Sponsored Content
The Lounge War Stories Do you trust your users to follow your instructions? Post 302940916 by edfair on Friday 10th of April 2015 12:25:37 AM
Old 04-10-2015
Do you trust your users to follow your instructions?

This happened a long time ago and some of the details may not be exact. Customer had obsolete hardware running an obsolete SCO OS and some type of database program with data scattered around the system. There were 2-1g SCSI drives, both split in half, with the 3 filesystems automatically loading on boot.

The non boot hard drive upchucked and went out to data recovery while I replaced the hard drive, partioned it, and created the filesystems awaiting Monday morning and left with instructions to not attempt a restore of the data if it came in.

When I got there on Monday the data had been restored. You can imagine that their data was totally corrupted. Some parts were good, some parts were bad, and they had no understanding of how it happened. Restoring to a system without the attached filesystem dropped the stuff to the assumed proper place on the root drive but only those files that would fit. And as the drive filled up less and less would fit.

The person who did the restore told me that the owner of the company had told him to restore it in spite of my instructions not to.

It ended up in the court system, my side to get reimbursed for some time and parts, his counter suit for $40,000 for the work to recover his data, later reduced to $25,000 so it would stay in small claims court. The second judge to hear it suggested that we kiss and make up since it was going to cost both of us more to proceed than we would win.

I understood what had happened, had insisted on backups that would have allowed full recovery, but wasn't interested in dealing with them any more.
 

7 More Discussions You Might Find Interesting

1. Shell Programming and Scripting

ssh2-keygen trust issue

I have two systems SysA & SysB having the same userid sharing the home directory via NFS mount. I need to know the steps to setup ssh trust between these two systems given that both share the home dir. I have tried all the steps to generate the keys & then creating identification &... (2 Replies)
Discussion started by: deo_kaustubh
2 Replies

2. Solaris

configuring user as trust

Hi Gurus, Got another issue. I am trying to configure a user to run some scripts through trusted user where in while logging to remote system it shouldn't ask for password. i did following to get it working but its not working i create a private and public key with the below... (1 Reply)
Discussion started by: kumarmani
1 Replies

3. HP-UX

Not Trust Host 10.10.10.10

I get a message similar to this, in the syslog file. Actually, I am trying to let the host at 10.10.10.10 access the HP-UX system. How do I get it trusted? Thanks! (2 Replies)
Discussion started by: instant000
2 Replies

4. UNIX for Dummies Questions & Answers

ssh trust issue

Hi, i am setting up ssh trust setup between two servers where SVRA is a solaris box and SVRB is a Red Hat Linux. It is asking for the password all the time. I have copied over the SVRA:/home/nagios/.ssh/id_dsa.pub as authorized_keys on to SVRB:/dat01/home/nagios/.ssh/ -bash-3.00$ ssh -vvv... (4 Replies)
Discussion started by: uxadmin007
4 Replies

5. Linux

SCEP and Trust Anchor

Hi Does anybody knows about the simple certificate enrollment protocol details ? if yes please provide me the details. And what is a trust anchor profile ? Thanks in advance. (0 Replies)
Discussion started by: chaitus.28
0 Replies

6. Shell Programming and Scripting

SSH Trust Testing

Hi, I want test the ssh trust between two host. It works fine if the trust is working fine but if the trust is not working fine it gets stuck. #!/bin/sh >/users/test/ssh.txt for i in `cat /users/test/host.txt`; do ssh test@$i uname -a >> /users/test/ssh.txt test=`cat... (0 Replies)
Discussion started by: Abhayman
0 Replies

7. What is on Your Mind?

Twitter Users: Follow the Forums on Twitter

Hey Twitter Users, You can follow the forums on Twitter: https://twitter.com/unixlinux @unixlinux Current Twitter Stats: TWEETS 76.4K FOLLOWERS 54.3K Comments or questions? Please post below. (1 Reply)
Discussion started by: Neo
1 Replies
rzdisk(8)						      System Manager's Manual							 rzdisk(8)

Name
       rzdisk - SCSI disk maintenance utility

Syntax
       /bin/rzdisk
       /bin/rzdisk -c [ ask ] special
       /bin/rzdisk -d [ bfi | sector | block ] special
       /bin/rzdisk -f [ vendor | known ] special
       /bin/rzdisk -g [ current | saved | default | changeable ] special
       /bin/rzdisk -h
       /bin/rzdisk -i special
       /bin/rzdisk -r LBN special
       /bin/rzdisk -s LBN length special

Description
       The  utility,  a  SCSI  disk maintenance program, formats a SCSI disk, scans a SCSI disk for bad blocks, and reassigns bad blocks on a SCSI
       disk.

       The special file argument is a raw device pathname.  You must specify an unmounted c partition of a  character  device  special	file  (for
       example, ).

Options
       -c     Changes disk parameters.

	      If the ask argument isn't specified, the disk parameters are set to the default values.  When the ask argument is specified, you are
	      interactively asked about each field marked as changeable in each page having changeable parameters.  In both cases, you	are  asked
	      whether  to  save  the parameters on disk.  If you answer 'yes', the parameters are saved on the disk which also updates the current
	      parameters.  Otherwise, only the current parameters are updated.	You must save the parameters if you want them to be preserved when
	      the system is power cycled.  Also note, the direct-access device format parameters (Page 3) and the rigid disk drive geometry param-
	      eters (Page 4) only get saved after formatting the drive, regardless of whether you said 'yes' to saving the parameters.

       -d     Reads defect lists from the disk.

	      This option allows you to get defect lists from the disk in either bfi (bytes from index), sector, or block, formats.

       -g     Gets parameters from the disk.

	      This option allows you to get either current, saved, default, or changeable parameters from the  disk.   The  program  displays  the
	      parameters of all pages supported by the drive.

	      The changeable parameters are those fields which can be changed.	Any field displayed with a '1' value in it can be changed.

	      The default parameters are those parameters supplied by the vendor.

	      The current parameters are those parameters which you are currently running with.  These parameters are setup from the saved parame-
	      ters when the drive is powered on.

	      The saved parameters are those parameters which have been saved on the disk.  These parameters get used each time the drive is  pow-
	      ered on.

       -f     Formats a SCSI hard disk (see Restrictions).

	      Since  Digital ships the SCSI disk already formatted, only use this option if you have encountered a serious problem and must refor-
	      mat the disk.  You can format a disk with the vendor (manufacturer) defect list or with the known (vendor and  grown)  defect  list.
	      The  grown  defect list contains any blocks that may have been reassigned during the life of the SCSI disk drive.  When formatting a
	      disk, you must specify the raw device pathname.

	      The following example reformats the disk on drive 3 using the known defect list:
	      /bin/rzdisk -f known /dev/rrz3c

	      The -f option with no additional arguments formats a SCSI floppy diskette (see Restrictions).  New floppy diskettes are usually  not
	      formatted.  You need to format each diskette before you can store data on it.

	      The following example shows how to format a floppy diskette in RX23 drive one:
	      /bin/rzdisk -f /dev/rrzlc
	      The floppy format operation is interactive.  The program will guide you through formatting the diskette.

       -i     Requests inquiry data from the disk.

	      This  option  causes  the  program  to dump all inquiry data returned from the disk.  The information dumped includes the peripheral
	      device type, the device type qualifier, the removable medium field, the version number, vendor identification, product  information,
	      and the firmware revision level.

       -h     Calls the HELP menu to the screen.

       -r     Reassigns a bad block on the disk (see Restrictions).

	      When  reassigning  a bad block, you must specify the LBN, which is a unique number (decimal notation) that represents the disk block
	      as reported in the errorlog file, and the raw device pathname.

	      The following example reassigns block 222658 on the c partition of drive 3:
	      /bin/rzdisk -r 222658 /dev/rrz3c

	      The program reads the specified LBN prior to reassigning the block.  If reads valid data from the block, then the block is not actu-
	      ally bad or the SCSI driver already reassigned the block.  In this case, asks if the reassignment should be canceled.  Answer yes to
	      cancel the reassignment.	 This prevents double reassignment and replacement of good blocks.

       -s     Scans for bad blocks on a specified area of the disk.

	      When scanning a disk, you must specify the LBN, which is a unique number (decimal notation) that represents the disk block  relative
	      to the start of the partition, the length, and the raw device pathname of the partition to scan.

	      To  start  scanning  from  the  first block of the specified partition, use the number 0 to represent the LBN.  When the number 0 is
	      specified, the scan starts at the first block of the specified partition.

	      The length is a decimal number that indicates how many 512-byte blocks to scan.  To scan up to and including the last block  of  the
	      specified  partition,  use the number -1 to represent the length.  By specifying the length, you define the scope of the scan within
	      the identified partition.

	      The following example scans the first ten blocks of the entire disk (c partition) on drive 3:
	      /bin/rzdisk 0 10 -s /dev/rrz3c
	      The following example scans the entire disk (c partion) on drive 3:
	      /bin/rzdisk 0 -1 -s /dev/rrz3c

Diagnostics
       The program generates messages when the user is not privileged, when the LBN is not in the specified partition, and when the length exceeds
       the size of the partition.

Restrictions
       You must have super-user privileges to run the program.

       You should not have to format your system disk.

       The system should be in single-user mode and the file systems on the disk should be unmounted when running the program, except when format-
       ting floppy diskettes.

       Use the -f option with caution and only if the SCSI hard disk drive seems corrupted.  Be aware that when you format a  disk,  all  resident
       data is destroyed.

       Digital	supports  formatting,  writing, and reading of High Density (HD) 3.5 inch diskettes in the RX23 and RX26 disk drives, or 5.25 inch
       diskettes in the RX33 disk drive.

       Digital supports reading, but not formatting or writing, of Double Density (DD) 3.5 inch diskettes in the RX23 and RX26 disk drives or 5.25
       inch  diskettes	in the RX33 disk drive.  Reliable reading of DD diskettes requires they be written only on a double density drive and have
       not been overwritten by an RX23, RX26, RX33, or other high density drive.  This restriction occurs because  of  differences  in	the  write
       heads between DD and HD drives.	Data written by a DD drive cannot be completely overwritten by a HD drive.

       This same restriction applies to the RX33 5.25 inch diskettes when using low density (48 TPI - Tracks Per Inch) diskettes.  Diskettes over-
       written on different machines may suffer read failures from reduced signal-to-noise ratios.  So long as a diskette is formatted and written
       on a single machine, there should be no problem.

       The -r option is supported only with those SCSI disks that support the reassign block command.

See Also
       dkio(4), rz(4), chpt(8), mount(8), restore(8), uerf(8)
       Guide to the Error Logger

																	 rzdisk(8)
All times are GMT -4. The time now is 09:35 PM.
Unix & Linux Forums Content Copyright 1993-2022. All Rights Reserved.
Privacy Policy