Re: [gentoo-user] flash changes
On Mon, Jun 21, 2010 at 05:40, W.Kenworthy wrote: > So is there anything better for flash when using firefox to browse the > web? You can try gnash. Ward
Re: [gentoo-user] Boot hangs after install, no error
> This is funny. I have NEVER got a genkernel to work on my system. > Actually, on any system. I'm not sure the OP would know that kernel is > any better then the one he makes. Dale, If you've never gotten genkernel to work, you should try this little script that I've used for the past few years. I put it in /usr/src/gk and I change into whatever /usr/src/kernel directory I'm going to compile. Then, I just call "../gk all" and off it goes. Of course, if you use lilo, it's a different story because I jumped out of the lilo life raft years ago and managed to swim to shore. Here's gk. Tweak to your environment: CFLAGS="-O2 -march=barcelona -pipe" \ genkernel --lvm --menuconfig --save-config --oldconfig \ --bootloader=grub --install --symlink --kerneldir=$PWD \ --makeopts="-s -j4" "$@" -- Bill
Re: [gentoo-user] Boot hangs after install, no error
Bill Longman wrote: This is funny. I have NEVER got a genkernel to work on my system. Actually, on any system. I'm not sure the OP would know that kernel is any better then the one he makes. Dale, If you've never gotten genkernel to work, you should try this little script that I've used for the past few years. I put it in /usr/src/gk and I change into whatever /usr/src/kernel directory I'm going to compile. Then, I just call "../gk all" and off it goes. Of course, if you use lilo, it's a different story because I jumped out of the lilo life raft years ago and managed to swim to shore. Here's gk. Tweak to your environment: CFLAGS="-O2 -march=barcelona -pipe" \ genkernel --lvm --menuconfig --save-config --oldconfig \ --bootloader=grub --install --symlink --kerneldir=$PWD \ --makeopts="-s -j4" "$@" -- Bill I don't use genkernel anymore. I just roll my own. That way, I know what is in there and what is not. Then if something doesn't work, I know if it is the kernel or something else. With genkernel, you won't have a clue what it is since you don't know much if anything about the kernel and how it is configured. I'll pass. As I have seen with others, genkernel doesn't work consistently enough. Dale :-) :-)
Re: [gentoo-user] Boot hangs after install, no error
Dale writes: > I don't use genkernel anymore. I just roll my own. That way, I know > what is in there and what is not. Then if something doesn't work, I > know if it is the kernel or something else. With genkernel, you won't > have a clue what it is since you don't know much if anything about the > kernel and how it is configured. That's not necessarily true. When I create a new kernel, I copy /usr/src/linux/.config into the new kernel directory, make oldconfig and menuconfig just as I like my kernel to be, and recreate the linux symlink to the new kernel directory. Then I do a genkernel --install --lvm --luks all && emerge -a @module-rebuild, and am done. I never noticed genkernel changing anything in my configuration, .config, /proc/config.gz and the stuff in /etc/kernels/ are identical. Until not long ago, I did not even know that genkernel was intended to create a working kernel from scratch. Wonko
Re: [gentoo-user] Boot hangs after install, no error
Alex Schuster wrote: Dale writes: I don't use genkernel anymore. I just roll my own. That way, I know what is in there and what is not. Then if something doesn't work, I know if it is the kernel or something else. With genkernel, you won't have a clue what it is since you don't know much if anything about the kernel and how it is configured. That's not necessarily true. When I create a new kernel, I copy /usr/src/linux/.config into the new kernel directory, make oldconfig and menuconfig just as I like my kernel to be, and recreate the linux symlink to the new kernel directory. Then I do a genkernel --install --lvm --luks all&& emerge -a @module-rebuild, and am done. I never noticed genkernel changing anything in my configuration, .config, /proc/config.gz and the stuff in /etc/kernels/ are identical. Until not long ago, I did not even know that genkernel was intended to create a working kernel from scratch. Wonko I always do mine this way. I copy the .config from the old kernel to the new kernel, run make oldconfig then afterwards make all && make modules_install and then copy the kernel to /boot with my own numbering system. That way I know which version and series the kernel is. After that, edit grub with the new kernel and I'm done. I have only had that fail once in the past six years or so and the kernel made some serious changes and I had to start from scratch that one time. They moved things around and oldconfig couldn't reorganize things on the new kernel. Point being, genkernal causes issues for people and they don't know how to fix it because they expect genkernel to do everything. Problem with that is that usually when someone has a kernel problem, they use genkernel. If they do their own, it just works. Now someone new to building a kernel may need some help but apparently genkernel needs some help anyway. May as well learn how to roll your own. This is Gentoo after all. Dale :-) :-)
Re: [gentoo-user] Boot hangs after install, no error
On 06/21/2010 12:01 PM, Dale wrote: > Alex Schuster wrote: >> Dale writes: >> >> >>> I don't use genkernel anymore. I just roll my own. That way, I know >>> what is in there and what is not. Then if something doesn't work, I >>> know if it is the kernel or something else. With genkernel, you won't >>> have a clue what it is since you don't know much if anything about the >>> kernel and how it is configured. >>> >> That's not necessarily true. When I create a new kernel, I copy >> /usr/src/linux/.config into the new kernel directory, make oldconfig and >> menuconfig just as I like my kernel to be, and recreate the linux symlink >> to the new kernel directory. Then I do a genkernel --install --lvm --luks >> all&& emerge -a @module-rebuild, and am done. >> I never noticed genkernel changing anything in my configuration, .config, >> /proc/config.gz and the stuff in /etc/kernels/ are identical. Until not >> long ago, I did not even know that genkernel was intended to create a >> working kernel from scratch. >> >> Wonko >> >> > > I always do mine this way. I copy the .config from the old kernel to > the new kernel, run make oldconfig then afterwards make all && make > modules_install and then copy the kernel to /boot with my own numbering > system. That way I know which version and series the kernel is. After > that, edit grub with the new kernel and I'm done. I have only had that > fail once in the past six years or so and the kernel made some serious > changes and I had to start from scratch that one time. They moved > things around and oldconfig couldn't reorganize things on the new kernel. > > Point being, genkernal causes issues for people and they don't know how > to fix it because they expect genkernel to do everything. Problem with > that is that usually when someone has a kernel problem, they use > genkernel. If they do their own, it just works. Now someone new to > building a kernel may need some help but apparently genkernel needs some > help anyway. May as well learn how to roll your own. This is Gentoo > after all. The only thing that genkernel would add is your initrd. The kernel is exactly the same, whether you compile it with "make" or through "genkernel". Do a test and you'll see. (I'm assuming we're both talking about gentoo-sources, not vanilla-sources. Either way, they'd be the same.) Some might be confused about what happens in the steps if they haven't been down the "kernel compilation trail" more than once or twice, but for folks who just want to compile their kernel and plop it into place, along with a hands-off initrd, it's rather handy.
[gentoo-user] evince stabilization request at bugzilla
I found out evince 2.30 made its way into the tree, but I see no "bug" in bugzilla regarding evince stabilization. As the commit message states "new version for Gnome 2.30", does this means I should look at the bug to stabilize Gnome 2.30 (#324077) instead? -- TIA, Nuno J. Silva gopher://sdf-eu.org/1/users/njsg
Re: [gentoo-user] Boot hangs after install, no error
Bill Longman wrote: On 06/21/2010 12:01 PM, Dale wrote: Alex Schuster wrote: Dale writes: I don't use genkernel anymore. I just roll my own. That way, I know what is in there and what is not. Then if something doesn't work, I know if it is the kernel or something else. With genkernel, you won't have a clue what it is since you don't know much if anything about the kernel and how it is configured. That's not necessarily true. When I create a new kernel, I copy /usr/src/linux/.config into the new kernel directory, make oldconfig and menuconfig just as I like my kernel to be, and recreate the linux symlink to the new kernel directory. Then I do a genkernel --install --lvm --luks all&& emerge -a @module-rebuild, and am done. I never noticed genkernel changing anything in my configuration, .config, /proc/config.gz and the stuff in /etc/kernels/ are identical. Until not long ago, I did not even know that genkernel was intended to create a working kernel from scratch. Wonko I always do mine this way. I copy the .config from the old kernel to the new kernel, run make oldconfig then afterwards make all&& make modules_install and then copy the kernel to /boot with my own numbering system. That way I know which version and series the kernel is. After that, edit grub with the new kernel and I'm done. I have only had that fail once in the past six years or so and the kernel made some serious changes and I had to start from scratch that one time. They moved things around and oldconfig couldn't reorganize things on the new kernel. Point being, genkernal causes issues for people and they don't know how to fix it because they expect genkernel to do everything. Problem with that is that usually when someone has a kernel problem, they use genkernel. If they do their own, it just works. Now someone new to building a kernel may need some help but apparently genkernel needs some help anyway. May as well learn how to roll your own. This is Gentoo after all. The only thing that genkernel would add is your initrd. The kernel is exactly the same, whether you compile it with "make" or through "genkernel". Do a test and you'll see. (I'm assuming we're both talking about gentoo-sources, not vanilla-sources. Either way, they'd be the same.) Some might be confused about what happens in the steps if they haven't been down the "kernel compilation trail" more than once or twice, but for folks who just want to compile their kernel and plop it into place, along with a hands-off initrd, it's rather handy. But only if it works. When I compile my kernel, I KNOW for sure what is in there. When genkernel does one, especially on a new install, I have no idea what is in it or what is not. If something goes wrong, I don't know where to start. Is it a kernel problem or is it something else? Who knows. Then you have to go back and start from the bottom, usually the kernel, and work your way back up to find out what is broken. Genkernel may work for you but that doesn't mean it does for everyone else. Should I mention hal here? When someone comes for help, your looking for the failure not the successes. If it was sucessful, they wouldn't need help. Dale :-) :-)
[gentoo-user] Samba freakiness; why isn't it reading MY /etc/samba/smb.conf file?
I just rebooted the computer to make sure this wasn't some weird RAM remnant, but the computer booted up: carter ~ # /etc/init.d/samba status * status: started carter ~ # cat /etc/samba/smb.conf [global] workgroup = MYGROUP security = user encrypt passwords = yes guest account = guest wins support = yes local master = yes os level = 99 domain master = yes preferred master = yes hosts allow = 192.168.1. 127. interfaces = eth0 [tmp] path=/tmp writeable=yes [homes] path=/samba/michael valid users=michael writable=yes carter ~ # testparm Load smb config files from /etc/samba/smb.conf rlimit_max: rlimit_max (1024) below minimum Windows limit (16384) Processing section "[tmp]" Processing section "[homes]" Loaded services file OK. Server role: ROLE_STANDALONE Press enter to see a dump of your service definitions [global] workgroup = MYGROUP interfaces = eth0 guest account = guest os level = 99 preferred master = Yes domain master = Yes wins support = Yes hosts allow = 192.168.1., 127. [tmp] path = /tmp read only = No [homes] path = /samba/michael valid users = michael read only = No At the top there you see that Samba IS started. Under that you see MY smb.conf, and at the bottom you see the smb.conf that's being loaded. Nothing really in the logs that would suggest what's going on here: carter ~ # cat /var/log/samba/* cat: /var/log/samba/cores: Is a directory [2010/06/21 16:39:34, 0] nmbd/nmbd.c:854(main) nmbd version 3.4.6 started. Copyright Andrew Tridgell and the Samba Team 1992-2009 [2010/06/21 16:39:34, 0] nmbd/asyncdns.c:155(start_async_dns) started asyncdns process 20980 [2010/06/21 16:39:34, 0] nmbd/nmbd_become_dmb.c:337(become_domain_master_browser_wins) become_domain_master_browser_wins: Attempting to become domain master browser on workgroup MYGROUP, subnet UNICAST_SUBNET. [2010/06/21 16:39:34, 0] nmbd/nmbd_become_dmb.c:351(become_domain_master_browser_wins) become_domain_master_browser_wins: querying WINS server from IP 192.168.1.2 for domain master browser name MYGROUP<1b> on workgroup MYGROUP [2010/06/21 16:39:40, 0] nmbd/nmbd_become_dmb.c:110(become_domain_master_stage2) * Samba server CARTER is now a domain master browser for workgroup MYGROUP on subnet UNICAST_SUBNET * [2010/06/21 16:39:40, 0] nmbd/nmbd_become_dmb.c:292(become_domain_master_browser_bcast) become_domain_master_browser_bcast: Attempting to become domain master browser on workgroup MYGROUP on subnet 192.168.1.2 [2010/06/21 16:39:40, 0] nmbd/nmbd_become_dmb.c:305(become_domain_master_browser_bcast) become_domain_master_browser_bcast: querying subnet 192.168.1.2 for domain master browser on workgroup MYGROUP [2010/06/21 16:39:48, 0] nmbd/nmbd_become_dmb.c:110(become_domain_master_stage2) * Samba server CARTER is now a domain master browser for workgroup MYGROUP on subnet 192.168.1.2 * [2010/06/21 16:39:56, 0] nmbd/nmbd_become_lmb.c:395(become_local_master_stage2) * Samba name server CARTER is now a local master browser for workgroup MYGROUP on subnet 192.168.1.2 * [2010/06/21 16:39:33, 0] smbd/server.c:1073(main) smbd version 3.4.6 started. Copyright Andrew Tridgell and the Samba Team 1992-2009 [2010/06/21 16:39:33, 0] printing/print_cups.c:103(cups_connect) Unable to connect to CUPS server /var/run/cups/cups.sock:631 - No such file or directory [2010/06/21 16:39:33, 0] printing/print_cups.c:103(cups_connect) Unable to connect to CUPS server /var/run/cups/cups.sock:631 - No such file or directory [2010/06/21 16:39:34, 0] smbd/server.c:457(smbd_open_one_socket) smbd_open_once_socket: open_socket_in: Address already in use [2010/06/21 16:39:34, 0] smbd/server.c:457(smbd_open_one_socket) smbd_open_once_socket: open_socket_in: Address already in use
Re: [gentoo-user] Boot hangs after install, no error
On 06/21/2010 01:23 PM, Dale wrote: >> The only thing that genkernel would add is your initrd. The kernel is >> exactly the same, whether you compile it with "make" or through >> "genkernel". Do a test and you'll see. (I'm assuming we're both talking >> about gentoo-sources, not vanilla-sources. Either way, they'd be the >> same.) Some might be confused about what happens in the steps if they >> haven't been down the "kernel compilation trail" more than once or >> twice, but for folks who just want to compile their kernel and plop it >> into place, along with a hands-off initrd, it's rather handy. >> >> > > But only if it works. When I compile my kernel, I KNOW for sure what is > in there. When genkernel does one, especially on a new install, I have > no idea what is in it or what is not. If something goes wrong, I don't > know where to start. Is it a kernel problem or is it something else? > Who knows. Then you have to go back and start from the bottom, usually > the kernel, and work your way back up to find out what is broken. By "But only if it works," I assume the antecedent "it" refers to is a kernel that we're attempting to boot correctly. (In other words, you're not talking about genkernel failing to create a kernel for you. Is that correct?) If someone has trouble on an initial install, then that just means they didn't configure the kernel correctly, is what I interpret that to mean. The result of "make" and the result of "genkernel kernel" are exactly the same. If your "make menuconfig" creates an invalid .config file for you, no sort of magic is going to make its resultant kernel valid. Do you mean to say that you just grab a kernel, jump into the directory and say "make" without an mrproper and some sort of config? You do realize that genkernel has --menuconfig, --xconfig and --gconfig exactly for this purpose, don't you? What sort of things do you believe genkernel is adding to your kernel? If you use "genkernel --menuconfig --no-install kernel", you can look and see what it did. It's no different than running "make menuconfig" followed by a "make; make modules". Just look in /usr/share/genkernel at the gen_compile.sh and you'll see that it does a make. > Genkernel may work for you but that doesn't mean it does for everyone > else. Should I mention hal here? When someone comes for help, your > looking for the failure not the successes. If it was sucessful, they > wouldn't need help. Which is why I mentioned genkernel in the first place. Most times a hang after boot is due to components that were missed in the kernel build -- from where? -- from a missing or incorrectly created initrd if the required modules weren't compiled into the kernel. The easiest way that I've seen is to use genkernel and get back to work. Then later on you can find out what an initrd is and why it's needed with modules but at least you'd have a running system. No, I don't think you should mention hal because it's probably OT for a thread about a hung boot. But you should apply to yourself a similar logic you ask of me: if others can use genkernel successfully, why can't I? Bill
[gentoo-user] Questions re swap and hibernate interaction on 8 gig machine
I just got a brand new custom-built 8 gig machine. There's an outfit in north Toronto that has MSI motherboards with PS/2 ports, so I can keep my genuine IBM PS/2 clickety-clack-keyboard; woho. And the integrated Intel graphics chip has *BOTH VGA AND DIGITAL OUTPUTS*! Anyhow, I have 8 gigs of ram on the sytem (will obviously be 64-bit Gentoo) and I want to know how much swap I need. The general rule of thumb is twice the ram. In this case, it would be 16 gigs. I think that it may not need swap when up, unless I do some heavy duty stuff. My main concern about a swap partition is how much I need for hibernate-to-disk to work. Is there a rule about this, or should I simply allocate 16 gigs out of my terabyte drive, and play it safe? -- Walter Dnes
Re: [gentoo-user] Questions re swap and hibernate interaction on 8 gig machine
On Tuesday 22 June 2010 00:04:14 Walter Dnes wrote: > I just got a brand new custom-built 8 gig machine. There's an outfit > in north Toronto that has MSI motherboards with PS/2 ports, so I can > keep my genuine IBM PS/2 clickety-clack-keyboard; woho. And the > integrated Intel graphics chip has *BOTH VGA AND DIGITAL OUTPUTS*! > > Anyhow, I have 8 gigs of ram on the sytem (will obviously be 64-bit > Gentoo) and I want to know how much swap I need. The general rule of > thumb is twice the ram. In this case, it would be 16 gigs. I think > that it may not need swap when up, unless I do some heavy duty stuff. > My main concern about a swap partition is how much I need for > hibernate-to-disk to work. Is there a rule about this, or should I > simply allocate 16 gigs out of my terabyte drive, and play it safe? With 8G of ram, you will likely never ever use a single bit of swap for the entire life of the machine. There is no such thing as a decent rule of thumb for how much swap. What does exist, is the following: "What intelligent-sounding (but actually dumb) answer can we give to this infernal question that keeps coming up that will make the user shut up and go away satisfied (regardless of the correctness and workability of the answer)?" That answer is, of course, "Twice your RAM". Said answer is also, bullshit[1]. If I said "42!" it would have made as much semantic sense. Your swap needs depend totally on your usage. There is no rule of thumb[2]. [1] Long ago when 386's were all the rage, 2 X RAM did make some sense. You do not have a 386 and 2 X RAM does not make sense with the hardware you have. [2] If you plan to suspend to disk you will need a certain minimum amount of swap for that. But you already know that, so I'd create that minimum amount. -- alan dot mckinnon at gmail dot com
Re: [gentoo-user] Samba freakiness; why isn't it reading MY /etc/samba/smb.conf file?
On Monday 21 June 2010 23:44:58 Michael Sullivan wrote: > I just rebooted the computer to make sure this wasn't some weird RAM > remnant, but the computer booted up: Everything below is correct. What makes you think its wrong? testparm does not dump a config file, it tells you what setting are IN EFFECT. "writeable=yes" and "read only = No" are exactly the same thing. You used the former, samba uses the latter internally and translated it. There is no problem here. > > carter ~ # /etc/init.d/samba status > * status: started > carter ~ # cat /etc/samba/smb.conf > [global] > workgroup = MYGROUP > security = user > encrypt passwords = yes > guest account = guest > wins support = yes > local master = yes > os level = 99 > domain master = yes > preferred master = yes > hosts allow = 192.168.1. 127. > interfaces = eth0 > > [tmp] > path=/tmp > writeable=yes > > [homes] > path=/samba/michael > valid users=michael > writable=yes > carter ~ # testparm > Load smb config files from /etc/samba/smb.conf > rlimit_max: rlimit_max (1024) below minimum Windows limit (16384) > Processing section "[tmp]" > Processing section "[homes]" > Loaded services file OK. > Server role: ROLE_STANDALONE > Press enter to see a dump of your service definitions > > [global] > workgroup = MYGROUP > interfaces = eth0 > guest account = guest > os level = 99 > preferred master = Yes > domain master = Yes > wins support = Yes > hosts allow = 192.168.1., 127. > > [tmp] > path = /tmp > read only = No > > [homes] > path = /samba/michael > valid users = michael > read only = No > > At the top there you see that Samba IS started. Under that you see MY > smb.conf, and at the bottom you see the smb.conf that's being loaded. > Nothing really in the logs that would suggest what's going on here: > > carter ~ # cat /var/log/samba/* > cat: /var/log/samba/cores: Is a directory > [2010/06/21 16:39:34, 0] nmbd/nmbd.c:854(main) > nmbd version 3.4.6 started. > Copyright Andrew Tridgell and the Samba Team 1992-2009 > [2010/06/21 16:39:34, 0] nmbd/asyncdns.c:155(start_async_dns) > started asyncdns process 20980 > [2010/06/21 16:39:34, 0] > nmbd/nmbd_become_dmb.c:337(become_domain_master_browser_wins) > become_domain_master_browser_wins: > Attempting to become domain master browser on workgroup MYGROUP, > subnet UNICAST_SUBNET. > [2010/06/21 16:39:34, 0] > nmbd/nmbd_become_dmb.c:351(become_domain_master_browser_wins) > become_domain_master_browser_wins: querying WINS server from IP > 192.168.1.2 for domain master browser name MYGROUP<1b> on workgroup > MYGROUP > [2010/06/21 16:39:40, 0] > nmbd/nmbd_become_dmb.c:110(become_domain_master_stage2) > * > > Samba server CARTER is now a domain master browser for workgroup > MYGROUP on subnet UNICAST_SUBNET > > * > [2010/06/21 16:39:40, 0] > nmbd/nmbd_become_dmb.c:292(become_domain_master_browser_bcast) > become_domain_master_browser_bcast: > Attempting to become domain master browser on workgroup MYGROUP on > subnet 192.168.1.2 > [2010/06/21 16:39:40, 0] > nmbd/nmbd_become_dmb.c:305(become_domain_master_browser_bcast) > become_domain_master_browser_bcast: querying subnet 192.168.1.2 for > domain master browser on workgroup MYGROUP > [2010/06/21 16:39:48, 0] > nmbd/nmbd_become_dmb.c:110(become_domain_master_stage2) > * > > Samba server CARTER is now a domain master browser for workgroup > MYGROUP on subnet 192.168.1.2 > > * > [2010/06/21 16:39:56, 0] > nmbd/nmbd_become_lmb.c:395(become_local_master_stage2) > * > > Samba name server CARTER is now a local master browser for workgroup > MYGROUP on subnet 192.168.1.2 > > * > [2010/06/21 16:39:33, 0] smbd/server.c:1073(main) > smbd version 3.4.6 started. > Copyright Andrew Tridgell and the Samba Team 1992-2009 > [2010/06/21 16:39:33, 0] printing/print_cups.c:103(cups_connect) > Unable to connect to CUPS server /var/run/cups/cups.sock:631 - No such > file or directory > [2010/06/21 16:39:33, 0] printing/print_cups.c:103(cups_connect) > Unable to connect to CUPS server /var/run/cups/cups.sock:631 - No such > file or directory > [2010/06/21 16:39:34, 0] smbd/server.c:457(smbd_open_one_socket) > smbd_open_once_socket: open_socket_in: Address already in use > [2010/06/21 16:39:34, 0] smbd/server.c:457(smbd_open_one_socket) > smbd_open_once_socket: open_socket_in: Address already in use -- alan dot mckinnon at gmail dot com
Re: [gentoo-user] Questions re swap and hibernate interaction on 8 gig machine
Walter Dnes writes: > I just got a brand new custom-built 8 gig machine. There's an outfit > in north Toronto that has MSI motherboards with PS/2 ports, so I can > keep my genuine IBM PS/2 clickety-clack-keyboard; woho. And the > integrated Intel graphics chip has *BOTH VGA AND DIGITAL OUTPUTS*! Hooray! > Anyhow, I have 8 gigs of ram on the sytem (will obviously be 64-bit > Gentoo) and I want to know how much swap I need. The general rule of > thumb is twice the ram. In this case, it would be 16 gigs. I think > that it may not need swap when up, unless I do some heavy duty stuff. I think this rule does not scale with todays amounts of system ram. If your system would need a similar amount of swap, swapping such a lot would make things really really slow. You could probably live without any swap, except for the purpose of hibernating to disk. > My main concern about a swap partition is how much I need for > hibernate-to-disk to work. Is there a rule about this, or should I > simply allocate 16 gigs out of my terabyte drive, and play it safe? The amount of swap needed is the amount RAM actually being used on your system, compressed. Add the values of the 'used' fields of Mem and Swap in your free -m output, divide by two, and that should be somewhere near the amount you need. Maybe even less if tuxonice frees caches and buffers. 4GB should be more than enough, I'd think. But hibernation also works with swap files, so there is no need to set the exact size already. And I suggest the usage of LVM, this way you can freely and very easily change the swap size as you like. I never install Linux without LVM these days, this flexibility makes things so much easier, and I do not have to care much about partition sizes. Wonko
Re: [gentoo-user] Boot hangs after install, no error
Bill Longman wrote: On 06/21/2010 01:23 PM, Dale wrote: The only thing that genkernel would add is your initrd. The kernel is exactly the same, whether you compile it with "make" or through "genkernel". Do a test and you'll see. (I'm assuming we're both talking about gentoo-sources, not vanilla-sources. Either way, they'd be the same.) Some might be confused about what happens in the steps if they haven't been down the "kernel compilation trail" more than once or twice, but for folks who just want to compile their kernel and plop it into place, along with a hands-off initrd, it's rather handy. But only if it works. When I compile my kernel, I KNOW for sure what is in there. When genkernel does one, especially on a new install, I have no idea what is in it or what is not. If something goes wrong, I don't know where to start. Is it a kernel problem or is it something else? Who knows. Then you have to go back and start from the bottom, usually the kernel, and work your way back up to find out what is broken. By "But only if it works," I assume the antecedent "it" refers to is a kernel that we're attempting to boot correctly. (In other words, you're not talking about genkernel failing to create a kernel for you. Is that correct?) If someone has trouble on an initial install, then that just means they didn't configure the kernel correctly, is what I interpret that to mean. The result of "make" and the result of "genkernel kernel" are exactly the same. If your "make menuconfig" creates an invalid .config file for you, no sort of magic is going to make its resultant kernel valid. Do you mean to say that you just grab a kernel, jump into the directory and say "make" without an mrproper and some sort of config? You do realize that genkernel has --menuconfig, --xconfig and --gconfig exactly for this purpose, don't you? What sort of things do you believe genkernel is adding to your kernel? If you use "genkernel --menuconfig --no-install kernel", you can look and see what it did. It's no different than running "make menuconfig" followed by a "make; make modules". Just look in /usr/share/genkernel at the gen_compile.sh and you'll see that it does a make. Genkernel may work for you but that doesn't mean it does for everyone else. Should I mention hal here? When someone comes for help, your looking for the failure not the successes. If it was sucessful, they wouldn't need help. Which is why I mentioned genkernel in the first place. Most times a hang after boot is due to components that were missed in the kernel build -- from where? -- from a missing or incorrectly created initrd if the required modules weren't compiled into the kernel. The easiest way that I've seen is to use genkernel and get back to work. Then later on you can find out what an initrd is and why it's needed with modules but at least you'd have a running system. No, I don't think you should mention hal because it's probably OT for a thread about a hung boot. But you should apply to yourself a similar logic you ask of me: if others can use genkernel successfully, why can't I? Bill I'm not saying you can't use it just that it doesn't always work. Thing is, when someone uses genkernel to make the kernel, when someone asks 'did you include some driver', the usual answer is 'I don't know, I used genkernel' and then nobody knows whether it is there or not. If a person builds their own kernel, they usually know if it is there and better yet how to check and make sure it is there. Also, I don't use initrd and not sure why most people need one. I don't use modules either, hence the reason I don't need initrd. Just build in the drivers and such that are needed to boot until the modules are loaded and that's it. It's not rocket science. Driver controller, file system that root uses and that's about it. I haven't used genkernel in a while. I have just seen where people have used it and it not work. Same as hal. It works for most but when it doesn't, no one can figure out why because few people know how the thing works and even fewer can figure out the config file. That's not quite as off topic as it appears. If you want to use genkernel, go for it. I just know this, when someone asks for help that may be kernel related and they use genkernel, there is very very little help I can provide. Some people here use genkernel but there are a lot that don't. There is another thread posted a day or so ago where they used genkernel, no one has been able to help them yet. Not one reply that I have seen. I want to help but with genkernel, I have no idea where to start. I'm sure it is a kernel issue but that's about it. It appears that on one else can help either. It's not like this is a small mailing list with few people on it. It's your choice. Use whatever makes you happy and gets you where you want to go. Dale :-) :-)
Re: [gentoo-user] Questions re swap and hibernate interaction on 8 gig machine
On Tue, 2010-06-22 at 00:27 +0200, Alex Schuster wrote: > Walter Dnes writes: > > > I just got a brand new custom-built 8 gig machine. There's an outfit > > in north Toronto that has MSI motherboards with PS/2 ports, so I can > > keep my genuine IBM PS/2 clickety-clack-keyboard; woho. And the > > integrated Intel graphics chip has *BOTH VGA AND DIGITAL OUTPUTS*! > > Hooray! > > > Anyhow, I have 8 gigs of ram on the sytem (will obviously be 64-bit > > Gentoo) and I want to know how much swap I need. The general rule of > > thumb is twice the ram. In this case, it would be 16 gigs. I think > > that it may not need swap when up, unless I do some heavy duty stuff. > > I think this rule does not scale with todays amounts of system ram. If > your system would need a similar amount of swap, swapping such a lot > would make things really really slow. You could probably live without > any swap, except for the purpose of hibernating to disk. > > > My main concern about a swap partition is how much I need for > > hibernate-to-disk to work. Is there a rule about this, or should I > > simply allocate 16 gigs out of my terabyte drive, and play it safe? > > The amount of swap needed is the amount RAM actually being used on your > system, compressed. Add the values of the 'used' fields of Mem and Swap > in your free -m output, divide by two, and that should be somewhere near > the amount you need. Maybe even less if tuxonice frees caches and > buffers. 4GB should be more than enough, I'd think. > > But hibernation also works with swap files, so there is no need to set > the exact size already. And I suggest the usage of LVM, this way you can > freely and very easily change the swap size as you like. I never install > Linux without LVM these days, this flexibility makes things so much > easier, and I do not have to care much about partition sizes. > > Wonko > google has lots on this - this is one with a few ideas. http://www.cyberciti.biz/tips/linux-swap-space.html Generally hibernate (I use TuxOnIce with compression) uses around half the occupied ram size - but you cant guarantee it, and if swap itself is heavily loaded it gets critical. In my case the most swap Ive ever used is 34Gb of a 145Gb swap partition (5G ram, number crunching :) but graphics editing will sometimes require up to 6gb swap in addition to ram. Keep in mind that the maximum swap available varies depending on physical architecture (I think!, cant find the original reference now, but its recapped here: http://ubuntuforums.org/showthread.php?t=606837) My experience is you can create as much swap as you want - but the cpu will only use as much as its capable of - wastage! Note that i386 can only address 2Gb swap in one partition, but you can have up to 32 swap partitions. So its a 'how long is a piece of string' type question. My 'algorithm' is something like: 1. swap is bad, get as much ram as possible 2. use a swap of 2xram 'just in case' that ram is not enough - better that than a crash or oom event - disk space is cheap! Added benefit is almost always enough space for hibernate. 3. use multiple swap partitions with each one on a different physical disk mounted with equal priority to allow striping for performance benefits. The more disks, the more swap partitions! Create them at the start of the disk as thats apparently the fastest place to put them - http://lissot.net/partition/partition-04.html 3. if swap/hibernate space looks like being a problem, hibernate to a file instead (see tuxonice, not sure if the flakey in-kernal hibernate works with a file) 4. on systems expected to occasionally need extra swap keep a swapfile around for instant use, or a script to create one on the fly 5. Tune the kernel swappable parameter to either force most everything to swap to keep ram free or the other way to make it less likely to swap if thats whats needed. The current parameter is a compromise that works in most cases, but there are uses that benefit one way or the other. Lastly, to reiterate, disk space is cheap and putting aside 16Gb for swap is a small price to pay for stability - you may only need it occasionally, but then you will REALLY need it! Have fun! BillK
[gentoo-user] Re: Boot hangs after install, no error
On 06/21/2010 03:37 PM, Dale wrote: I don't use initrd and not sure why most people need one... Good point, and I'd like to know why, too. AFAIK the major use of initrd is on installation CD's, where the maker of the install disk has no way of knowing in advance what hardware/filesystems the "installee" is using, and thus must be ready for any surprise. If this point of view is misguided or incomplete, please hit me with the cluestick, asap :) (BTW, hal is never off-topic for Dale. Don't ask why, it's ancient history now, just like hal. I don't think Dale is ancient history, though, quite yet, well, almost ;)
[gentoo-user] Questions regarding the usage of multiple locales
I've setup my Gentoo box to use en_GB as the default locale in /etc/env.d/02locale with tips from the Gentoo Localisation Guide[0]. Is it at all possible to set a locale, i.e. cy_GB to be the primary LANG parameter but if there is no .po for cy_GB or the .po is incomplete to use en_GB as a backup instead of the default en_US? So for example if gedit were fully translated into Welsh (cy_GB) I could use gedit in Welsh but emerge not being translated into Welsh would resort to en_GB instead of en_US? Many thanks / Diolch yn fawr iawn, Chris [0] http://www.gentoo.org/doc/en/guide-localization.xml -- Christopher Swift
Re: [gentoo-user] Re: Boot hangs after install, no error
walt wrote: On 06/21/2010 03:37 PM, Dale wrote: I don't use initrd and not sure why most people need one... Good point, and I'd like to know why, too. AFAIK the major use of initrd is on installation CD's, where the maker of the install disk has no way of knowing in advance what hardware/filesystems the "installee" is using, and thus must be ready for any surprise. If this point of view is misguided or incomplete, please hit me with the cluestick, asap :) (BTW, hal is never off-topic for Dale. Don't ask why, it's ancient history now, just like hal. I don't think Dale is ancient history, though, quite yet, well, almost ;) It used to be that if you used lvm or something, you needed a initrd. I'm not sure that is the case now. I think I read somewhere that you can use lvm and not use initrd since there is something in the kernel for that now. I have not seen that in the kernel tho so I may have been confusing that with something else. It does happen sometimes. Me being confused that is. lol Thing about genkernel, when a person needs help with a kernel issue, it limits the number of people that can help with the problem. Most people on here can help with kernel issues if the user compiled their own. With genkernel, you don't really know what you are getting. I have even read where people used genkernel and got two different kernels on the same machine. They should be pretty much the same since the hardware is the same. Heck, I rarely change anything when I run make oldconfig. It is more of saying no to stuff I don't have than yes. Hal is much like anything else, it "just works" for most people but others have "issues". I had serious issues with hal. It seemed nothing I did would get it to work. Funny thing is, it works with everything except xorg. Hal works great and always has for everything else. That is one reason I have also told folks doing a fresh install to try it with hal. It works with most people and if it doesn't, everyone knows how to get rid of it. You can't exactly get rid of a kernel now can we? ;-) Dale :-) :-)
Re: [gentoo-user] Questions re swap and hibernate interaction on 8 gig machine
Bill Kenworthy wrote: << SNIP >> 5. Tune the kernel swappable parameter to either force most everything to swap to keep ram free or the other way to make it less likely to swap if thats whats needed. The current parameter is a compromise that works in most cases, but there are uses that benefit one way or the other. Lastly, to reiterate, disk space is cheap and putting aside 16Gb for swap is a small price to pay for stability - you may only need it occasionally, but then you will REALLY need it! Have fun! BillK Just for notes, I have 2Gbs of ram. I have ~1Gb of swap. I don't think I have ever seen it use over 100Mbs or so of swap since I built this thing. Most of this depends on the programs you are using. If they use more memory than you have, you need more swap. If not, then maybe not. ;-) If you only want it to use swap when really needed, set it like this or something close: r...@smoker-new / # cat /proc/sys/vm/swappiness 20 r...@smoker-new / # The lower the number, the more it tries not to use swap. The higher the number, the more it will try to use swap. Since I rarely use more than 1Gb, compiling OOo may be a exception, I set mine to 20. My drives are the old IDE and are not as fast as the new SATA drives. Use echo to change that setting just in case you don't know already. Hope that little bit of info helps. This is one of those "it depends" situations. No matter what you set it up for, you will want to change it later. :-) Dale :-) :-)
Re: [gentoo-user] kernel configuration problem -- genkernel: no irq handler; manual config: root filesystem could not be mounted r/w
On Mon, Jun 21, 2010 at 02:57:58PM +0800, rocwhite168 wrote > On the other hand, I'm trying to configure the kernel manually. I > did this according to several online tutorials, but it still won't > even start up: "Root filesystem could not be mounted read/write." Can > anyone please have a look at my configuration? Thank you very much! One more helpful piece of info; boot from the install CD (or a Knoppix Live CD for that matter) and mount a USB stick and run... lsmod > info.txt ...and post the contents of info.txt. That will give us an idea of what modules are used when the system successfully boots. Also, do you have AHCI support enabled? -- Walter Dnes
[gentoo-user] Re: kernel configuration problem -- genkernel: no irq handler; manual config: root filesystem could not be mounted r/w
Kaddeh gmail.com> writes: > > it would help to also put your /etc/fstaband let us know what FS you have > root setup asin addition to your /boot/grub/grub.conf > 2010/6/20 rocwhite168 163.com> > I used genkernel to configure the kernel. It complained every time at > shutdown that > > "Code: Bad EIP value. > > ... > do_IRQ: 0.43 No irq handler for vector (irq -1)". > I had to press the power button to shut down the machine. How should I solve this problem? > > On the other hand, I'm trying to configure the kernel manually. I did this according to several online tutorials, but it still won't even start up: "Root filesystem could not be mounted read/write." Can anyone please have a look at my configuration? Thank you very much! > > Roc fstab -- /dev/sda1 /boot ext4noauto,noatime 1 2 /dev/sda5 / ext4noatime 0 1 /dev/sda3 /usrext4noatime 0 2 /dev/sda4 /home ext4noatime 0 2 /dev/sda7 /tmpext4noatime 0 0 /dev/sda6 noneswapsw 0 0 shm /dev/shmtmpfs nodev,nosuid,noexec 0 0 grub.conf -- default 0 timeout 3 #splashimage=(hd0,0)/boot/grub/splash.xpm.gz title Gentoo Linux 2.6.31-r10 root (hd0,0) kernel /boot/kernel-2.6.31-gentoo-r10 root=/dev/sda5 title Gentoo Linux 2.6.31-r10 -- genkernel root (hd0,0) kernel /boot/kernel-genkernel-x86-2.6.31-gentoo-r10 root=/dev/ram0 init=/linuxrc ramdisk=8192 real_root=/dev/sda5 initrd /boot/initramfs-genkernel-x86-2.6.31-gentoo-r10 The first entry is for the manual configuration, while the second one is for the kernel generated by genkernel. Thanks, Roc
[gentoo-user] Re: kernel configuration problem -- genkernel: no irq handler; manual config: root filesystem could not be mounted r/w
Walter Dnes waltdnes.org> writes: > > On Mon, Jun 21, 2010 at 02:57:58PM +0800, rocwhite168 wrote > > > On the other hand, I'm trying to configure the kernel manually. I > > did this according to several online tutorials, but it still won't > > even start up: "Root filesystem could not be mounted read/write." Can > > anyone please have a look at my configuration? Thank you very much! > > One more helpful piece of info; boot from the install CD (or a Knoppix > Live CD for that matter) and mount a USB stick and run... > > lsmod > info.txt > > ...and post the contents of info.txt. That will give us an idea of what > modules are used when the system successfully boots. Also, do you have > AHCI support enabled? > lsmod from a Knoppix Live CD Module Size Used by ext4 173499 4 jbd2 35345 1 ext4 ipv6 180370 24 ppdev 3946 0 lp 5936 0 uinput 4681 1 dcdbas 3844 0 8250_pnp3292 0 8250 14440 1 8250_pnp parport_pc 24136 1 parport21974 3 ppdev,lp,parport_pc serio_raw 2926 0 serial_core11779 1 8250 snd_intel8x0 19802 1 i2c_i8016149 0 snd_ac97_codec 77631 1 snd_intel8x0 ac97_bus 702 1 snd_ac97_codec tg386894 0 aufs 107889 1 cloop 11854 1 lsmod from my manually configured kernel -- Module Size Used by tg394300 0 libphy 20600 1 tg3 e1000 102212 0 fuse 51712 0 jfs 145856 0 raid10 19112 0 raid45645132 0 async_memcpy1748 1 raid456 async_xor 2844 1 raid456 xor13952 2 raid456,async_xor async_tx2936 3 raid456,async_memcpy,async_xor raid6_pq 80412 1 raid456 raid1 18432 0 raid0 6876 0 dm_bbr 9476 0 dm_snapshot21668 0 dm_crypt 11020 0 dm_mirror 12488 0 dm_region_hash 10280 1 dm_mirror dm_log 8424 2 dm_mirror,dm_region_hash dm_mod 56208 5 dm_bbr,dm_snapshot,dm_crypt,dm_mirror,dm_log scsi_wait_scan 1020 0 sbp2 19252 0 ohci1394 25428 0 ieee1394 73612 2 sbp2,ohci1394 sl811_hcd 9100 0 usbhid 22436 0 ohci_hcd 19692 0 uhci_hcd 18732 0 usb_storage60504 0 ehci_hcd 30380 0 usbcore 118068 7 sl811_hcd,usbhid,ohci_hcd,uhci_hcd,usb_storage,ehci_hcd aic94xx61996 0 libsas 38148 1 aic94xx lpfc 317620 0 qla2xxx 192236 0 megaraid_sas 27500 0 megaraid_mbox 26148 0 megaraid_mm 7968 1 megaraid_mbox megaraid 36556 0 aacraid59084 0 sx812952 0 DAC960 59284 0 cciss 35056 0 3w_9xxx27440 0 3w_21188 0 mptsas 43196 0 scsi_transport_sas 24016 3 aic94xx,libsas,mptsas mptfc 14052 0 scsi_transport_fc 38224 3 lpfc,qla2xxx,mptfc scsi_tgt 10740 1 scsi_transport_fc mptspi 15168 0 mptscsih 28432 3 mptsas,mptfc,mptspi mptbase77084 4 mptsas,mptfc,mptspi,mptscsih atp870u25092 0 dc395x 28308 0 sim710 3040 0 53c700 22968 1 sim710 qla128019612 0 dmx3191d9388 0 sym53c8xx 64348 0 qlogicfas4086256 0 gdth 75712 0 aha1740 6232 0 advansys 50808 0 initio 15052 0 BusLogic 20164 0 arcmsr 19220 0 aic7xxx 105888 0 aic79xx 117228 0 scsi_transport_spi 19936 6 mptspi,53c700,dmx3191d,sym53c8xx,aic7xxx,aic79xx sg 24248 0 pdc_adma6100 0 sata_inic162x 8172 0 sata_mv24156 0 ata_piix 22096 5 ahci 31644 0 sata_qstor 6200 0 sata_vsc4848 0 sata_uli3756 0 sata_sis4812 0 sata_sx48804 0 sata_nv19824 0 sata_via8196 0 sata_svw4764 0 sata_sil24 11780 0 sata_sil8256 0 sata_promise9760 0 pata_sl82c105 4108 0 pata_cs5535 3116
[gentoo-user] Re: kernel configuration problem -- genkernel: no irq handler; manual config: root filesystem could not be mounted r/w
Walter Dnes waltdnes.org> writes: > > On Mon, Jun 21, 2010 at 02:57:58PM +0800, rocwhite168 wrote > > > On the other hand, I'm trying to configure the kernel manually. I > > did this according to several online tutorials, but it still won't > > even start up: "Root filesystem could not be mounted read/write." Can > > anyone please have a look at my configuration? Thank you very much! > > One more helpful piece of info; boot from the install CD (or a Knoppix > Live CD for that matter) and mount a USB stick and run... > > lsmod > info.txt > > ...and post the contents of info.txt. That will give us an idea of what > modules are used when the system successfully boots. Also, do you have > AHCI support enabled? > lsmod from a Knoppix Live CD -- Module Size Used by ext4 173499 4 jbd2 35345 1 ext4 ipv6 180370 24 ppdev 3946 0 lp 5936 0 uinput 4681 1 dcdbas 3844 0 8250_pnp3292 0 8250 14440 1 8250_pnp parport_pc 24136 1 parport21974 3 ppdev,lp,parport_pc serio_raw 2926 0 serial_core11779 1 8250 snd_intel8x0 19802 1 i2c_i8016149 0 snd_ac97_codec 77631 1 snd_intel8x0 ac97_bus 702 1 snd_ac97_codec tg386894 0 aufs 107889 1 cloop 11854 1 lsmod from my manual configured kernel -- Module Size Used by ipv6 202124 24 snd_intel8x0 26140 0 snd_ac97_codec 90036 1 snd_intel8x0 ac97_bus1408 1 snd_ac97_codec snd_pcm58024 2 snd_intel8x0,snd_ac97_codec snd_timer 17124 1 snd_pcm snd47768 4 snd_intel8x0,snd_ac97_codec,snd_pcm,snd_timer soundcore 6224 1 snd snd_page_alloc 7768 2 snd_intel8x0,snd_pcm ata_generic 4308 0 rtc 9072 0 pata_acpi 3544 0 processor 32616 0 thermal12532 0 button 5012 0 dcdbas 6592 0 tg394300 0 libphy 20600 1 tg3 e1000 102212 0 fuse 51712 0 jfs 145856 0 raid10 19112 0 raid45645132 0 async_memcpy1748 1 raid456 async_xor 2844 1 raid456 xor13952 2 raid456,async_xor async_tx2936 3 raid456,async_memcpy,async_xor raid6_pq 80412 1 raid456 raid1 18432 0 raid0 6876 0 dm_bbr 9476 0 dm_snapshot21668 0 dm_crypt 11020 0 dm_mirror 12488 0 dm_region_hash 10280 1 dm_mirror dm_log 8424 2 dm_mirror,dm_region_hash dm_mod 56208 5 dm_bbr,dm_snapshot,dm_crypt,dm_mirror,dm_log scsi_wait_scan 1020 0 sbp2 19252 0 ohci1394 25428 0 ieee1394 73612 2 sbp2,ohci1394 sl811_hcd 9100 0 usbhid 22436 0 ohci_hcd 19692 0 uhci_hcd 18732 0 usb_storage60504 0 ehci_hcd 30380 0 usbcore 118068 7 sl811_hcd,usbhid,ohci_hcd,uhci_hcd,usb_storage,ehci_hcd aic94xx61996 0 libsas 38148 1 aic94xx lpfc 317620 0 qla2xxx 192236 0 megaraid_sas 27500 0 megaraid_mbox 26148 0 megaraid_mm 7968 1 megaraid_mbox megaraid 36556 0 aacraid59084 0 sx812952 0 DAC960 59284 0 cciss 35056 0 3w_9xxx27440 0 3w_21188 0 mptsas 43196 0 scsi_transport_sas 24016 3 aic94xx,libsas,mptsas mptfc 14052 0 scsi_transport_fc 38224 3 lpfc,qla2xxx,mptfc scsi_tgt 10740 1 scsi_transport_fc mptspi 15168 0 mptscsih 28432 3 mptsas,mptfc,mptspi mptbase77084 4 mptsas,mptfc,mptspi,mptscsih atp870u25092 0 dc395x 28308 0 sim710 3040 0 53c700 22968 1 sim710 qla128019612 0 dmx3191d9388 0 sym53c8xx 64348 0 qlogicfas4086256 0 gdth 75712 0 aha1740 6232 0 advansys 50808 0 initio 15052 0 BusLogic 20164 0 arcmsr 19220 0 aic7xxx 105888 0 aic79xx 117228 0 scsi_trans
