I used pentium 4 mechine with 256 Mb ram and fedora core 3 for web server and four IDE HDDs [160 Gbx4] on both primary ans secondary IDE. The bios also detected all 4 HDD but after starting FC3 the system mount secondary slave HDD /dev/hdd1 as /mnt/idedisk1 /dev/hda1 / ext3 defaults 1 1 none /dev/pts devpts gid=5,mode=620 0 0 none /dev/shm tmpfs defaults 0 0 /dev/hdc1 /mec ext3 defaults 1 2 none /proc proc defaults 0 0 none /sys sysfs defaults 0 0 /dev/hdb1 /var ext3 defaults 1 2 /dev/hda2 swap swap defaults 0 0
/dev/hdd1 /mnt/idedisk1 auto amconsole,exec,noauto,managed 0 0
which appeared to be the read only device. I cannot umount this device and mount it to /dev/hdd1 ext3 . I would like to use this HDD to store the nackup data. Does any one tell me how to do this?
-- Amrit Angsusingh Nakornsawan Thailand
On Mon, Feb 20, 2006 at 23:53:47 +0700, Amrit Angsusingh amritangs@gmail.com wrote:
I used pentium 4 mechine with 256 Mb ram and fedora core 3 for web server and four IDE HDDs [160 Gbx4] on both primary ans secondary IDE. The bios also detected all 4 HDD but after starting FC3 the system mount secondary slave HDD /dev/hdd1 as /mnt/idedisk1 /dev/hda1 / ext3 defaults 1 1 none /dev/pts devpts gid=5,mode=620 0 0 none /dev/shm tmpfs defaults 0 0 /dev/hdc1 /mec ext3 defaults 1 2 none /proc proc defaults 0 0 none /sys sysfs defaults 0 0 /dev/hdb1 /var ext3 defaults 1 2 /dev/hda2 swap swap defaults 0 0
/dev/hdd1 /mnt/idedisk1 auto amconsole,exec,noauto,managed 0 0
which appeared to be the read only device. I cannot umount this device and mount it to /dev/hdd1 ext3 . I would like to use this HDD to store the nackup data. Does any one tell me how to do this?
Have you put a file system on it yet? If not you can use mkfs to do that.
Although I certainly sure that I do already use the fdisk and mkfs on Hdd , I will do it again. But is there any other reason for that for ex. setting jumper on hdd?
After I do 'mkfs' it and set the fstab to ext3 then I reboot the system , the FC3 cannot mount the /dev/hdd1 and the system hang up for a long time [more than 1/2 hour] and it asked for root password to enter the system to recover and I must marked the fstab /dev/hdd1 and then reboot , the server will boot properly but the fstab will then automatically create the line dev/hdd1 /mnt/idedisk1 auto amconsole,exec,noauto,managed 0 0 and the fourth system will be read only and could not be unmount. Amrit
2006/2/21, Bruno Wolff III bruno@wolff.to:
On Mon, Feb 20, 2006 at 23:53:47 +0700, Amrit Angsusingh amritangs@gmail.com wrote:
I used pentium 4 mechine with 256 Mb ram and fedora core 3 for web
server
and four IDE HDDs [160 Gbx4] on both primary ans secondary IDE. The bios also detected all 4 HDD but after starting FC3 the system mount
secondary
slave HDD /dev/hdd1 as /mnt/idedisk1 /dev/hda1 /
ext3 defaults 1 1
none /dev/pts devpts gid=5,mode=620 0 0
none /dev/shm tmpfs
defaults 0 0
/dev/hdc1
/mec ext3 defaults 1 2
none /proc
proc defaults 0 0
none /sys sysfs
defaults 0 0
/dev/hdb1
/var ext3 defaults 1 2
/dev/hda2
swap swap defaults 0 0
/dev/hdd1 /mnt/idedisk1 auto amconsole,exec,noauto,managed
0 0
which appeared to be the read only device. I cannot umount this device
and
mount it to /dev/hdd1 ext3 . I would like to use this HDD to store
the
nackup data. Does any one tell me how to do this?
Have you put a file system on it yet? If not you can use mkfs to do that.
-- Amrit Angsusingh
Thailand
I do the mkfs again and here is the result of mkfs ------------------------------------------- [root@web ~]# umount /backup [root@web ~]# mkfs -V -t ext3 /dev/hdd1 mkfs version 2.12a (Oct 19 2004) mkfs.ext3 /dev/hdd1 mke2fs 1.35 (28-Feb-2004) max_blocks 4294967295, rsv_groups = 0, rsv_gdb = 1024 Filesystem label= OS type: Linux Block size=4096 (log=2) Fragment size=4096 (log=2) 24428544 inodes, 48839600 blocks 2441980 blocks (5.00%) reserved for the super user First data block=0 Maximum filesystem blocks=50331648 1491 block groups 32768 blocks per group, 32768 fragments per group 16384 inodes per group Superblock backups stored on blocks: 32768, 98304, 163840, 229376, 294912, 819200, 884736, 1605632, 2654208, 4096000, 7962624, 11239424, 20480000, 23887872
Writing inode tables: done ext2fs_mkdir: Attempt to read block from filesystem resulted in short read while creating root dir
----------------------- After I do this I cannot mount or fdisk the /dev/hdd1.
[root@web /]# mount -t ext3 /dev/hdd1 /backup1 mount: mount point /backup1 does not exist [root@web /]# mount -t ext3 /dev/hdd1 /backup mount: wrong fs type, bad option, bad superblock on /dev/hdd1, or too many mounted file systems [root@web /]# fdisk /dev/hdd
Unable to read /dev/hdd [root@web /]#
------------------------------------
What happen to me? Any suggestion.
Amrit
Writing inode tables: done ext2fs_mkdir: Attempt to read block from filesystem resulted in short read while creating root dir [root@web ~]#
2006/2/21, Amrit Angsusingh amritangs@gmail.com:
Although I certainly sure that I do already use the fdisk and mkfs on Hdd , I will do it again. But is there any other reason for that for ex. setting jumper on hdd?
After I do 'mkfs' it and set the fstab to ext3 then I reboot the system , the FC3 cannot mount the /dev/hdd1 and the system hang up for a long time [more than 1/2 hour] and it asked for root password to enter the system to recover and I must marked the fstab /dev/hdd1 and then reboot , the server will boot properly but the fstab will then automatically create the line dev/hdd1 /mnt/idedisk1 auto amconsole,exec,noauto,managed 0 0 and the fourth system will be read only and could not be unmount. Amrit
2006/2/21, Bruno Wolff III bruno@wolff.to:
On Mon, Feb 20, 2006 at 23:53:47 +0700, Amrit Angsusingh <amritangs@gmail.com > wrote:
I used pentium 4 mechine with 256 Mb ram and fedora core 3 for web
server
and four IDE HDDs [160 Gbx4] on both primary ans secondary IDE. The
bios
also detected all 4 HDD but after starting FC3 the system mount
secondary
slave HDD /dev/hdd1 as /mnt/idedisk1 /dev/hda1 /
ext3 defaults 1 1
none /dev/pts devpts gid=5,mode=620 0 0
none /dev/shm tmpfs
defaults 0 0
/dev/hdc1
/mec ext3 defaults 1 2
none /proc
proc defaults 0 0
none /sys sysfs
defaults 0 0
/dev/hdb1
/var ext3 defaults 1 2
/dev/hda2
swap swap defaults 0 0
/dev/hdd1 /mnt/idedisk1
auto amconsole,exec,noauto,managed 0 0
which appeared to be the read only device. I cannot umount this device
and
mount it to /dev/hdd1 ext3 . I would like to use this HDD to store
the
nackup data. Does any one tell me how to do this?
Have you put a file system on it yet? If not you can use mkfs to do that.
-- Amrit Angsusingh
Thailand
-- Amrit Angsusingh Thailand
On Tue, Feb 21, 2006 at 14:34:26 +0700, Amrit Angsusingh amritangs@gmail.com wrote:
I do the mkfs again and here is the result of mkfs
Writing inode tables: done ext2fs_mkdir: Attempt to read block from filesystem resulted in short read while creating root dir
You might want to check the cables to make sure they are seated properly. Also you might be able to use smartctl to check for problems with the disk.
On Tue, 2006-02-21 at 11:01 +0700, Amrit Angsusingh wrote:
Although I certainly sure that I do already use the fdisk and mkfs on Hdd , I will do it again. But is there any other reason for that for ex. setting jumper on hdd?
You might check jumper settings on all drives. I have had good luck with modern drives and cables with using cable select for the drive jumpering. Others say that all drives should be jumpered to master/slave even when using the CS cables. If a drive is attached to the end connection of a CS cable (black connector) it must be jumpered to master or CS. Similarly if connected to the middle connector (usually grey) if must be jumpered to slave or CS.
Those connectors are wired as master (end) and slave (middle) so the jumpering must match.
I recommend always using CS for the drive jumpering based on my experience with some problems seen when using master/slave jumpering on the CS cables. YMMV
After I do 'mkfs' it and set the fstab to ext3 then I reboot the system , the FC3 cannot mount the /dev/hdd1 and the system hang up for a long time [more than 1/2 hour] and it asked for root password to enter the system to recover and I must marked the fstab /dev/hdd1 and then reboot , the server will boot properly but the fstab will then automatically create the line dev/hdd1 /mnt/idedisk1 auto amconsole,exec,noauto,managed 0 0 and the fourth system will be read only and could not be unmount. Amrit
This certainly seems like a problem addressing the drive and may be jumper related or drive/cable related.
Jeff Vian wrote:
On Tue, 2006-02-21 at 11:01 +0700, Amrit Angsusingh wrote:
Although I certainly sure that I do already use the fdisk and mkfs on Hdd , I will do it again. But is there any other reason for that for ex. setting jumper on hdd?
You might check jumper settings on all drives. I have had good luck with modern drives and cables with using cable select for the drive jumpering. Others say that all drives should be jumpered to master/slave even when using the CS cables.
Sounds like a good idea to check jumpering. Whatever you do, don't try to mix the jumpering so one drive is jumpered as M/S and the other as CS. I've not had good experience with CS, and am one of those Jeff warned you would say use M/S jumpering.
If a drive is attached to the end connection of a CS cable (black connector) it must be jumpered to master or CS. Similarly if connected to the middle connector (usually grey) if must be jumpered to slave or CS.
Eh? Don't think so.
Those connectors are wired as master (end) and slave (middle) so the jumpering must match.
I recommend always using CS for the drive jumpering based on my experience with some problems seen when using master/slave jumpering on the CS cables. YMMV
[snip]
This certainly seems like a problem addressing the drive and may be jumper related or drive/cable related.
On this we are agreed.
Mike
On Tue, 2006-02-21 at 18:25 -0600, Jeff Vian wrote:
If a drive is attached to the end connection of a CS cable (black connector) it must be jumpered to master or CS. Similarly if connected to the middle connector (usually grey) if must be jumpered to slave or CS.
No, that's not correct. You're oversimplifying the conditions of what must be done in combination.
If you jumper your drives as master and slave, it doesn't matter where they're plugged in. The cable will be ignored by all but broken hard drives.
If you jumper them both as cable select, the cable determines which is which (master on the end, normally).
It is highly inadvisable to jumper one drive as master or slave and the other as cable select, and that's the only time you're going to have to worry about what position on the cable the master or slave jumpered drive is placed.
There is about only thing you do have to do with an 80-wire cable, and that's plug the motherboard end in at the motherboard. It's only at that end that the interspersed ground wires (the extra 40 ones) are grounded.
Fianlly,after setting the jumper I found that the fourth HDD was out of order and when I changed it to the newer one ,I could use it properly. Thanks for every one to help me. Amrit
2006/2/21, Bruno Wolff III bruno@wolff.to:
On Mon, Feb 20, 2006 at 23:53:47 +0700, Amrit Angsusingh <amritangs@gmail.com > wrote:
I used pentium 4 mechine with 256 Mb ram and fedora core 3 for web
server
and four IDE HDDs [160 Gbx4] on both primary ans secondary IDE. The bios also detected all 4 HDD but after starting FC3 the system mount
secondary
slave HDD /dev/hdd1 as /mnt/idedisk1 /dev/hda1 /
ext3 defaults 1 1
none /dev/pts devpts gid=5,mode=620 0 0
none /dev/shm tmpfs
defaults 0 0
/dev/hdc1
/mec ext3 defaults 1 2
none /proc
proc defaults 0 0
none /sys sysfs
defaults 0 0
/dev/hdb1
/var ext3 defaults 1 2
/dev/hda2
swap swap defaults 0 0
/dev/hdd1 /mnt/idedisk1 auto amconsole,exec,noauto,managed
0 0
which appeared to be the read only device. I cannot umount this device
and
mount it to /dev/hdd1 ext3 . I would like to use this HDD to store
the
nackup data. Does any one tell me how to do this?
Have you put a file system on it yet? If not you can use mkfs to do that.
-- Amrit Angsusingh ICT center Saweanpracharuck Hospital Nakornsawan Thailand
Amrit Angsusingh wrote:
I used pentium 4 mechine with 256 Mb ram and fedora core 3 for web server and four IDE HDDs [160 Gbx4] on both primary ans secondary IDE. The bios also detected all 4 HDD but after starting FC3 the system mount secondary slave HDD /dev/hdd1 as /mnt/idedisk1 /dev/hda1 / ext3 defaults 1 1 none /dev/pts devpts gid=5,mode=620 0 0 none /dev/shm tmpfs defaults 0 0 /dev/hdc1 /mec ext3 defaults 1 2 none /proc proc defaults 0 0 none /sys sysfs defaults 0 0 /dev/hdb1 /var ext3 defaults 1 2 /dev/hda2 swap swap defaults 0 0
/dev/hdd1 /mnt/idedisk1 auto amconsole,exec,noauto,managed 0 0
which appeared to be the read only device. I cannot umount this device and mount it to /dev/hdd1 ext3 . I would like to use this HDD to store the nackup data. Does any one tell me how to do this?
First of all, did you partition the drive, and create a file system on it? (What does "fdisk -l /dev/hdd" show?)
If this is a new drive, then it may have a partition on it from the factory. If so, it is probably a vfat partition, and it may not be the entire drive. You will need to unmount the drive, and then use fdisk to change the partition type to type 83, and then run mke2fs on it to create a file ext3 file system. If the partition is not the entire drive, delete the current partition, and create a new one.
Mikkel
On Mon, 2006-02-20 at 23:53 +0700, Amrit Angsusingh wrote:
I used pentium 4 mechine with 256 Mb ram and fedora core 3 for web server and four IDE HDDs [160 Gbx4] on both primary ans secondary IDE. The bios also detected all 4 HDD but after starting FC3 the system mount secondary slave HDD /dev/hdd1 as /mnt/idedisk1 /dev/hda1 / ext3 defaults 1 1 none /dev/pts devpts gid=5,mode=620 0 0 none /dev/shm tmpfs defaults 0 0 /dev/hdc1 /mec ext3 defaults 1 2 none /proc proc defaults 0 0 none /sys sysfs defaults 0 0 /dev/hdb1 /var ext3 defaults 1 2 /dev/hda2 swap swap defaults 0 0
/dev/hdd1 /mnt/idedisk1 auto amconsole,exec,noauto,managed 0 0
which appeared to be the read only device. I cannot umount this device and mount it to /dev/hdd1 ext3 . I would like to use this HDD to store the nackup data. Does any one tell me how to do this?
It seems that the partition is likely not an ext3. It also seems to be mounted by the hot plug manager (it is set up as 'managed' as you can see by the options). Did you just recently install it?
If there are no problems try using fdisk to check/change the partitons to what you want (ext3 I assume) and format the filesystem, then change the line in fstab to reflect it properly and it should work.
On Mon, 2006-02-20 at 23:53 +0700, Amrit Angsusingh wrote:
/dev/hdd1 /mnt/idedisk1 auto amconsole,exec,noauto,managed 0 0
which appeared to be the read only device. I cannot umount this device and mount it to /dev/hdd1 ext3 . I would like to use this HDD to store the nackup data. Does any one tell me how to do this?
A quick guess: You weren't the root user when you tried to unmount it. Try "su -" to become root before unmounting it. Then you can go about partitioning/formatting it, if you need to.
I do shut down the server and found that the cable was set to master and cable select on both primary and secondary IDE as your comment, so I change it to master and slave on both IDE but there is still some error ie. an echo " hdd interupt" and the linux still try to mount the system automatically. After I tried to umount the /dev/hdd1 ,there was some error echo on screen ie. bad sector.... . Is that a physical error on this HDD? [It is seagate 7200 IDE 200 Gb with 5 years warranty]? -- Amrit Angsusingh
Thailand
On Wed, 2006-02-22 at 22:59 +0700, Amrit Angsusingh wrote:
I do shut down the server and found that the cable was set to master and cable select on both primary and secondary IDE as your comment, so I change it to master and slave on both IDE but there is still some error
Huh? A cable isn't set to something, a cable is a cable. A drive can be set to master/slave/single/cable-select, but only one of them (it can't be master *and* cable-select). And what you've described is too vague to really know what you've done.
Two drives on the same cable have to be jumpered appropriately (one as master, the other as slave, or both as cable-select).
Two drives on different cables should both be their own masters, though some models of drive need to be set as a "single" drive, rather than master (i.e. they have no slave), when they'r the only drive on that cable.
Single drives (those not sharing their cable), whether or not there's other drives on the system, are treated as single drives. Sometimes they need setting as a single drive, sometimes they're fine set as a master (which can be by the jumpers, or being connected to the master plug on a cable-select system).
More than two drives requires the ones sharing a cable to be appropriately set up as master or slave in one way or another, and the other drives as appropriate (master and slave if sharing, but if you only have one other drive it should be set up as a single or master drive).
Debugging: Try connecting each drive by itself, and check that you're drives are working properly. Also check both cables individually. Check that your jumpers aren't broken (there should be metal contacts inside the plastic plug). Check the power leads to each drive are working. These days you should have a power supply big enough to handle multiple drives, but it's still possible that you don't.
-
Amrit Angsusingh -
Bruno Wolff III -
Jeff Vian -
Mike McCarty -
Mikkel -
Tim