SliTaz Forum
You are not logged in.
Pages: 1
Hi
I did get the old play PC from son. he did preinstall for me completely fresh an ubuntu 14.04 on it before he did give it to me. ubuntu did use the disk as UEFI...
As I did prefer Debian and it was fresh new stable (Jessie), I did remove ubuntu completely but reorganize the disk to permit more including an fat32 partition (I need Windows for Internet banking... no way to use only Linux: that bank uses more than 10 years a driver build with activeX and activeX is announced for linux years ago but not coming or coming slowly.
So I did erase the system creating a new DOS partition table and not UEFI, as UEFI would be completely new for me.
sda1 is a little boot partition containing only grub and a little linux starting frugal
sda2 is the fat32 partition with boot flag
sda3 is the extended for all Linux within sda5 debian with my /home/f (f is for files, fichiers on French, my tongue) and sda6 all the rest (data etc.) in my /home/f(iles)/my linked in partition sda4 through /etc/fstab
sda4 is swap.
since the day I did do that and get the PC, divers new Linux versions did appear (SID is becoming now more an more different from Jessie, a new Ubuntu did appear, a new SliTaz did appear, a new NuTyx did appear, so I did move the content of my initial /mnt/sda5 into /mnt/sda6, amend the values in /etc/ and sda5 is now free for test (SID, Lubuntu 15.10, Slitaz5RC).
each new installation in sda5, the Debian installer did warn: the begin of (logical) sda5 (in extended sda3) is bad.
I did reajust it moving all data and letting the Debian installer do the job...
today reading on such problems on the web, I did try:
CODE: SELECT ALL
fdisk -l
oh! wonder:
sda2 (was re-installed from Debian installer, and reformated after that as fat32) seems to be, perhaps, ok
sda5 (was re-installed from Debian installer to install SID) seems to be ok
sda6 (was re-installed from Debian installer) seems to be ok
but her container, the extended sda3 not: it doesn't on a boundary.
same thing with sda1. but creating sda1, I did certainly never enter myself some value: I did create the new partition table with gparted and in the menu of gparted I did only enter the end marking "on boundary" as "true"....
how can that happen?
is the disk only operable as UEFI?
this is the SliTaz fdisk -l message made with the actual live CD FOR THE SAME HARDDISK:
Disk /dev/sda: 750.1 GB, 750156374016 bytes
255 heads, 63 sectors/track, 91201 cylinders
Units = cylinders of 16065 * 512 = 8225280 bytes
Device Boot Start End Blocks Id System
/dev/sda1 1 38 305203+ 83 Linux
/dev/sda2 * 39 1254 9764864 83 Linux
/dev/sda3 1254 40195 312787969 5 Extended
/dev/sda4 90551 91202 5230592 82 Linux swap
/dev/sda5 1254 2470 9764864 83 Linux
/dev/sda6 2470 40195 303022080 83 Linux
his is the Debian fdisk -l message:
Disk /dev/sda: 750.2 GB, 750156374016 bytes
255 têtes, 63 secteurs/piste, 91201 cylindres, total 1465149168 secteurs
Unités = secteurs de 1 * 512 = 512 octets
Taille de secteur (logique / physique) : 512 octets / 4096 octets
taille d'E/S (minimale / optimale) : 4096 octets / 4096 octets
Identifiant de disque : 0x0006cc76
Périphérique Amorçage Début Fin Blocs Id. Système
/dev/sda1 63 610469 305203+ 83 Linux
La partition 1 ne commence pas sur une frontière de cylindre physique.
/dev/sda2 * 612352 20142079 9764864 83 Linux
/dev/sda3 20144126 645720063 312787969 5 Étendue
La partition 3 ne commence pas sur une frontière de cylindre physique.
/dev/sda4 1454686208 1465147391 5230592 82 partition d'échange Linux / Solaris
/dev/sda5 20144128 39673855 9764864 83 Linux
/dev/sda6 39675904 645720063 303022080 83 Linux
as those operations happens really rarely, I am not fit in CLI commands for that !
how to check which of both gives the real value?
how to find really a boundary ?
how to shring a partition and after that move it as an existing partition (to preserve the content, important for sda6: all data are in it ) to the found boundary ?
Offline
There's no need to be on a cylinder boundary: the disk geometry of MSDOS partition tables is fictitious, the 255/63 limits have been obsoleted by "modern" hard disks long ago.
I've read that aligning on round numbers of MB is better, but I'm not sure whether it applies to all hard disks or only SDDs. Anyway, on a "magnetic" (traditional) hard disk lack of alignment will at worst cause a small loss of performance.
Offline
Hi
Thank you llev! But...
... I suppose it can't be right if one, the «SliTaz fdisk -l» says that my sda1 starts at position «1»
and different other fdisk (Debian, Ubuntu, divers Puppy) say that my sda1 starts at position «63»!
Uses SliTaz a different arithmetic?
Without consequences for user?
Offline
Hi oui,
Answering your question about "1" and "63". Remove all the mess and compare:
First:
[c]/dev/sda1 1 38 305203+ 83 Linux
/dev/sda2 * 39 1254 9764864 83 Linux
/dev/sda3 1254 40195 312787969 5 Extended
/dev/sda4 90551 91202 5230592 82 Linux swap
/dev/sda5 1254 2470 9764864 83 Linux
/dev/sda6 2470 40195 303022080 83 Linux[/c]
Second:
[c]/dev/sda1 63 610469 305203+ 83 Linux
/dev/sda2 * 612352 20142079 9764864 83 Linux
/dev/sda3 20144126 645720063 312787969 5 Étendue
/dev/sda4 1454686208 1465147391 5230592 82 partition d'échange Linux / Solaris
/dev/sda5 20144128 39673855 9764864 83 Linux
/dev/sda6 39675904 645720063 303022080 83 Linux[/c]
Here's not only difference between 1 and 63, but all numbers are different,
As you have "63 sectors per track" seems that it calculated numbers
[*]in the tracks in the first place
[*]in the sectors in the second place
Offline
Look carefully :-)
[c]Units = cylinders of 16065 * 512 = 8225280 bytes
Device Boot Start End Blocks Id System
/dev/sda1 1 38 305203+ 83 Linux[/c]
vs.
[c]Unités = secteurs de 1 * 512 = 512 octets
Périphérique Amorçage Début Fin Blocs Id. Système
/dev/sda1 63 610469 305203+ 83 Linux[/c]
SliTaz uses busybox to supply many GNU tools, it's lighter but not fully compatible. You can force units=sectors with option "-u".
Offline
Pages: 1
[ Generated in 0.021 seconds, 7 queries executed - Memory usage: 1.55 MiB (Peak: 1.77 MiB) ]