Re: [gentoo-user] Re: [OT] LENOVO Z510 + Dual Boot + Gentoo == True ?
On Mar 15, 2014, at 19:17, »Q« boxc...@gmx.net wrote: On Mon, 10 Mar 2014 13:33:20 + Mick michaelkintz...@gmail.com wrote: On Saturday 08 Mar 2014 20:22:12 »Q« wrote: On Sat, 08 Mar 2014 08:23:21 +0100 grub booted Gentoo just fine, but Windows booting failed, something about not finding partitions or files. Instead of troubleshooting that, I disabled os probing for grub (GRUB_DISABLE_OS_PROBER=true in /etc/default/grub) and added Windows via /etc/grub.d/40_custom , like so: If you moved the MSWindows OS or boot partitions then the UUIDs would have changed. I moved the OS partition, and it's UUID did indeed change. I have swaped the hard drive from my dual boot box and ran into the same problem trying get windows 7 to boot. As you also quite fast realice by reading different forums that changing windows boot parameters is a quite big hassle. I would not go that way! You have another simpler solution. Change the hard disk device ID to the same value as the old disk. It is written on MBR. Change the UUID of the windows partition to the same as on the old partition. UUID on NTFS partition is written at the beginning of the partition at 0x48-4F. So by changing 2x16 bytes of data your machine should boot again correctly. Also if you grub is not on the same physical disk as windows then you need trick windows by changing the order with grub before booting (see map command) You'll need to edit the MSWindows boot menu (in the MSWindows boot partition) and change their entrie(s) accordingly. If somebody can post a link to a recipe for doing that, I'd appreciate it. I don't understand the Windows boot stuff.
Re: [gentoo-user] Re: [OT] LENOVO Z510 + Dual Boot + Gentoo == True ?
On Sunday 16 Mar 2014 09:07:49 Matti Nykyri wrote: Change the hard disk device ID to the same value as the old disk. It is written on MBR. Change the UUID of the windows partition to the same as on the old partition. UUID on NTFS partition is written at the beginning of the partition at 0x48-4F. Can you give more detail please? How would you change disk and partition UUIDs? -- Regards, Mick signature.asc Description: This is a digitally signed message part.
Re: [gentoo-user] Re: [OT] LENOVO Z510 + Dual Boot + Gentoo == True ?
On Mar 16, 2014, at 12:38, Mick michaelkintz...@gmail.com wrote: On Sunday 16 Mar 2014 09:07:49 Matti Nykyri wrote: Change the hard disk device ID to the same value as the old disk. It is written on MBR. Change the UUID of the windows partition to the same as on the old partition. UUID on NTFS partition is written at the beginning of the partition at 0x48-4F. Can you give more detail please? How would you change disk and partition UUIDs? -- Regards, Mick Well when you purchase a new blank disk it is full with null's. When you first time open that drive with for example with fdisk it complains about incorrect mbr. If you in that situation print the partition table you will see that the device id is null. When you create a partition these errors will be corrected by write. Fdisk creates a new device id from random data. It is then written to the mbr. Just explore the disk with hexedit and you'll find the device id. Just remember endianess. By the same way a UUID is created when you format a new NTFS partition. It is also just random data written to the disk. It can easily edited with hexedit. At least my win7 booted normally when i moved it from a disk to another and fixed the UUID's of the new drive. Windows didn't notice anything. After i switched the motherboard of the machine then windows required a new activation. Actually if you copy the windows partition with dd the uuid of the NTFS partition will not change. If you also copy the beginning of the old disk to a new one it will copy the device id to the new disk. Instead if you make a new partition table the device id will change. There is nothing magical with partitioning and moving data on disk or to another disk. You can completely wipe mbr and partition table and then write a new partition table with partitions pointing to the beginning of your data and all your data will be intact. Just experiment with hexedit. I can give you correct addresses when i'm back at home tomorrow. Or just google your self, if you are unable to find it with hexedit. -- Matti
[gentoo-user] dracut: mount: special device /dev/disk/by-label/usr does not exist
Howdy, I got this when rebooting after we had a power outage. I have a UPS so I was able to perform a regular shutdown. [2.567061] hub 8-1:1.0: state 7 ports 4 chg evt [2.567069] hub 8-1:1.0: hub_suspend [2.579644] usb 8-1: usb auto-suspend, wakeup 1 [2.591677] hub 8-0:1.0: hub_suspend [2.591682] usb usb8: bus auto-suspend, wakeup 1 [3.362374] dracut: root has been mounted 29 times without being checked, check forced. [3.363014] dracut: root: 28323/1525920 files (0.4% non-contiguous), 580665/6102684 blocks [3.364957] dracut: Mounting /dev/disk/by-uuid/888352dd-9c91-4a9f-9595-cd0e74b74ee7 with -o defaults,ro [3.474224] EXT4-fs (sda6): mounted filesystem with ordered data mode. Opts: (null) [3.522894] dracut: Mounted root filesystem /dev/sda6 [3.568630] dracut: Mounting /usr with -o defaults,ro [3.600028] dracut: mount: special device /dev/disk/by-label/usr does not exist [3.601749] dracut Warning: Mounting /usr to /sysroot/usr failed [3.602452] dracut Warning: *** Dropping you to a shell; the system will continue [3.603419] dracut Warning: *** when you leave the shell. [3.604892] dracut Warning: [3.849621] blkid (2070) used greatest stack depth: 4576 bytes left + '[' -f /run/initramfs/init.log ']' root@fireball / # It seems to me that the / file system needed to be checked. For that reason, it couldn't mount /usr after the check was performed. That's my thinking on this. Anyone think otherwise? Is this a one off event or should I be concerned about this? Is there some way to avoid this in the future without disabling file system check for /? Another related LVM question. I have some partitions on LVM. If I moved the drives to another system, would the new LVMs be found on the new system or is there some magic involved to find and get them mounted? Example. My /home is on its own LVM partition. If I moved the drive that has that on it, would the new system see it or would I have to do something to make it see it? I suspect and wouldn't want it to mount automatically. I'd just want to be able to see it and mount it if needed. Sort of a question I have always wondered about. Thanks much. Dale :-) :-) P. S. For those who recall my brother having cancer. He is close to the end of his treatments. Lost a LOT of weight but hanging in there.
Re: [gentoo-user] dracut: mount: special device /dev/disk/by-label/usr does not exist
2014-03-16 8:24 GMT-06:00 Dale rdalek1...@gmail.com: Howdy, I got this when rebooting after we had a power outage. I have a UPS so I was able to perform a regular shutdown. [2.567061] hub 8-1:1.0: state 7 ports 4 chg evt [2.567069] hub 8-1:1.0: hub_suspend [2.579644] usb 8-1: usb auto-suspend, wakeup 1 [2.591677] hub 8-0:1.0: hub_suspend [2.591682] usb usb8: bus auto-suspend, wakeup 1 [3.362374] dracut: root has been mounted 29 times without being checked, check forced. [3.363014] dracut: root: 28323/1525920 files (0.4% non-contiguous), 580665/6102684 blocks [3.364957] dracut: Mounting /dev/disk/by-uuid/888352dd-9c91-4a9f-9595-cd0e74b74ee7 with -o defaults,ro [3.474224] EXT4-fs (sda6): mounted filesystem with ordered data mode. Opts: (null) [3.522894] dracut: Mounted root filesystem /dev/sda6 [3.568630] dracut: Mounting /usr with -o defaults,ro [3.600028] dracut: mount: special device /dev/disk/by-label/usr does not exist Seems like the block device for /usr couldn't be found by it's label, root partition seems fine after check and properly mounted, so I'd say it has nothing to do. [3.601749] dracut Warning: Mounting /usr to /sysroot/usr failed [3.602452] dracut Warning: *** Dropping you to a shell; the system will continue [3.603419] dracut Warning: *** when you leave the shell. [3.604892] dracut Warning: [3.849621] blkid (2070) used greatest stack depth: 4576 bytes left + '[' -f /run/initramfs/init.log ']' root@fireball / # It seems to me that the / file system needed to be checked. For that reason, it couldn't mount /usr after the check was performed. That's my thinking on this. Anyone think otherwise? Is this a one off event or should I be concerned about this? Is the block device corresponding to /usr available under another directory in /dev? if not something wrong might be going on with that block device. I personally prefer using UUIDs for finding partitions at boot, they are more fail-proof. Is there some way to avoid this in the future without disabling file system check for /? Again, maybe UUIDs. Another related LVM question. I have some partitions on LVM. If I moved the drives to another system, would the new LVMs be found on the new system or is there some magic involved to find and get them mounted? Example. My /home is on its own LVM partition. If I moved the drive that has that on it, would the new system see it or would I have to do something to make it see it? I suspect and wouldn't want it to mount automatically. I'd just want to be able to see it and mount it if needed. Sort of a question I have always wondered about. On my experience as long, as udev and lvm are running on the receiving system, they should be found and placed for access under /dev, not mounted automatically. if for some reason it doesnt happen, its easy to do a 'pvscan' to see if the physical volume is recognized, and if it is, 'vgchange -ay volume_group_name ' activates all LVs. Thanks much. Dale :-) :-) P. S. For those who recall my brother having cancer. He is close to the end of his treatments. Lost a LOT of weight but hanging in there.
Re: [gentoo-user] dracut: mount: special device /dev/disk/by-label/usr does not exist
Jc García wrote: 2014-03-16 8:24 GMT-06:00 Dale rdalek1...@gmail.com mailto:rdalek1...@gmail.com: Howdy, I got this when rebooting after we had a power outage. I have a UPS so I was able to perform a regular shutdown. [2.567061] hub 8-1:1.0: state 7 ports 4 chg evt [2.567069] hub 8-1:1.0: hub_suspend [2.579644] usb 8-1: usb auto-suspend, wakeup 1 [2.591677] hub 8-0:1.0: hub_suspend [2.591682] usb usb8: bus auto-suspend, wakeup 1 [3.362374] dracut: root has been mounted 29 times without being checked, check forced. [3.363014] dracut: root: 28323/1525920 files (0.4% non-contiguous), 580665/6102684 blocks [3.364957] dracut: Mounting /dev/disk/by-uuid/888352dd-9c91-4a9f-9595-cd0e74b74ee7 with -o defaults,ro [3.474224] EXT4-fs (sda6): mounted filesystem with ordered data mode. Opts: (null) [3.522894] dracut: Mounted root filesystem /dev/sda6 [3.568630] dracut: Mounting /usr with -o defaults,ro [3.600028] dracut: mount: special device /dev/disk/by-label/usr does not exist Seems like the block device for /usr couldn't be found by it's label, root partition seems fine after check and properly mounted, so I'd say it has nothing to do. [3.601749] dracut Warning: Mounting /usr to /sysroot/usr failed [3.602452] dracut Warning: *** Dropping you to a shell; the system will continue [3.603419] dracut Warning: *** when you leave the shell. [3.604892] dracut Warning: [3.849621] blkid (2070) used greatest stack depth: 4576 bytes left + '[' -f /run/initramfs/init.log ']' root@fireball / # It seems to me that the / file system needed to be checked. For that reason, it couldn't mount /usr after the check was performed. That's my thinking on this. Anyone think otherwise? Is this a one off event or should I be concerned about this? Is the block device corresponding to /usr available under another directory in /dev? if not something wrong might be going on with that block device. I personally prefer using UUIDs for finding partitions at boot, they are more fail-proof. Well, after that, it booted fine. I forgot to mention that it did boot without me doing anything but letting it proceed. I *think* I typed exit or something to get it to keep going. Is there some way to avoid this in the future without disabling file system check for /? Again, maybe UUIDs. I tried that once and grub didn't like it. May need to see if things have improved in that area since. Another related LVM question. I have some partitions on LVM. If I moved the drives to another system, would the new LVMs be found on the new system or is there some magic involved to find and get them mounted? Example. My /home is on its own LVM partition. If I moved the drive that has that on it, would the new system see it or would I have to do something to make it see it? I suspect and wouldn't want it to mount automatically. I'd just want to be able to see it and mount it if needed. Sort of a question I have always wondered about. On my experience as long, as udev and lvm are running on the receiving system, they should be found and placed for access under /dev, not mounted automatically. if for some reason it doesnt happen, its easy to do a 'pvscan' to see if the physical volume is recognized, and if it is, 'vgchange -ay volume_group_name ' activates all LVs. Thanks much. Dale :-) :-) That's my thinking to but I have never had the chance to test it. I figured the info is stored on the drive and moves wherever the drives goes as long as LVM is running. Thanks much. Dale :-) :-) -- I am only responsible for what I said ... Not for what you understood or how you interpreted my words!
Re: [gentoo-user] dracut: mount: special device /dev/disk/by-label/usr does not exist
2014-03-16 9:40 GMT-06:00 Dale rdalek1...@gmail.com: Jc García wrote: Is there some way to avoid this in the future without disabling file system check for /? Again, maybe UUIDs. I tried that once and grub didn't like it. May need to see if things have improved in that area since. For grub2 works fine for me, but the issue here was after grub had done its job, is the way dracut or more accurately the initrd, was trying to mount partitions before calling init. Not sure at all, but the fstab inside the initrd would have had something to do, so maybe, setting UUIDs in fstab and reconstructing the initrd, might prevent the issue from happening again. Another related LVM question. I have some partitions on LVM. If I moved the drives to another system, would the new LVMs be found on the new system or is there some magic involved to find and get them mounted? Example. My /home is on its own LVM partition. If I moved the drive that has that on it, would the new system see it or would I have to do something to make it see it? I suspect and wouldn't want it to mount automatically. I'd just want to be able to see it and mount it if needed. Sort of a question I have always wondered about. On my experience as long, as udev and lvm are running on the receiving system, they should be found and placed for access under /dev, not mounted automatically. if for some reason it doesnt happen, its easy to do a 'pvscan' to see if the physical volume is recognized, and if it is, 'vgchange -ay volume_group_name ' activates all LVs. Thanks much. Dale :-) :-) That's my thinking to but I have never had the chance to test it. I figured the info is stored on the drive and moves wherever the drives goes as long as LVM is running. I think it's metadata stored in the PV. Thanks much. Dale :-) :-) -- I am only responsible for what I said ... Not for what you understood or how you interpreted my words!
Re: [gentoo-user] dracut: mount: special device /dev/disk/by-label/usr does not exist
On Sun, 16 Mar 2014 09:05:01 -0600, Jc García wrote: Another related LVM question. I have some partitions on LVM. If I moved the drives to another system, would the new LVMs be found on the new system or is there some magic involved to find and get them mounted? Example. My /home is on its own LVM partition. If I moved the drive that has that on it, would the new system see it or would I have to do something to make it see it? I suspect and wouldn't want it to mount automatically. I'd just want to be able to see it and mount it if needed. Sort of a question I have always wondered about. On my experience as long, as udev and lvm are running on the receiving system, they should be found and placed for access under /dev, not mounted automatically. Unless there is already a VG on the other system with the same name. LVM doesn't handle VG name clashes, yet some distros still give them generic names. -- Neil Bothwick ... I just forgot to increment the counter, Tom said, nonplussed. signature.asc Description: PGP signature
[gentoo-user] Where is /etc/conf.d/net.example?
Where is the proper place to specify the gentoo network configuration nowadays? I do not have a file called /etc/conf.d/net.example on my hard drive. That surprised me. The handbook talks all about eth0, but my machine does not have a eth0. It has eno1. Perhaps the handbook is not up to date? I'm using wicd now but I want to ditch wicd and replace it with the generally accepted correct gentoo way. Thank you, Chris
Re: [gentoo-user] Where is /etc/conf.d/net.example?
/usr/share/doc/netifrc-0.1/net.example.bz2 On Sun, Mar 16, 2014 at 8:27 PM, Chris Stankevitz chrisstankev...@gmail.com wrote: Where is the proper place to specify the gentoo network configuration nowadays? I do not have a file called /etc/conf.d/net.example on my hard drive. That surprised me. The handbook talks all about eth0, but my machine does not have a eth0. It has eno1. Perhaps the handbook is not up to date? I'm using wicd now but I want to ditch wicd and replace it with the generally accepted correct gentoo way. Thank you, Chris
Re: [gentoo-user] Where is /etc/conf.d/net.example?
On 03/16/2014 08:27 PM, Chris Stankevitz wrote: Where is the proper place to specify the gentoo network configuration nowadays? I do not have a file called /etc/conf.d/net.example on my hard drive. That surprised me. The handbook talks all about eth0, but my machine does not have a eth0. It has eno1. Perhaps the handbook is not up to date? I'm using wicd now but I want to ditch wicd and replace it with the generally accepted correct gentoo way. Thank you, Chris As far as eth0 goes, the handbook does talk about network interface names other than eth0. See below for details. http://www.gentoo.org/doc/en/handbook/handbook-x86.xml?style=printablefull=1#book_part1_chap3 Automatically Start Networking at Boot To have your network interfaces activated at boot, you need to add them to the default runlevel. Code Listing 2.7: Adding net.eth0 to the default runlevel # cd /etc/init.d # ln -s net.lo net.eth0 # rc-update add net.eth0 default If you have several network interfaces, you need to create the appropriate net.* files just like you did with net.eth0. If you later find out the assumption about the network interface name (which we currently document as eth0) was wrong, then 1. update the /etc/conf.d/net file with the correct interface name (like enp3s0 instead of eth0), 2. create new symbolic link (like /etc/init.d/net.enp3s0), 3. remove the old symbolic link (rm /etc/init.d/net.eth0), 4. add the new one to the default runlevel, and 5. remove the old one using rc-update del net.eth0 default.
Re: [gentoo-user] Where is /etc/conf.d/net.example?
On 16/03/2014 20:27, Chris Stankevitz wrote: [snip] I'm using wicd now but I want to ditch wicd and replace it with the generally accepted correct gentoo way. Thank you, Chris There is no generally accepted correct gentoo way, there is only whatever you feel like using. You have various choices - an orthodox network manager like wicd or nm - a minimal network manager like connman - /etc/init.d/net* scripts supplied by OpenRc - no manager, do it manually OpenRc is the default for no real reason other than it is and has been for some time. It can also be gotten to work in every case known to man, the same can't be said for the other options. If you want to ditch wicd because you think it's not the supported or approved way, you might want to revisit that choice -- Alan McKinnon alan.mckin...@gmail.com
Re: [gentoo-user] Where is /etc/conf.d/net.example?
On Sunday 16 Mar 2014 18:27:45 Chris Stankevitz wrote: Where is the proper place to specify the gentoo network configuration nowadays? I do not have a file called /etc/conf.d/net.example on my hard drive. That surprised me. There used to be a /etc/conf.d/net.example, but that stopped some way back since sys-apps/openrc-0.12.4 I seem to recall. I never understood why. Since then every new openrc version updates its net.example file, but instead of doing so in /etc/conf.d/, it does it in /usr/share/doc/netifrc*/ BTW, this is the gentoo way only insofar that gentoo happens to use openrc. If wicd, networkmanager, connman suit your needs, there's no obligation to use openrc's scripts. -- Regards, Mick signature.asc Description: This is a digitally signed message part.
Re: [gentoo-user] dracut: mount: special device /dev/disk/by-label/usr does not exist
Jc García wrote: 2014-03-16 9:40 GMT-06:00 Dale rdalek1...@gmail.com mailto:rdalek1...@gmail.com: Jc García wrote: Is there some way to avoid this in the future without disabling file system check for /? Again, maybe UUIDs. I tried that once and grub didn't like it. May need to see if things have improved in that area since. For grub2 works fine for me, but the issue here was after grub had done its job, is the way dracut or more accurately the initrd, was trying to mount partitions before calling init. Not sure at all, but the fstab inside the initrd would have had something to do, so maybe, setting UUIDs in fstab and reconstructing the initrd, might prevent the issue from happening again. I can't recall if I was on the old grub or the new grub. I switched to the new grub a while back but can't recall which came first. H. On another note, we had another power fail and when I started up again, it didn't complain about that problem. Of course it didn't like the shutdown since I wasn't here to do it and forgot to set it to start the UPS driver stuff. It was the same as a power plug pull. Anyway, it seems that when it has to fsck the / file system, it can't pick up from there the same as it does when it doesn't need to do a fsck on /. Bug maybe??? Dale :-) :-) -- I am only responsible for what I said ... Not for what you understood or how you interpreted my words!
Re: [gentoo-user] dracut: mount: special device /dev/disk/by-label/usr does not exist
Neil Bothwick wrote: On Sun, 16 Mar 2014 09:05:01 -0600, Jc García wrote: Another related LVM question. I have some partitions on LVM. If I moved the drives to another system, would the new LVMs be found on the new system or is there some magic involved to find and get them mounted? Example. My /home is on its own LVM partition. If I moved the drive that has that on it, would the new system see it or would I have to do something to make it see it? I suspect and wouldn't want it to mount automatically. I'd just want to be able to see it and mount it if needed. Sort of a question I have always wondered about. On my experience as long, as udev and lvm are running on the receiving system, they should be found and placed for access under /dev, not mounted automatically. Unless there is already a VG on the other system with the same name. LVM doesn't handle VG name clashes, yet some distros still give them generic names. This is good to know. It seems this would work like I was thinking. If this rig were to die or something, I could unplug the drive with /home from this system, hook up to my spare rig and go from there. I may have to mount it manually at first but hey, at least the data is still there. Neato. Thanks. Dale :-) :-) -- I am only responsible for what I said ... Not for what you understood or how you interpreted my words!
[gentoo-user] Re: Where is /etc/conf.d/net.example?
On Sun, 16 Mar 2014 22:15:59 +0200, Alan McKinnon alan.mckin...@gmail.com wrote: You have various choices - an orthodox network manager like wicd or nm - a minimal network manager like connman - /etc/init.d/net* scripts supplied by OpenRc - no manager, do it manually Why doesn't anyone ever mention using dhcpcd for managing connections? It and its accompanying openrc init script are installed on almost every gentoo box anyway. For simple setups it should Just Work(TM) out-of-box. -- eroen signature.asc Description: PGP signature
