On 10/28/2014 11:39 PM, Ruben Safir wrote:

You have to be KIDDING ME to even ask this question

That is totally incomprehensible and doesn't tell me any useful information.... list where the hell sda2 is mounted.

On my system, *everything* except /dev/disk/by-label/BOOT is mounted on /dev/sda3. /dev/sda3 is the LVM. On the lvm /home is mounted on /dev/mapper/vgmain-vHome That seems very meaningful to me :-) -- All of the great leaders have had one characteristic in common: it was the willingness to confront unequivocally the major anxiety of their people in their time. This, and not much else, is the essence of leadership. - John Kenneth Galbraith, U.S. economist ìThe Age of Uncertaintyî -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

Le 29/10/2014 08:26, Carlos E. R. a écrit :

On 2014-10-29 04:39, Ruben Safir wrote:

...

That is totally incomprehensible and doesn't tell me any useful information.... list where the hell sda2 is mounted.

You must be daft.

Why would I care where is sda2 mounted? Why would I care if it is the second or the third partition in the first or the fourth disk? I just want the partition that I reserve for home to be mounted on /home and not on /usr.

may be because you need to use a rescue system any time? with several linux on the same drive it's difficult to guess who is where jdd -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

Le 29/10/2014 09:39, Carlos E. R. a écrit :

On 2014-10-29 08:50, jdd wrote:

...

Le 29/10/2014 08:26, Carlos E. R. a écrit :

...

...

may be because you need to use a rescue system any time? with several linux on the same drive it's difficult to guess who is where

I don't say it have to be used always, but there are use cases. When possible, use df to know what sdx is used jdd -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

Hello, On Wed, 29 Oct 2014, Carlos E. R. wrote:

On 2014-10-29 08:50, jdd wrote:

...

may be because you need to use a rescue system any time? with several linux on the same drive it's difficult to guess who is where

I do, but that is not the reason. [..] What is sda one boot, is sdb on the next, and sda becomes one of the externals instead.

That, in my experience, depends on whether the BIOS/UEFI manipulates what is what. I have an external docking station (USB/eSATA, doesn't matter for this). Apart from that, I've got 6 SATA on the AMD SB710 southbridge, 2 SATA on a Marvel 9128 and 2 eSATA on a JMicronJMB362 (both via PCIe on the AMD 770 northbridge) and 2 USB3.0 on a NEC also via PCIe on the AMD770. When I boot from it via grub, all is as usual, with hd0 in grub being the /dev/sda being SATA0 on the SB710. When I change via the BIOS-boot-menu to one of the eSATA or USB ports, that device becomes hd0 and /dev/sda (and IIRC the other port on that chip becomes hd1 and /dev/sdb), regardless of which chip I switch to (Marvel, JMB or the NEC USB one). IIRC my old MoBo/BIOS did not do that, it always presented the devices in the same order.

On some machines the names like sda remain stable, but on many they don't, and can not be made stable. They gave up on the attempts, and invented that other system.

Well, depends on how you boot and select your boot-device ;)

Another reason for using labels or uuids, is that not only you can move around the disks, but that you can move around the partitions! You can add new partitions to the disk, or remove, resize or join some, and both labels and uuids remain untouched, so that booting works the same, and fstab doesn't need to be edited. The partition numbers change, but the names do not.

Yes. Actually, I just recently swapped out a LABEL=FOO (just one partition on that disk) with a new drive. Temporarily named the new partition FOO_NEW, moved the data (or copied? CNR), removed the old drive, put in the new one, ran 'e2label /dev/sdXy FOO', et voila. With stable /dev/sd*, it would be a bit easier (BTDT): just pop in the new drive externally, copy data, switch drives (i.e. remove the old drive on the SATA0-MoBo-Connector, connect the new drive, done (as the SATA0 Connector always got /dev/sda. At least since ahci was compiled in the kernel ;) If modules are involved, you gotta take care of a stable sequence, which is easy enough using the former above/below feature of modprobe. I wrote a little translator script for that[1]. Some samples from my fstab: ### part-by-disk-id+part (stable): /dev/disk/by-id/ata-SAMSUNG_HD204UI_[..]-part1 swap swap defaults,pri=42 0 0 ### part-by-label (stable): LABEL=ROOT_SSD_1 / ext4 acl,user_xattr,noatime,nodiratime,errors=remount-ro 1 1 LABEL=R_HOME /home ext3,ext2 acl,user_xattr,strictatime,errors=remount-ro 1 0 ### image mount (stable): /data/spool/news_reiserfs.img /data/spool/news \ reiserfs loop,acl,user_xattr,strictatime,barrier=flush 0 0 ### tmpfs mount (stable): tmpfs /data/build tmpfs nosuid,size=75% 0 0 #### udev-symlink mounts of /dev/sr0 /dev/cdrom /cdrom auto noauto,ro,user,users 0 0 /dev/dvd /dvd auto noauto,ro,user,users 0 0 /dev/dvdram /dvdram auto noauto,rw,user,users 0 0 ### nfs mount (stable): foo:/ /foo/root nfs defaults,noauto,[..] 0 0 ### by id mount of one USBstick (stable): /dev/disk/by-id/usb-USB_2.0_Flash_Disk_[..]-0:0-part1 \ /usbadata4 auto defaults,user,users,iocharset=iso8859-15,noauto 0 0 ### one of the front-USB2.0-ports by path (stable): /dev/disk/by-path/pci-0000:00:13.2-usb-0:1:1.0-scsi-0:0:0:0-part1 \ /usb6p1 auto defaults,user,users,iocharset=iso8859-15,noauto 0 0 HTH, -dnh [1] a whopping 1330 bytes of awk :) To e.g. always load sata_sil after pata_atiixp: $ echo 'above pata_atiixp sata_sil' | ~/src/modprobe_gen/modprobe_gen ### above pata_atiixp sata_sil install pata_atiixp /sbin/modprobe --ignore-install pata_atiixp && {\ /sbin/modprobe sata_sil;\ } remove pata_atiixp {\ /sbin/modprobe --remove sata_sil;\ }; /sbin/modprobe --ignore-remove --remove pata_atiixp; or the other way around (you could switch the modules too ;) first sata_sil, then pata_atiixp: $ echo 'below pata_atiixp sata_sil' | ~/src/modprobe_gen/modprobe_gen ### below pata_atiixp sata_sil install pata_atiixp {\ /sbin/modprobe sata_sil;\ }; /sbin/modprobe --ignore-install pata_atiixp remove pata_atiixp /sbin/modprobe --ignore-remove --remove pata_atiixp && {\ /sbin/modprobe --remove sata_sil;\ } -- [Mouse problems] So, delete /bin/cat -- I tried that. I had to "rm /bin/cat" nine times before it disappeared. (Gabor Lenart, Mans Rullgard) -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

Carlos E. R. composed on 2014-10-29 15:12 (UTC+0100):

...It happens if I boot with only the internal hard disks powered, and next time I power up one external. The external becomes sda, and what was sda, the boot disk, becomes sdb, and still boots fine.

That is, I use hotplug.

It either is the BIOS which automatically changes the identification (number) of each device, or the kernel finding one disk before another, and naming the first found one sda. In my machine I think it is this second issue.

I think, but I'd have to verify, that even with hibernation this change may trigger, instead of a full reboot.

Those who experience this and don't like it and whose external disks are on different controllers than the internal disk(s) may be able to avoid or reduce the unstable device naming by excluding the driver for the external drive controller(s) from their initrds. Through 13.1, remove the external driver from /etc/sysconfig/kernel's INITRD_MODULES= list. Post-13.1, in dracut.conf include the unwanted driver in the omit_drivers+= list. That should cause the internal disk's driver to be the first to find disks and start with sda. It's how I keep disks attached to PCI SATA cards' eSATA ports from stealing sda and sdb from the motherboard controllers if powered up and connected at boot time. -- "The wise are known for their understanding, and pleasant words are persuasive." Proverbs 16:21 (New Living Translation) Team OS/2 ** Reg. Linux User #211409 ** a11y rocks! Felix Miata *** http://fm.no-ip.com/ -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 2014-10-29 21:24, Ruben Safir wrote:

On Wed, Oct 29, 2014 at 03:12:31PM +0100, Carlos E. R. wrote:

...

and next time I power up one external. The external becomes sda,

And what ELSE would you expect to happen if you set your bios up to function in that method.

In what method, pray? What other method do you think I can setup MY BIOS?

...

and what was sda, the boot disk, becomes sdb, and still boots fine.

That is, I use hotplug.

You don't hotplug your boot system and core OS.

Of course not. You can not umount it live.

I won't even ask why you would think of doing this but the system is working 100% predicably and correctly and the problem is between the keyboard and the chair.

There is no problem at all. I don't see any problem. My machine works reliably and boots no matter how I shuffle my disks. It is properly designed, together with Linux, to handle modern life properly. -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" at Telcontar)

On Wed, Oct 29, 2014 at 08:26:09AM +0100, Carlos E. R. wrote:

On 2014-10-29 04:39, Ruben Safir wrote:

...

That is totally incomprehensible and doesn't tell me any useful information.... list where the hell sda2 is mounted.

You must be daft.

Why would I care where is sda2 mounted? Why would I care if it is the second or the third partition in the first or the fourth disk? I just want the partition that I reserve for home to be mounted on /home and not on /usr.

You probably don't care. For the rest of use, we want to know what out hardware is doing. A) we know what our hard is B) we want to know what it is doing C) we want it to do what the to do and expect it to do so without obfucation If I wasn't reading this, I wouldn't believe it was possible that computer users and developers would say such idiot things like, "I don't care what my partition is, or what my hard drive geometry is" This is getting depressing at this point.

-- Cheers / Saludos,

Carlos E. R. (from 13.1 x86_64 "Bottle" at Telcontar)

-- So many immigrant groups have swept through our town that Brooklyn, like Atlantis, reaches mythological proportions in the mind of the world - RI Safir 1998 http://www.mrbrklyn.com DRM is THEFT - We are the STAKEHOLDERS - RI Safir 2002 http://www.nylxs.com - Leadership Development in Free Software http://www2.mrbrklyn.com/resources - Unpublished Archive http://www.coinhangout.com - coins! http://www.brooklyn-living.com Being so tracked is for FARM ANIMALS and and extermination camps, but incompatible with living as a free human being. -RI Safir 2013 -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On Wed, Oct 29, 2014 at 03:25:38PM +0100, Carlos E. R. wrote:

On 2014-10-29 12:32, Ruben Safir wrote:

...

On Wed, Oct 29, 2014 at 08:26:09AM +0100, Carlos E. R. wrote:

...

...

Why would I care where is sda2 mounted? Why would I care if it is the second or the third partition in the first or the fourth disk? I just want the partition that I reserve for home to be mounted on /home and not on /usr.

You probably don't care. For the rest of use, we want to know what out hardware is doing.

A) we know what our hard is

And I do.

...

B) we want to know what it is doing

And I do.

...

C) we want it to do what the to do and expect it to do so without obfucation

And I do.

I don't care where sda2 is mounted, because the name sda2 IS NOT GUARANTEED stable.

Actually it is. sda2 will always be sda2 until you make a change in the bios. It is the first device on your sata chain.

It can change from boot to boot,

if you say so...

and this is the kernel way of doing things since several years. The Linux kernel is not static, it evolves fast, and you have to adapt to it, not the kernel to you. If you can not stand this situation,

dmesg says sda BTW - why are you rebooting the system?

then perhaps you should consider BS, solaris, whatever, or not upgrade Linux.

(If you want to complain about this, complain to Linus himself)

I WILL talk to him about this and other things the next time I see him.

I repeat: if I use names like sda2 in my fstab, my machine will boot one day, and not the next.

Not if I was running it.

It is not reliable for every machine. It is static only for machines with static hardware config.

Yeah, so let me get this straight. You hardware is not in a static configuration so you break the software?

...

If I wasn't reading this, I wouldn't believe it was possible that computer users and developers would say such idiot things like, "I don't care what my partition is, or what my hard drive geometry is"

I consider that it is you who is behaving idiotically in this respect.

No, but this is actually getting boring. When something this stupid is being defended, it is time to shrug your shoulders and feel bad for those that believe such crap. There is nothing that can be done for someone who doesn't want to know what his hardware configuration is prior to booting the OS.

Of course I do care what my partition is. I just don't care about what /dev/sdaXY UNSTABLE name it gets.

It has been stable now for 2 decades.

I care about the STABLE name it has.

Well, try fixing your hardware... then because sda is assigned by the bios. [ 1.621444] sd 0:0:0:0: [sda] 488281250 512-byte logical blocks: (250 GB/232 GiB) [ 1.621511] sd 0:0:0:0: [sda] Write Protect is off [ 1.621514] sd 0:0:0:0: [sda] Mode Sense: 00 3a 00 00 [ 1.621543] sd 0:0:0:0: [sda] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA [ 1.667459] sda: sda1 sda2 sda4 [ 1.667893] sd 0:0:0:0: [sda] Attached SCSI disk as for the new terminalogy, that is entirly meaningless. -- So many immigrant groups have swept through our town that Brooklyn, like Atlantis, reaches mythological proportions in the mind of the world - RI Safir 1998 http://www.mrbrklyn.com DRM is THEFT - We are the STAKEHOLDERS - RI Safir 2002 http://www.nylxs.com - Leadership Development in Free Software http://www2.mrbrklyn.com/resources - Unpublished Archive http://www.coinhangout.com - coins! http://www.brooklyn-living.com Being so tracked is for FARM ANIMALS and and extermination camps, but incompatible with living as a free human being. -RI Safir 2013 -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On Wed, Oct 29, 2014 at 10:26:42PM +0200, ellanios82 wrote:

On 10/29/2014 10:10 PM, Ruben Safir wrote:

...

Actually it is. sda2 will always be sda2 until you make a change in the bios. ................

- my experience : having 2 SATA SCSI disks seen as sda & sdb

then,

adding an ATA disk on 80 channel ribbon cable caused that added ATA to become sda

when you added the ata either A) you entered your bios to get the ATA bus up B) The bios program CHANGED with the additional hardware, so in effect, you did change your bios. This is no better than telling me you took your SATA drive out, or flipped them on the chain. Or are you saying that every time you boot, now that the ATA bus has been added, you get different assignments by your bios. If that is the case, change hardware....to hardware that works.

: this occured without making any manual change to BIOS

YOU added the word manual.... just to point it out.

newer syntax: /dev/disk/by-label/sata01root / ext4 ...

Really, what device is this?

/dev/disk/by-label/sata03home /home ext4 ... And what device is this?

/dev/disk/by-label/pata01oldroot /media/oldroot ext3 ... What partition is this on?

/dev/disk/by-label/pata03oldhome /media/oldhome ext3 ...

uselss syntax....

-- "

.......................

regards

-- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

-- So many immigrant groups have swept through our town that Brooklyn, like Atlantis, reaches mythological proportions in the mind of the world - RI Safir 1998 http://www.mrbrklyn.com DRM is THEFT - We are the STAKEHOLDERS - RI Safir 2002 http://www.nylxs.com - Leadership Development in Free Software http://www2.mrbrklyn.com/resources - Unpublished Archive http://www.coinhangout.com - coins! http://www.brooklyn-living.com Being so tracked is for FARM ANIMALS and and extermination camps, but incompatible with living as a free human being. -RI Safir 2013 -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On Wed, Oct 29, 2014 at 06:27:29PM -0400, Felix Miata wrote:

Ruben Safir composed on 2014-10-29 17:13 (UTC-0400):

...

...

/dev/disk/by-label/pata03oldhome /media/oldhome ext3 ...

...

uselss syntax....

No one is making you use it.

OTOH, it is available to those who comprehend utility in it.

I'm really not oposed to that. Choice is good and it allows for people to customized the use and behaviors of their systems for their own particular situations. I was content that I was able to run fdisk on the uuid names, which is at least it works! Good acept you still have touble know which device is say, failing, or causing troubles. If you want to have this name added to the table, that would have been a better solution. Two things though: I'm sure it is not really a choice and this trend of movement away from the hardware sucks. I don't want or need long drawnout human unreadable computer generated names for disks or ehternet cards or anything else. THIS for example, is totally useless BS ruben@laptop~\> ifconfig enp3s0f2: flags=4099 mtu 1500 ether 10:c3:7b:be:72:53 txqueuelen 1000 (Ethernet) RX packets 0 bytes 0 (0.0 B) RX errors 0 dropped 0 overruns 0 frame 0 TX packets 0 bytes 0 (0.0 B) TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0 lo: flags=73 mtu 65536 inet 127.0.0.1 netmask 255.0.0.0 inet6 ::1 prefixlen 128 scopeid 0x10<host> loop txqueuelen 0 (Local Loopback) RX packets 1385 bytes 112364 (109.7 KiB) RX errors 0 dropped 0 overruns 0 frame 0 TX packets 1385 bytes 112364 (109.7 KiB) TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0 wlp2s0: flags=4163 mtu 1500 inet 10.0.0.45 netmask 255.255.255.0 broadcast 10.0.0.255 inet6 fe80::5627:1eff:fee0:7c8f prefixlen 64 scopeid 0x20<link> ether 54:27:1e:e0:7c:8f txqueuelen 1000 (Ethernet) RX packets 1287287 bytes 1611179064 (1.5 GiB) RX errors 0 dropped 0 overruns 0 frame 0 TX packets 832116 bytes 110236677 (105.1 MiB) TX errors 0 dropped 0 overruns 0 carrier 0 collisions 0 ruben@laptop~\>

Contemplate if you will ramifications of device names and UUIDs among the following excerpts

I don't even know the point of that is. At this point, I really don't want to spend the time teasing out what tht means. I'm busy with EER graphs, and Parity checks on the memory bus.

from a list of files in /dev/disk/by-label/ here: lrwxrwxrwx 1 11 Oct 29 17:56 fedora22 -> ../../sda16 lrwxrwxrwx 1 11 Oct 29 17:56 fedora20 -> ../../sda17 lrwxrwxrwx 1 11 Oct 29 17:56 fedora21 -> ../../sda18 lrwxrwxrwx 1 11 Oct 29 17:56 19mageia -> ../../sda19 lrwxrwxrwx 1 11 Oct 29 17:56 os131p20 -> ../../sda20 lrwxrwxrwx 1 11 Oct 29 17:56 os132p21 -> ../../sda21 lrwxrwxrwx 1 11 Oct 29 17:56 22cauldrn -> ../../sda22 lrwxrwxrwx 1 11 Oct 29 17:56 24buntu -> ../../sda24 lrwxrwxrwx 1 11 Oct 29 17:56 MyDocs -> ../../sda27 lrwxrwxrwx 1 11 Oct 29 17:56 WINDRIVERS -> ../../sda29 lrwxrwxrwx 1 11 Oct 29 17:56 os114hs80 -> ../../sdb20 lrwxrwxrwx 1 11 Oct 29 17:56 os121hs80 -> ../../sdb21 lrwxrwxrwx 1 11 Oct 29 17:56 os122hs80 -> ../../sdb22 lrwxrwxrwx 1 11 Oct 29 17:56 os123hs80 -> ../../sdb23 lrwxrwxrwx 1 11 Oct 29 17:56 os131hs80 -> ../../sdb24 lrwxrwxrwx 1 11 Oct 29 17:56 os132hs80 -> ../../sdb25 lrwxrwxrwx 1 11 Oct 29 17:56 os133hs80 -> ../../sdb26 -- "The wise are known for their understanding, and pleasant words are persuasive." Proverbs 16:21 (New Living Translation)

Team OS/2 ** Reg. Linux User #211409 ** a11y rocks!

Felix Miata *** http://fm.no-ip.com/ -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

-- So many immigrant groups have swept through our town that Brooklyn, like Atlantis, reaches mythological proportions in the mind of the world - RI Safir 1998 http://www.mrbrklyn.com DRM is THEFT - We are the STAKEHOLDERS - RI Safir 2002 http://www.nylxs.com - Leadership Development in Free Software http://www2.mrbrklyn.com/resources - Unpublished Archive http://www.coinhangout.com - coins! http://www.brooklyn-living.com Being so tracked is for FARM ANIMALS and and extermination camps, but incompatible with living as a free human being. -RI Safir 2013 -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

Ruben Safir composed on 2014-10-29 16:10 (UTC-0400):

sda2 will always be sda2 until you make a change in the bios.

Never say "never", and never say "always". Most times you do you can be proven wrong.

It is the first device on your sata chain.

Where the chain starts can depend on the installed hardware, which can change when external devices are added or removed. The BIOS may or may not be capable of ordering such that external devices will not change the order. Maybe you have hardware that makes sda stay sda. I have hardware with BIOS that doesn't and can't. To keep sda sda here on those installations requires effort be put into the OS start procedure to work around the BIOS shortcoming. -- "The wise are known for their understanding, and pleasant words are persuasive." Proverbs 16:21 (New Living Translation) Team OS/2 ** Reg. Linux User #211409 ** a11y rocks! Felix Miata *** http://fm.no-ip.com/ -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 2014-10-29 21:54, Anton Aylward wrote:

Worse than that. In the past Felix has also pointed out that using fdisk to change the partitions can alter what partition is called #2 even if you didn't make any alterations to the partition you used to call #2.

I said this too, earlier in this thread. -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" at Telcontar)

On Wed, Oct 29, 2014 at 04:33:16PM -0400, Felix Miata wrote:

Ruben Safir composed on 2014-10-29 16:10 (UTC-0400):

...

sda2 will always be sda2 until you make a change in the bios.

Never say "never", and never say "always". Most times you do you can be proven wrong.

...

It is the first device on your sata chain.

Where the chain starts can depend on the installed hardware, which can change when external devices are added or removed.

irrelevent. My ceiling fans aren't controlled and numbered by my bios either. My bios isn't going to assing it /dev/sda (although udev might do that. udev can do anything) If someone complains that their temporary hard drives are not always the same name, good! It everything is as expected then. dmesg it and mount it by hand. Ruben

The BIOS may or may not be capable of ordering such that external devices will not change the order. Maybe you have hardware that makes sda stay sda. I have hardware with BIOS that doesn't and can't. To keep sda sda here on those installations requires effort be put into the OS start procedure to work around the BIOS shortcoming. -- "The wise are known for their understanding, and pleasant words are persuasive." Proverbs 16:21 (New Living Translation)

Team OS/2 ** Reg. Linux User #211409 ** a11y rocks!

Felix Miata *** http://fm.no-ip.com/ -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

-- So many immigrant groups have swept through our town that Brooklyn, like Atlantis, reaches mythological proportions in the mind of the world - RI Safir 1998 http://www.mrbrklyn.com DRM is THEFT - We are the STAKEHOLDERS - RI Safir 2002 http://www.nylxs.com - Leadership Development in Free Software http://www2.mrbrklyn.com/resources - Unpublished Archive http://www.coinhangout.com - coins! http://www.brooklyn-living.com Being so tracked is for FARM ANIMALS and and extermination camps, but incompatible with living as a free human being. -RI Safir 2013 -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 2014-10-29 22:38, Felix Miata wrote:

Ruben Safir composed on 2014-10-29 17:18 (UTC-0400):

Again you glossed over or failed to recognize an important implicit point. I didn't write anything about *when* external devices were added or removed. Here when an external HD is added and powered up while the PC is powered down, and then powering up the PC, my BIOS (actually most, maybe all, of the many BIOS I have) will find the external device first, and assign it sda instead of an internal device. That makes your assertion of irrelevance incorrect WRT any BIOS that behaves like mine instead of like yours.

He probably expects us to edit fstab everytime we connect a disk on eSATA. Maybe he never connects/disconnects disk, because it is such a pain to have to edit fstab to conform. Better not boot, no disks changes. Whell, I do disk changes, hotplug, and I don't have to edit anything in bios, fstab, or grub. It simply works as designed. -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" at Telcontar)

Ruben Safir composed on 2014-10-30 02:05 (UTC-0400):

explain again why if a device is 'hot swappable' that you reboot your machine with it plugged in?

Probably because all of us on this list are human. That impacts how devs write code, why some options exist, and how distro packagers select defaults. -- "The wise are known for their understanding, and pleasant words are persuasive." Proverbs 16:21 (New Living Translation) Team OS/2 ** Reg. Linux User #211409 ** a11y rocks! Felix Miata *** http://fm.no-ip.com/ -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On Wed, Oct 29, 2014 at 05:38:44PM -0400, Felix Miata wrote:

Ruben Safir composed on 2014-10-29 17:18 (UTC-0400):

...

On Wed, Oct 29, 2014 at 04:33:16PM -0400, Felix Miata wrote:

...

...

Ruben Safir composed on 2014-10-29 16:10 (UTC-0400):

...

...

...

sda2 will always be sda2 until you make a change in the bios.

...

...

Never say "never", and never say "always". Most times you do you can be proven wrong.

...

...

...

It is the first device on your sata chain.

...

...

Where the chain starts can depend on the installed hardware, which can change when external devices are added or removed.

...

irrelevent. My ceiling fans aren't controlled and numbered by my bios either.

...

My bios isn't going to assing it /dev/sda (although udev might do that. udev can do anything)

...

If someone complains that their temporary hard drives are not always the same name, good! It everything is as expected then.

Again you glossed over or failed to recognize an important implicit point. I didn't write anything about *when* external devices were added or removed. Here when an external HD is added and powered up while the PC is powered down, and then powering up the PC, my BIOS (actually most, maybe all, of the many BIOS I have) will find the external device first, and assign it sda instead of an internal device.

I understand the point and thanks for making it even clearer. I would need to sit with your bios and look at it BUT, archetectually, the USB or Memory slots are not on the main system bus. They do NOT start with sda1 EXCEPT, and here is the big one, if you configured the bios to boot from the external port rather than your drive. Don't boot your box with the flashdrive/external attached, because it is behaving just like you programmed the bios to. Now if this is not the situation, then I agree with you. This is a bug in the bios, because the bios difinitely internal system bus before the software driven usb or firewire connections. You have have it do a network boot and even a diskless system. Are you then going to complain that sda moved on you because your bios is doing what you tell it too?

That makes your assertion of irrelevance incorrect WRT any BIOS that behaves like mine instead of like yours.

Eh - sorry - I was being cranky without a clear reason.

-- "The wise are known for their understanding, and pleasant words are persuasive." Proverbs 16:21 (New Living Translation)

Team OS/2 ** Reg. Linux User #211409 ** a11y rocks!

Felix Miata *** http://fm.no-ip.com/ -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

-- So many immigrant groups have swept through our town that Brooklyn, like Atlantis, reaches mythological proportions in the mind of the world - RI Safir 1998 http://www.mrbrklyn.com DRM is THEFT - We are the STAKEHOLDERS - RI Safir 2002 http://www.nylxs.com - Leadership Development in Free Software http://www2.mrbrklyn.com/resources - Unpublished Archive http://www.coinhangout.com - coins! http://www.brooklyn-living.com Being so tracked is for FARM ANIMALS and and extermination camps, but incompatible with living as a free human being. -RI Safir 2013 -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

Le 30/10/2014 08:13, Felix Miata a écrit :

Many BIOS by default behave exactly like the BIOS of 3 decades ago, booting from a removable device if a bootable removable device (floppy then; USB, Firewire, eSATA, or internal OM on ATA now) is present at boot time. Since it's a common if not dominant default, saying "I" configured it so is inapt.

not dominant, no. I very often install openSUSE on other people computer and I often have to press whatever key to bring up the boot menu. and why will I have to keep a *bootable* external device if I don't need it on purpose? I have lot of *data* disks, not bootable ones! that said, at least on esata it's possible, and stable. When I first tried my new SSD, I plugged it in a esata dock and runned it that way for more than a month before deciding to set it inside the box. All this without problem. I do not use /dev/sdx, but I could (and did some time ago) jdd -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On Thu, Oct 30, 2014 at 12:01 PM, Hans Witvliet wrote:

But I agree, in whatever USB-slot you put the drive, or regardless how many other usb-drives are connected, there should be an fixed identifier for each drive.

USB does not have "fixed identifiers" for attached devices. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 2014-10-30 10:01, Hans Witvliet wrote:

No more horror-scenario's from the past, where external scsi-drives ended up somewhere else, only because one of the powercables were disconnected. If a drive is (momentarily) not detected, it just should mean that its content is not available. It should not have any consequence where the data of other drives might end up...

LOL. That's precisely the issue solved by using UUIDs ;.)

** afaik, you cannot convert a factory-supplied disk with a ms-partition table into a GPT-partition table. Only a compete reinstall, and mostly pc's are delivered with a rescue partition instead on installable media.

I heard of a tool. Dunno. But MS may refuse to boot on the grounds of "changed hardware, you are a pirate!". -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" at Telcontar)

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On 2014-11-03 01:37, Linda Walsh wrote:

Carlos E. R. wrote:

Vs. -- if you use UUID, and try to boot from it (aka boot=/dev/disk/by-uuid1 root=/dev/disk/by-uuid2), the kernel won't boot as it doesn't recognize anything in /dev/disk without a system being already booted (aka an init-ramdisk that runs 'udev'(or *equiv*)) that assigns and interprets those UUID's.

WRONG. This is my line in grub 1 menu.lst: kernel /vmlinuz-3.11.10-21-desktop root=/dev/disk/by-label/a_main resume=/dev/disk/by-label/b_swap showopts splash=verbose console=tty0 vga=0x31a Because there it is the *kernel* which does the reading, not grub. No matter that you thing that you need udev, it is obviously working. Grub 1 uses this to find the kernel itself: root (hd0,1) which depends on hd0 always being hd0, as handled of by the *BIOS*. Then, depending on what is plugged that day, it becomes either sda or sdb, depending on the *kernel's* choice of the day. On another system using grub2, I see: menuentry 'openSUSE 13.1' --class 'opensuse-13-1' --class gnu-linux --class gnu --class os $menuentry_id_option 'gnulinux-simple-00eb9a40-d067-459e-a22f-1d3b667dddbb' { load_video set gfxpayload=keep insmod gzio insmod part_msdos insmod ext2 set root='hd0,msdos3' if [ x$feature_platform_search_hint = xy ]; then search --no-floppy --fs-uuid --set=root --hint-bios=hd0,msdos3 --hint-efi=hd0,msdos3 --hint-baremetal=ahci0,msdos3 --hint='hd0,msdos3' 6d9d4270-3fd1-4027-9316-c614d6c090e4 else search --no-floppy --fs-uuid --set=root 6d9d4270-3fd1-4027-9316-c614d6c090e4 fi echo 'Loading Linux 3.11.10-21-desktop ...' linux /vmlinuz-3.11.10-21-desktop root=UUID=00eb9a40-d067-459e-a22f-1d3b667dddbb resume=/dev/disk/by-label/b_swap splash=verbose showopts echo 'Loading initial ramdisk ...' initrd /initrd-3.11.10-21-desktop } which as you can see, uses 'hd0' initially, but if that fails it searches for the listed uuid. Thus, grub2 may boot a system based on the UUID, regardless of where it is.

...

...

If a drive is (momentarily) not detected, it just should mean that its content is not available. It should not have any consequence where the data of other drives might end up...

LOL.

That's precisely the issue solved by using UUIDs ;.)

---

That's really up to the sysadmin who should know to put fixed drives in /etc/fstab and drives that might become "'momentarily'-not-detected" in the /etc/automounter hierarchy that expects mounts to "come and go"...

I don't expect "/home" or "/" (both internal and fixed disks) to come and go. If either disk is missing, for whatever reason, I do not want the boot to proceed. This can happen if I accidentally shuffled the cables around a bit and forgot to plug one. I do not want "/usr/local/" to appear mounted in "/var/spool/" because I swapped a cable in mistake, inside the box. All of those examples are permanent drives, internal. Using, as I do, 'label's, or 'uuid's, that type of accident simply doesn't happen. I simply do not notice, the system boots happily and working correctly, every thing mounted on the expected place. That's progress to me.

...

I heard of a tool. Dunno. But MS may refuse to boot on the grounds of "changed hardware, you are a pirate!".

---

While there are many reasons MS won't boot, changing HW causing Windows to not validate is not one of them -- it does boot, but operates in some "restricted mode" (that does something like reboot every few minutes or something -- that can give you a chance to recover a valid installation, but do little else; admittedly with all the Win updates, this might have changed, but I seem to remember that even MS, recognized as folly, piracy yielding the same feedback as other problems.

I call that "non-booting". :-| - -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" at Telcontar) -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.22 (GNU/Linux) iEYEARECAAYFAlRW5FwACgkQtTMYHG2NR9X0RgCfQtf/S5RO0XkNhF7u5l/42j9F Ec4An1EEUdGG+MvHrTdJI9usP2binV/+ =wXd3 -----END PGP SIGNATURE----- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

В Sun, 02 Nov 2014 23:56:34 -0500 Anton Aylward пишет:

On 11/02/2014 09:11 PM, Carlos E. R. wrote:

...

This is my line in grub 1 menu.lst:

kernel /vmlinuz-3.11.10-21-desktop root=/dev/disk/by-label/a_main resume=/dev/disk/by-label/b_swap showopts splash=verbose console=tty0 vga=0x31a

Because there it is the *kernel* which does the reading, not grub. No matter that you thing that you need udev, it is obviously working.

The kernel, that vmlinux (which leads to the initrd on my system) has other smarts as well. My grub2 menu line has

/vmlinuz-3.11.10-101.g966c1cc-desktop root=/dev/mapper/vgmain-vROOT resume=/dev/disk/by-label/SWAP splash=silent quiet showopts

Note that "/dev/mapper". Smarts to deal with LVM.

It is not kernel, it is initrd. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

В Mon, 03 Nov 2014 12:54:10 +0100 "Carlos E. R." пишет:

On 2014-11-03 06:54, Andrei Borzenkov wrote:

...

В Sun, 02 Nov 2014 23:56:34 -0500 Anton Aylward <> пишет:

...

...

/vmlinuz-3.11.10-101.g966c1cc-desktop root=/dev/mapper/vgmain-vROOT resume=/dev/disk/by-label/SWAP splash=silent quiet showopts

Note that "/dev/mapper". Smarts to deal with LVM.

It is not kernel, it is initrd.

You mean scripts in it?

Of course.

Oh.

Curious: would it handle encryption in the same manner at that step?

Yes.

Placing a "/dev/mapper/cr_root" in there? That's my missing link, I believe.

What about 13.2 with dracut?

Same, module different implementation and different bugs.

Heh. If Linda removes initrd, then that's why she thinks uuid are not supported at boot ;-)

Carlos E. R. wrote:

On 2014-11-03 01:37, Linda Walsh wrote:

...

Vs. -- if you use UUID, and try to boot from it (aka boot=/dev/disk/by-uuid1 root=/dev/disk/by-uuid2), the kernel won't boot as it doesn't recognize anything in /dev/disk without a system being already booted (aka an init-ramdisk that runs 'udev'(or *equiv*)) that assigns and interprets those UUID's.

[sic] WRONG.

This is my line in grub 1 menu.lst:

kernel /vmlinuz-3.11.10-21-desktop root=/dev/disk/by-label/a_main resume=/dev/disk/by-label/b_swap showopts splash=verbose console=tty0 vga=0x31a

[sic] Because there it is the *kernel* which does the reading, not grub. No matter that you thing that you need udev, it is obviously working.

On 2014-11-03 06:54, Andrei Borzenkov wrote:

...

It is not kernel, it is initrd.

Carlos E. R. wrote:

You mean scripts in it? Oh.

Heh. If Linda removes initrd, then that's why she thinks uuid are not supported at boot ;-)

*Ahem*, Carlos, Do you wish to retract your "WRONG" now? I.e. As the paragraph you quote, says: "the kernel won't boot as it doesn't recognize anything in /dev/disk without a system being already booted (aka an init-ramdisk that runs 'udev'(or *equiv*)) that assigns and interprets those UUID's. I.e. if linda specified that the features of fancy name booting (label, uuid, path...etc.) are not available unless your system is already booted (as from an init-RamDisk (initrd)). Linda's question was based on the premise of not prebooting with a minikernel and starting udev (or similar) off of a ramdisks (initrd). Using device names that don't require a kernel+scripts already be running and available before 'boot' was a primary goal in my choosing a reliable naming scheme for boot devices. Your systems only works if you are already booted (as from an initrd). At which point, talking about whether or not something is supported by the kernel before boot is moot (because in your case, your system is already booted -- which is not to say all normal runtime services are running). -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On 2014-11-06 02:50, jdebert wrote:

Wait. There's a kernel in initrd?

No. -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" at Telcontar)

On Thu, Nov 6, 2014 at 4:18 AM, Linda Walsh wrote:

Using device names that don't require a kernel+scripts already be running and available before 'boot' was a primary goal in my choosing a reliable naming scheme for boot devices.

Linux kernel does not offer reliable naming scheme for boot devices which is one of reasons to use initrd. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On Thu, Oct 30, 2014 at 05:30:54AM +0100, Carlos E. R. wrote:

On 2014-10-29 21:10, Ruben Safir wrote:

...

On Wed, Oct 29, 2014 at 03:25:38PM +0100, Carlos E. R. wrote:

...

...

I don't care where sda2 is mounted, because the name sda2 IS NOT GUARANTEED stable.

Actually it is. sda2 will always be sda2 until you make a change in the bios.

Nope.

...

It is the first device on your sata chain.

which changes when you hotplug another device.

No it doesn't. but this is a useless discssion at this point. You don't need a guarantee anyway. That is utter nonsense. What you need is a bios and system that behaves as programmed and is predictable. This idea that everytime you turn your computer on you have no idea where your hard drives show up untrue. Learn how your hardwae works or take a computing class.

...

BTW - why are you rebooting the system?

I don't, in weeks. When I do, it is because updates force me, or because the kernel crashes.

It does for you. My workation uptime prior to the 13.1 upgrade was over 19 months. I think my housekeeper disconnected it on accident. This can be quite entertaining. Without changing your hardware, I want you to reboot your system 30 times and tell use the number of times each device shows up on each device node. Then, if it changes as you say, please break down the statistical frequency of /dev/sdx or /dev/sgx or whatever. Please inform us of the bios you are using and the current hardware configuration. If it is all true, as you describe, I'd like to avoid that broken hardware configuration. A computer that can't be guaranteed to find the MBR is certainly worthy of a class action lawsuite of some sort. If the bios doesn't remember where its hardware is, that is bad.

Oh, yes, it does. I have been tracking certain issue which has been solved recently, but the patch has not yet been published as an update.

...

...

then perhaps you should consider BS, solaris, whatever, or not upgrade Linux.

(If you want to complain about this, complain to Linus himself)

I WILL talk to him about this and other things the next time I see him.

Ok. Please post the video, I want to see it.

...

...

I repeat: if I use names like sda2 in my fstab, my machine will boot one day, and not the next.

Not if I was running it.

I dare you.

...

...

It is not reliable for every machine. It is static only for machines with static hardware config.

Yeah, so let me get this straight. You hardware is not in a static configuration so you break the software?

I don't break software nor hardware. It is designed to work this way, and it works just fine.

...

It has been stable now for 2 decades.

You are free to contribute to the kernel and make it work properly, if you think you can. I'm sure they'll welcome your ideas.

...

Well, try fixing your hardware... then because sda is assigned by the bios. [ 1.621444] sd 0:0:0:0: [sda] 488281250 512-byte logical blocks: (250 GB/232 GiB) [ 1.621511] sd 0:0:0:0: [sda] Write Protect is off [ 1.621514] sd 0:0:0:0: [sda] Mode Sense: 00 3a 00 00 [ 1.621543] sd 0:0:0:0: [sda] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA [ 1.667459] sda: sda1 sda2 sda4 [ 1.667893] sd 0:0:0:0: [sda] Attached SCSI disk

/var/log/messages-20141001.xz:<0.5> 2014-09-26 15:38:55 Telcontar kernel - - - [ 2.732770] scsi 8:0:0:0: Direct-Access ATA ST3500418AS CC37 PQ: 0 ANSI: 5 /var/log/messages-20141001.xz:<0.5> 2014-09-26 15:38:55 Telcontar kernel - - - [ 2.738473] sd 8:0:0:0: [sda] 976773168 512-byte logical blocks: (500 GB/465 GiB) /var/log/messages-20141001.xz:<0.5> 2014-09-26 15:38:55 Telcontar kernel - - - [ 2.738491] scsi 8:0:1:0: Direct-Access ATA ST3500418AS CC37 PQ: 0 ANSI: 5 /var/log/messages-20141001.xz:<0.5> 2014-09-26 15:38:55 Telcontar kernel - - - [ 2.738606] sd 8:0:1:0: [sdb] 976773168 512-byte logical blocks: (500 GB/465 GiB) /var/log/messages-20141001.xz:<0.5> 2014-09-26 15:38:55 Telcontar kernel - - - [ 2.738637] sd 8:0:1:0: [sdb] Write Protect is off /var/log/messages-20141001.xz:<0.7> 2014-09-26 15:38:55 Telcontar kernel - - - [ 2.738638] sd 8:0:1:0: [sdb] Mode Sense: 00 3a 00 00 /var/log/messages-20141001.xz:<0.5> 2014-09-26 15:38:55 Telcontar kernel - - - [ 2.738652] sd 8:0:1:0: [sdb] Write cache: enabled, read cache: enabled, doesn't support DPO or FUA /var/log/messages-20141001.xz:<0.5> 2014-09-26 15:38:55 Telcontar kernel - - - [ 2.766596] sd 8:0:0:0: [sda] Write Protect is off /var/log/messages-20141001.xz:--

/var/log/messages-20141001.xz:<0.5> 2014-10-01 22:22:35 Telcontar kernel - - - [ 0.989913] scsi 2:0:0:0: Direct-Access ATA ST31000528AS CC46 PQ: 0 ANSI: 5 /var/log/messages-20141001.xz:<0.5> 2014-10-01 22:22:35 Telcontar kernel - - - [ 0.994784] sd 2:0:0:0: [sda] 1953525168 512-byte logical blocks: (1.00 TB/931 GiB) /var/log/messages-20141001.xz:<0.5> 2014-10-01 22:22:35 Telcontar kernel - - - [ 0.999616] sd 2:0:0:0: [sda] Write Protect is off /var/log/messages-20141001.xz:<0.6> 2014-10-01 22:22:35 Telcontar kernel - - - [ 1.000212] usb 1-2: New USB device found, idVendor=05e3, idProduct=0608 /var/log/messages-20141001.xz:<0.6> 2014-10-01 22:22:35 Telcontar kernel - - - [ 1.000213] usb 1-2: New USB device strings: Mfr=0, Product=1, SerialNumber=0 /var/log/messages-20141001.xz:<0.6> 2014-10-01 22:22:35 Telcontar kernel - - - [ 1.000214] usb 1-2: Product: USB2.0 Hub /var/log/messages-20141001.xz:--

Two different entries for sda within days, and the first "ST3500418AS" is my boot disk, which I have not changed in months. Not the disk, nor the cable, board, bios config, anything.

I booted on 2014-09-26 15:38:54. I needed to copy something to an external disk, so I connected in on the caddy, made the copy, and did not extract the disk, there were more things to do. I stopped the machine on 2014-09-30 14:47:03 Started it again on 2014-10-01 22:22:32

Why did I boot, you ask? Maybe because I went on a field trip to check flood damage that afternoon. No kidding, I just looked up my logs. Emergency. As it turned out, I came back much sooner than expected. You ask too much. Why do I boot? Because I wish to!

The 1 TB HD was still on the caddy, and it takes precedence in BIOS over the boot disk, it appears earlier, and it is named sda... still, the machine boots without a complain, and everything mounts fine. I don't even notice that what on 2014-09-30 was sda, on 2014-10-01 was... wait:

<0.5> 2014-10-01 22:22:35 Telcontar kernel - - - [ 2.742632] scsi 8:0:0:0: Direct-Access ATA ST3500418AS CC37 PQ: 0 ANSI: 5 <0.6> 2014-10-01 22:22:35 Telcontar kernel - - - [ 2.745320] ata10.01: configured for UDMA/133 <0.5> 2014-10-01 22:22:35 Telcontar kernel - - - [ 2.753744] sd 8:0:0:0: [sdb] 976773168 512-byte logical blocks: (500 GB/465 GiB) <0.5> 2014-10-01 22:22:35 Telcontar kernel - - - [ 2.753775] scsi 8:0:1:0: Direct-Access ATA ST3500418AS CC37 PQ: 0 ANSI: 5 <0.5> 2014-10-01 22:22:35 Telcontar kernel - - - [ 2.753874] sd 8:0:1:0: [sdc] 976773168 512-byte logical blocks: (500 GB/465 GiB)

sdb. Or perhaps sdc :-P

-- Cheers / Saludos,

Carlos E. R. (from 13.1 x86_64 "Bottle" at Telcontar)

-- So many immigrant groups have swept through our town that Brooklyn, like Atlantis, reaches mythological proportions in the mind of the world - RI Safir 1998 http://www.mrbrklyn.com DRM is THEFT - We are the STAKEHOLDERS - RI Safir 2002 http://www.nylxs.com - Leadership Development in Free Software http://www2.mrbrklyn.com/resources - Unpublished Archive http://www.coinhangout.com - coins! http://www.brooklyn-living.com Being so tracked is for FARM ANIMALS and and extermination camps, but incompatible with living as a free human being. -RI Safir 2013 -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On Wed, Oct 29, 2014 at 10:58:26AM -0300, Cristian Rodríguez wrote:

El 28/10/14 a las #4, Carl Spitzer escribió:

...

with all the talk against systemd which should be dumped or the distros forked to remove it I have another older problem uuid is a pain to use .

the old way of /dev/* was easier to use and remember if you made a backup of this it would work on another drive whereas uuid ties you to one drive.

cat /etc/fstab /dev/hda3 / ext3 acl,user_xattr 1 1 /dev/hda7 /boot ext2 acl,user_xattr 1 2 /dev/hdb1 /home reiserfs defaults 1 2 /dev/hda8 swap swap defaults 0 0

is there a way back for 11.4 and 13.1??

It still *may* work, but it is not recommended for a very simple reason, kernel device names are not predictably named. if you want that changed, complain in LKML but I will not hold my breath.

Actually, it has for 20 plus years. Those hard drives have been predictably named for 2 decades. Ruben

-- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

-- So many immigrant groups have swept through our town that Brooklyn, like Atlantis, reaches mythological proportions in the mind of the world - RI Safir 1998 http://www.mrbrklyn.com DRM is THEFT - We are the STAKEHOLDERS - RI Safir 2002 http://www.nylxs.com - Leadership Development in Free Software http://www2.mrbrklyn.com/resources - Unpublished Archive http://www.coinhangout.com - coins! http://www.brooklyn-living.com Being so tracked is for FARM ANIMALS and and extermination camps, but incompatible with living as a free human being. -RI Safir 2013 -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org

On Thu, Oct 30, 2014 at 05:43:15AM +0100, Carlos E. R. wrote:

On 2014-10-29 21:25, Ruben Safir wrote:

...

On Wed, Oct 29, 2014 at 10:58:26AM -0300, Cristian Rodríguez wrote:

...

...

It still *may* work, but it is not recommended for a very simple reason, kernel device names are not predictably named. if you want that changed, complain in LKML but I will not hold my breath.

Actually, it has for 20 plus years.

Those hard drives have been predictably named for 2 decades.

Till manufacturers invented SATA and eSATA, so that we can connect a dozen internal disks, instead of 4 on good machines on 2 wide ribbon cables. And they changed hardware so that we can even add or remove disk while the machine is running.

external scsi would like to wave hello at you. so would a dozen or so evernal scis zip drives.

Those times of one or two PATA disks are long gone by.

-- Cheers / Saludos,

Carlos E. R. (from 13.1 x86_64 "Bottle" at Telcontar)

-- So many immigrant groups have swept through our town that Brooklyn, like Atlantis, reaches mythological proportions in the mind of the world - RI Safir 1998 http://www.mrbrklyn.com DRM is THEFT - We are the STAKEHOLDERS - RI Safir 2002 http://www.nylxs.com - Leadership Development in Free Software http://www2.mrbrklyn.com/resources - Unpublished Archive http://www.coinhangout.com - coins! http://www.brooklyn-living.com Being so tracked is for FARM ANIMALS and and extermination camps, but incompatible with living as a free human being. -RI Safir 2013 -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse+owner@opensuse.org