How to obtain UUID of drives (squeeze udev lacks vol_id)
in [Debian]
|
From: Lubos Rendek on 24 Mar 2010 23:50 Hi Paul, > For me, linux device names have become very intuitive. Certainly more > intuitive than UUIDs. I've been thinking about UUIDs for a little over > a week. To me, the idea of UUIDs suffers from a great excess of > pseudo-intellectual baggage. A convention for naming things that is > deliberately designed to generate names that give no clue as to what > the thing is, or which one of several similar things this one actually > is --- is crazy. I agree, HDXn is certainly easy to understand and it gives a sys admin more transparent view into the system which also allows him/her to react to the emergency situation with better speed. What is more is that this is yet another example where MS windows people would not move from their click, drag and drop approach as to convince them that this is a better way to do it would be very hard. Anyway what is this UUID business all about? If its is so great to use multiple character strings of complete gibberish to name a partition so why is it not done somewhere on the lower kernel level and make this process completely transparent on the user level? lubo On Thu, Mar 25, 2010 at 9:38 AM, Paul E Condon <pecondon(a)mesanetworks.net> wrote: > On 20100322_175115, Ron Johnson wrote: >> On 2010-03-22 17:05, Paul E Condon wrote: >> [snip] >> > >> >You answered my question --- I didn't know about blkid utility. >> >With this one can give every partition a meaningful labels like --- >> > >> >LABEL="hda1" >> > >> >or whatever is displayed as the system device one one runs blkid >> >without options in order to get a full listing. e.g. >> >cmpq:~# blkid >> >/dev/hda1: UUID="990189bc-f7ce-41a6-9d8f-64a35349875e" TYPE="ext3" >> >/dev/hda2: LABEL="HDA2" UUID="5a02e986-8aa3-4790-aa3f-41f7f565533f" >> >TYPE="ext3" /dev/hda3: LABEL="HDA3" >> >UUID="63025622-0210-49d4-89ff-f038e6c218b6" TYPE="ext3" /dev/hda5: >> >TYPE="swap" /dev/hda6: LABEL="HDA6" >> >UUID="12148027-bc0b-4260-853e-64a8d79c6fe7" TYPE="ext3" /dev/hdb1: >> >UUID="29b36e91-32e7-46bd-b8d3-831f5dcbd337" TYPE="ext3" /dev/sda1: >> >LABEL="WDMB-1" UUID="bc1d2170-d790-4952-bbee-c3c3efdd8413" >> >TYPE="ext3" /dev/sdb1: LABEL="WDP-5" >> >UUID="8c6c3be3-4a72-45cd-929f-2dec9f185427" TYPE="ext3" >> >> Gah!!! "HDAn" are *horrible* labels. >> >> When you add another hard drive, the kernel might make it hda and >> your existing drive hdb. Or libata might convert them to sdx and >> then you'd have sda2 with label HDA2. *Very* unintuitive! > > For me, linux device names have become very intuitive. Certainly more > intuitive than UUIDs. I've been thinking about UUIDs for a little over > a week. To me, the idea of UUIDs suffers from a great excess of > pseudo-intellectual baggage. A convention for naming things that is > deliberately designed to generate names that give no clue as to what > the thing is, or which one of several similar things this one actually > is --- is crazy. > > The HDXn convention tells me which metal-box-thing I have to pull out > of the computer box and replace, if Debian/GNU/Linux informs me that > this particular HDXn is broken. I like that. > > I know this doesn't work for USB which has SCSI derived names. For > this I have been using labels for removeable drives. I have no > permanently installed USB drives, but I expect that when I do, I > will label them with some conventional names that can be put on > a sticker that is stuck on the physical thing. > >> >> Or you might need to move the drive to a machine that already has a >> partition labeled HDA2. > > If I need to move a drive to another machine, I will use blkid on > that machine to inspect the drive before I compose the new line > that will be needed in /etc/fstab. This saves some time. If > blkid can't report the labels on the HD, there is little hope of > mounting it and little point in it having a distinct ID. > >> >> Give them meaningful *symbolic* names. >> >> >This is from a system that is running Lenny. Note that the swap partition doesn't >> >have either a label or a uuid, but I had specified labels each time I installed >> >a distro. >> > >> >> That's weird. >> >> -- >> "History does not long entrust the care of freedom to the weak >> or the timid." Dwight Eisenhower >> >> >> -- >> To UNSUBSCRIBE, email to debian-user-REQUEST(a)lists.debian.org with a >> subject of "unsubscribe". Trouble? Contact >> listmaster(a)lists.debian.org >> Archive: http://lists.debian.org/4BA7F463.8010906(a)cox.net >> > > -- > Paul E Condon > pecondon(a)mesanetworks.net > > > -- > To UNSUBSCRIBE, email to debian-user-REQUEST(a)lists.debian.org > with a subject of "unsubscribe". Trouble? Contact listmaster(a)lists.debian..org > Archive: http://lists.debian.org/20100324223819.GG29954(a)big.lan.gnu > > -- http://www.linuxconfig.org -- To UNSUBSCRIBE, email to debian-user-REQUEST(a)lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmaster(a)lists.debian.org Archive: http://lists.debian.org/7d2c465d1003242040ubaa1939recd1a1d9254d6c6a(a)mail.gmail.com
From: Ron Johnson on 25 Mar 2010 08:10 On 2010-03-24 22:40, Lubos Rendek wrote: > Hi Paul, > >> For me, linux device names have become very intuitive. Certainly more >> intuitive than UUIDs. I've been thinking about UUIDs for a little over >> a week. To me, the idea of UUIDs suffers from a great excess of >> pseudo-intellectual baggage. A convention for naming things that is >> deliberately designed to generate names that give no clue as to what >> the thing is, or which one of several similar things this one actually >> is --- is crazy. > > I agree, HDXn is certainly easy to understand and it gives a sys admin > more transparent view into the system which also allows him/her to > react to the emergency situation with better speed. What is more is > that this is yet another example where MS windows people would not > move from their click, drag and drop approach as to convince them that > this is a better way to do it would be very hard. As opposed to descriptive symbolic labels like DADS_ROOT, KIDS_HOME, WORK_DATA? > Anyway what is this UUID business all about? If you knew what UUID means, I think you'd understand. > If its is so great to use > multiple character strings of complete gibberish to name a partition > so why is it not done somewhere on the lower kernel level and make > this process completely transparent on the user level? > The kernel (TTBOMK) doesn't even deal with hda1, hda2, sdb4, etc. It uses major and minor numbers which are unique to a running instance of the system. -- "History does not long entrust the care of freedom to the weak or the timid." Dwight Eisenhower -- To UNSUBSCRIBE, email to debian-user-REQUEST(a)lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmaster(a)lists.debian.org Archive: http://lists.debian.org/4BAB524A.2050905(a)cox.net
From: Tom H on 25 Mar 2010 08:30 >> >> You can cnahe the UUID of any ext partition with >> > cnahe??? >> It is Sunday and my first weekend off in a few weeks so my fingers are >> on holiday... > Yes - excuse accepted. But that still leaves the question marks. Is it meant > to be "change"?? I never have been good at anagrams - let alone anagrams > that don't contain all the right letters. (Assuming that they are the right > letters in the wrong order makes no sense at all as far as I can see.) Sorry, no idea! It was my first weekend off in either 6 or 7 weeks (I would have to check my invoices to find out!)... -- To UNSUBSCRIBE, email to debian-user-REQUEST(a)lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmaster(a)lists.debian.org Archive: http://lists.debian.org/6d4219cc1003250525l2176f195ka8dc00baff69a7af(a)mail.gmail.com
First
|
Prev
|
Pages: 1 2 3 4 5 Prev: TCP wrapper and hostname issue Next: Panner and Gnome Terminal lag time. |