safe? Multiple simultaneous emerge package-x
On Fri, 13 Feb 2004 11:44:14 +0800 senectus [EMAIL PROTECTED] wrote: Just out of interest.. Is it smart enough on a multi processor machine to dedicate one CPU for each emerge/compile? Also would it do a similar trick for hyper threaded CPU's? If by it, you mean emerge, then no. Allocation of CPU's to processes is totally under control of the kernel dispatching routines. -- Collins - Denver Area - Gentoo stable kernel 2.6.2-rc1 -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] perfomance 2.6.3 = 2.4 B.S.!!!!!
On Thu, 12 Feb 2004 23:51:43 -0500 Kurt Bechstein [EMAIL PROTECTED] wrote: [ snips ] Now having said that I'll throw out my $.02 here. The first thing I've noticed is that there is quite a difference between the 2.4 kernels depending on what you are using. Keep in mind the major improvements in 2.6 were not to make your mozilla launch faster. However, using vanilla-sources-2.4.24 versus stock 2.6.2 I see huge improvements in performance and responsiveness on my system. For example I started running some compiles on my system which taxed the system fairly well. Running vanilla-2.4.24 doing anything was pretty painful. Starting up the same compiles under 2.6.2 and then trying to do other things is a whole different story. I couldn't even tell I was running the compile which is nice. So I guess that is where I see major improvement under 2.6.x, that being when the system is under heavy load. Unfortunately, I've been running 2.5/2.6 too long now to remember 2.4 results all that well, but I can certainly echo your description of the compile process. I'm running an 'emerge -e system' to a clone of my system in a chroot right now. Idle time is consistently near 0%, but I scarcely notice that when reading email or browsing in another window. One of the things to keep in mind is that any system runs better with more memory. This machine is faster (P4 2.4 Gz) than my other (AthlonXP 1800+), but I have just under 256MB memory available, so I don't get quite as good results as on the Athlon (512MB). That machine (also a 2.6 system) positively screams through compiles. I would be curious to know the memory size of the machines whose owners are singing the praises of, or conversely complaining about, 2.6. -- Collins - Denver Area - Gentoo stable kernel 2.6.3-rc2 -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Starting problem after 'emerge -u world'
On Wed, 11 Feb 2004 13:23:47 +0800 Stephen Liu [EMAIL PROTECTED] wrote: special device /dev/hda3 does not exist * Checking root filesystem . Failed to open the filesystem. If the partition table has not changed, and the partition is valid and it really contains a reiserfs partition, then the superblock is corrupted and you need to run this utility with --rebuild-sb Warning... fsck.reiserfs for device /dev/ROOT exited with signal 6. * Filesystem coulddn't be fixed : ( [!!] The cure for this is simple, but not quick, and permanent. unmount /dev/hdxn (if mounted) mke2fs -j /dev/hdxn tune2fs -c 0 -i 0 /dev/hdxn reinstall software on /dev/hdxn I encountered exactly this same problem. reiserfs lost its marbles after changing unrelated partitions via fdisk, and there was no way to use the reiserfs partition again. So, I reverted to ext3, and I will never touch reiserfs again. ext3 doesn't loose its marbles. HTH, -- Collins - Denver Area - Gentoo stable kernel 2.6.2-rc1 -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Starting problem after 'emerge -u world'
On Wed, 11 Feb 2004 18:55:07 +0600 (LKT) Grendel [EMAIL PROTECTED] wrote: N! Please DONT install ext3, ext3 is just a ext2 file system with journalling support. ext2 is a slow file system and ext3 is even slower, it also has to keep a kjournald thread running taking up your CPU. PLease dont use a slow decadent file system like ext3. If you want the fastest performance ie you are handling large files then XFS is the best. Second would be reiserfs. Reiserfs is a stable file system, if you run a 2.4.20 or above kernel reiserfs should be stabler than ext3. I can't disagree with much of what you have to say, but my experience (and I've read numerous other reports) with reiserfs has been less than sterling. When it fails, it fails big time, i.e. not just a few files, but loss of fs. Most of my respondents on another mailing list wouldn't use anything except XFS. I'm waiting for stability on 2.6. There are a lot of XFS patches going into 2.6 as we speak. My recommendation for ext3 was not based on performance, just on years of rock-solid, error free operation. ext3 may be somewhat slow, but I've never lost any data with ext3. Thanks, -- Collins - Denver Area - Gentoo stable kernel 2.6.2-rc1 -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] No /dev/dsp created with Alsa?
On Wed, 11 Feb 2004 08:40:19 -0600 Thomas T. Veldhouse [EMAIL PROTECTED] wrote: Has anybody else had this happen? I have OSS emulation setup. I try to run mpg123 and it complains that there is no /dev/dsp ... and there isn't. [EMAIL PROTECTED] modules.d # /etc/init.d/alsasound start * Loading ALSA drivers... * Using ALSA OSS emulation * Loading: snd-emu10k1 * Loading: snd-emu10k1-synth * Loading: snd-seq-midi * Running card-dependent scripts * Restoring Mixer Levels [ ok ] [EMAIL PROTECTED] modules.d # * media-sound/alsa-driver Latest version available: 1.0.2c Latest version installed: 1.0.2c Size of downloaded files: 1,664 kB Homepage:http://www.alsa-project.org/ Description: Advanced Linux Sound Architecture kernel modules License: GPL-2 LGPL-2.1 Any fix for this? Is it a bug? Not the same card, but I have ens1371 (SB16 PCI), and with alsa I have to modprobe- rmmod -modprobe ens1371 to get /dev/dsp. When the module is initially loaded, devfsd does not generate the device file! Reported as bug, but no fix in sight. No problem using OSS, however. HTH, -- Collins - Denver Area - Gentoo stable kernel 2.6.2-rc1 -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] No /dev/dsp created with Alsa?
On Wed, 11 Feb 2004 10:19:52 -0600 Thomas T. Veldhouse [EMAIL PROTECTED] wrote: Collins Richey wrote: Not the same card, but I have ens1371 (SB16 PCI), and with alsa I have to modprobe- rmmod -modprobe ens1371 to get /dev/dsp. When the module is initially loaded, devfsd does not generate the device file! Reported as bug, but no fix in sight. No problem using OSS, however. HTH, Well, doesn't that modprobe sequence actually load the OSS module and not the alsa module? ... that is why /dev/dsp was created I assume. Tom Veldhouse To be more exact, snd_ens1371, which is the alsa module. The problem is real and alsa, I assure you. -- Collins - Denver Area - Gentoo stable kernel 2.6.2-rc1 -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Starting problem after 'emerge -u world'
On Thu, 12 Feb 2004 01:07:13 +0800 Stephen Liu [EMAIL PROTECTED] wrote: The cause of my havoc differs from yours. It happened only after running 'emerge -u world' and rebooted automatically. I strongly believe 'upgrading betagenkernel' is critical which was also done automatically without asking a question. Can I copy files from /home/user directory by attaching this Gentoo hard drive as slave to another Linux box? Or is there any other ways to get back the files in that directory. If I understand your problem correctly, you are going to need to perform a successful reiserfs recovery, and I'm not the right person to consult about that, since every thing I tried failed miserably. If you can successfully mount the partition in read only mode (I was able to do so at my failure), you can certainly mount the partition on another linux system with reiserfs support and try to copy off the files you need. Knoppix (CD based system) is an excellent recovery tool. I wish you the best of luck. I'm sure you must have read the replies to my suggestion. Filesystem discussions most often result in a flame war. I know lots of people use reiserfs successfully, but my two attempts over a period of 2-3 years resulted in disaster, so I have signed off permanently. The fact that the filesystem can (under somewhat diverse circumstances) fail to locate the superblock makes me question the total design, since the purpose of a journaled filesystem is to recover properly when the system is out of sync. Once you get out of the woods, you can decide what to do. I personally have never never had a single problem with ext3 (oops, see below); others swear by XFS (running servers for many years). My only objection to XFS (until recently) was that you had to obtain separate kernel patches, and these had to be reevaluated each time the kernel changed. XFS has now been ported into the 2.6 kernel, but the level of XFS activity recently in kernel releases makes me suspect that it is not quite ready for prime time. Now that I remember, there was one oddity on my system with ext3 at least 1 1/2 years ago. I had a single directory with a small number of files under my home directory that would not permit writes. I had to copy off the files (all were intact) and rebuild the directory. This pales in comparison to what you are experiencing. -- Collins - Denver Area - Gentoo stable kernel 2.6.2-rc1 -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] No /dev/dsp created with Alsa?
On Wed, 11 Feb 2004 10:40:05 -0600 Thomas T. Veldhouse [EMAIL PROTECTED] wrote: Collins Richey wrote: Sorry, it was worth a try. Just to be sure. What I observed was: 1) alsa module loaded automatically (brief pop to indicate the card had been touched). 2) no /dev/dsp 3) rmmod and modprobe again (brief pop to indicate that the sound card had been touched) 4) /dev/dsp now exists. The situation still exists on my other machine. On latest 2.6.2 I get the same behavior, but now the card produces no sound. alsa has great intent, but poor delivery for some cards, most especially since the card worked fine with OSS! OTOH, I'm not really sure whether this is an alsa or a devfs problem. -- Collins - Denver Area - Gentoo stable kernel 2.6.2-rc1 -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] No /dev/dsp created with Alsa?
On Thu, 12 Feb 2004 00:25:34 +0600 (LKT) Grendel [EMAIL PROTECTED] wrote: point in it. Alsa setup is also a bit of a PITA. No, its a major PITA. -- Collins - Denver Area - Gentoo stable kernel 2.6.2-rc1 -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] portage -u world broken by /var/cache/edb/world
On Wed, 11 Feb 2004 10:58:41 -0700 Jared Thirsk [EMAIL PROTECTED] wrote: On February 10, 2004 15:42, phil wrote: Are you running reiserfs or have had any lockups? when i first installed gentoo i was using rfs and locked it a few times (badly configured kernel ,acpi-nforce2 issue) which badly messed up some files and forced me to reinstall with ext3,which seems to be a bit more forgiving. Yes I'm running reiserfs, and have gotten a few times, although I'm not sure what from. Maybe I'll try ext3 next time. Jeez, I must be doing something right. g Of course, I was immediately beaten about the head and shoulders for suggesting this! -- Collins - Denver Area - Gentoo stable kernel 2.6.2-rc1 -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] Pretty CUPS frontend
On Wed, 11 Feb 2004 15:56:08 -0500 Barry Marler [EMAIL PROTECTED] wrote: On Wed, 11 Feb 2004 04:53:32 -0800 Anthony Hoppe [EMAIL PROTECTED] wrote: Is there a pretty CUPS frontend I can emerge...or a printer config tool that interfaces with CUPS? What's wrong with the standard cups http://localhost:631 interface? Looks pretty enough for my needs. -- Collins - Denver Area - Gentoo stable kernel 2.6.2-rc1 -- [EMAIL PROTECTED] mailing list
Re: [gentoo-user] No /dev/dsp created with Alsa?
On Thu, 12 Feb 2004 08:31:46 +1100 Andrew Cowie [EMAIL PROTECTED] wrote: On Thu, 2004-02-12 at 04:05, Thomas T. Veldhouse wrote: ALSA does not build the OSS compat code into it unless OSS is set in the USE variable. Is this the way we want it? Somewhat tangentially, If I have 2.4 system, built happily a year ago with USE=-alsa, and I want to upgrade to 2.6 and switch to ALSA, then how do I approach that? Switching the USE variable is obvious, sure :) [as is configure ALSA in 2.6 - thanks] but I'm wondering how to reach through the package stack and recompile that which needs recompiling. Other than 2.6 configuration, the only thing you need any more is alsa-lib and alsa-tools (set your /usr/src/linux symlink). The alsa aliases, etc. are practically (entirely?) identical. -- Collins - Denver Area - Gentoo stable kernel 2.6.2-rc1 -- [EMAIL PROTECTED] mailing list
