10-07-2005
Well, I'm not sure. I have been looking at the source code for
format. It looks like the format program issued a SCSI Reserve command to the disk via a uscsi ioctl() call and this failed. Previous to this, the format program issued a SCSI Inquiry command which succeeded. Looking the
SCSI-2 Standard, it does not look like the Reserve command will actually do I/O. And you can do an Inquiry while another Inititator has a device reserved.
So I am guessing that somehow the system thinks that the drive is reserved. This could be legitimate if the SCSI chain connected to two hosts in a cluster or something like that. Maybe the drive just is in a confused state... a power cycle is susposed to reset any reservations. Maybe the drive is broken is some way that makes it always look reserved. Few of us actually ever use the "format" menu item in the format command, so maybe it has a bug.
But I would try some other stuff before I conclude that the drive is bad.
8 More Discussions You Might Find Interesting
1. Linux
How do you debug a perl script non interactively, similar to bash -x? (1 Reply)
Discussion started by: marcpascual
1 Replies
2. Solaris
We have several containers on one machine and would like to reserve some memory for the global zone. capped-memory only allows max physical/swap and setting a max on each container isn't an option. The server has 32GB physical and 30GB swap. Currently there are ten containers on it. Normally... (6 Replies)
Discussion started by: kharjahn
6 Replies
3. Filesystems, Disks and Memory
Hi All,
I want to make a 3GB of space reserve on Solaris. Let me know whether there is a way by creating empty file of 3GB so that i can delete that file in future to utilize that space. Or any other better ways for space reserve.
-Vinodh' Kumar (4 Replies)
Discussion started by: vino_hymi
4 Replies
4. Shell Programming and Scripting
I have a shell script which sets some variables and then calls modules of a program in succession, one by one. Problem is that the script is executed on servers with many users, so sometimes the script starts running, runs for 10 minutes and then breaks due to lack of resources when other users run... (1 Reply)
Discussion started by: tetreb
1 Replies
5. Solaris
Hi,
i got the following error when i tried to access the cygwin x server from a windows XP PC.
"xdmcp fatal error session failed session 23 failed for display"
Alternatively, when i tried to access the same Cygwin X Server from another windows XP PC which is on a different LAN... (3 Replies)
Discussion started by: HarishKumarM
3 Replies
6. AIX
Hi
I have a ticket to Reserve ports for SAS install in SAS servers.
Does anyone knows the procedure how to do it.
it will be helpful for me please
Thanks in Advance (4 Replies)
Discussion started by: gulamibrahim
4 Replies
7. UNIX for Advanced & Expert Users
my apps use port 40001; however, for example, firstly, I ftp to other server, it made a high port locally, remote is port 21, unfortunately, it hit my port 40001 and my apps is unable to startup. This chance is very very little, but I hit it. Can resevse my port 40001? otherwise command don't use it (5 Replies)
Discussion started by: goodbid
5 Replies
8. Solaris
I have two servers with a fresh install of Solaris 11, and having problems when doing rpcinfo between them. There is no firewall involved, so everything should theoretically be getting through. Does anyone have any ideas? I did a lot of Google searches, and haven't found a working solution yet.
... (2 Replies)
Discussion started by: christr
2 Replies
SD(4) Linux Programmer's Manual SD(4)
NAME
sd - driver for SCSI disk drives
SYNOPSIS
#include <linux/hdreg.h> /* for HDIO_GETGEO */
#include <linux/fs.h> /* for BLKGETSIZE and BLKRRPART */
CONFIGURATION
The block device name has the following form: sdlp, where l is a letter denoting the physical drive, and p is a number denoting the parti-
tion on that physical drive. Often, the partition number, p, will be left off when the device corresponds to the whole drive.
SCSI disks have a major device number of 8, and a minor device number of the form (16 * drive_number) + partition_number, where drive_num-
ber is the number of the physical drive in order of detection, and partition_number is as follows:
partition 0 is the whole drive
partitions 1-4 are the DOS "primary" partitions
partitions 5-8 are the DOS "extended" (or "logical") partitions
For example, /dev/sda will have major 8, minor 0, and will refer to all of the first SCSI drive in the system; and /dev/sdb3 will have
major 8, minor 19, and will refer to the third DOS "primary" partition on the second SCSI drive in the system.
At this time, only block devices are provided. Raw devices have not yet been implemented.
DESCRIPTION
The following ioctls are provided:
HDIO_GETGEO
Returns the BIOS disk parameters in the following structure:
struct hd_geometry {
unsigned char heads;
unsigned char sectors;
unsigned short cylinders;
unsigned long start;
};
A pointer to this structure is passed as the ioctl(2) parameter.
The information returned in the parameter is the disk geometry of the drive as understood by DOS! This geometry is not the physical
geometry of the drive. It is used when constructing the drive's partition table, however, and is needed for convenient operation of
fdisk(1), efdisk(1), and lilo(1). If the geometry information is not available, zero will be returned for all of the parameters.
BLKGETSIZE
Returns the device size in sectors. The ioctl(2) parameter should be a pointer to a long.
BLKRRPART
Forces a reread of the SCSI disk partition tables. No parameter is needed.
The SCSI ioctl(2) operations are also supported. If the ioctl(2) parameter is required, and it is NULL, then ioctl(2) will fail
with the error EINVAL.
FILES
/dev/sd[a-h]: the whole device
/dev/sd[a-h][0-8]: individual block partitions
COLOPHON
This page is part of release 3.44 of the Linux man-pages project. A description of the project, and information about reporting bugs, can
be found at http://www.kernel.org/doc/man-pages/.
Linux 2012-05-03 SD(4)