Veritas netbackup sg driver failed to attach

Hi,
I am installing veritas netbackup on my V245 server (solaris 10).
After the installation I am trying to do sg.build and sg.install, but I am getting an error message
Warning: Driver (sg) successfully added to system but failed to attach
No Scsi devices on your system
the solaris sgen driver is already loaded, and when i do mt -f /dev/rmt/1, i can see the tape drive detected
and on cfgadm -a lso shows the drive and the changer
i have a C2 tape autoloader attached to my machine
If i try to manullay load the driver by modload /kernel/drv/sparcv9/sg, It fails saying
sg: WARNING: sg: _init failed on linkage 6
No Scsi devices attached to the system
Please help me in resolving this issue
Thanking you
Ushas Symon

# ./sg.install
Copied files to /kernel/drv to /kernel/drv/sparcv9.
Doing add_drv of the sg driver
devfsadm: driver failed to attach: sg
Warning: Driver (sg) successfully added to system but failed to attach
WARNING: /usr/sbin/add_drv failed.
There may be no SCSI devices on this machine
root@bkp-server # cfgadm -a
c3 scsi-bus connected configured unknown
c3::rmt/1 tape connected configured unknown
c3::scsi/changer/c3t5d1 med-changer connected configured unknown

Similar Messages

Backup MaxDB 7.7.06.09 to Veritas Netbackup 6.5 fails

We want to backup MaxDB to Veritas Netbackup server.
According documentation found about this topic the backup is configured
by creating 3 files: bsi.env, initWDO.par (for backint for MaxDB) and
initWDO.utl (for backint for Oracle). The contents of these files are shown below.
When running the backup, it seems that backint for MaxDB ends with return code 2,
the error message received is:
-24920,ERR_BACKUPOP: backup operation was unsuccessful
The backup tool failed with 2 as sum of exit codes.
Unfortunately no further errors/warnings are shown that points in any direction
what's going wrong here.
We start the backup via the following dbmcli-command:
dbmcli -u control,****** -d WDO -uUTL control,****** backup_start BACKData_WDO
(for the time being without xuser entries, first make sure the backup is ok)
The files dbm.ebl, dbm.prt and several others do not show any additional details/hints.
It seems as if the "save data quick" to the named pipe is ok, the sql-code returns "0".
After this the backint for MaxDB should read the named pipe and write it to stage-files,
but backint (for MaxDB) ends with return code 2.
Any idea what might go wrong here?
With kind regards,
R. Hoogeveen
The Netherlands
File bsi.env:
=========
BACKINT                /sapdb/WDO/db/bin/backint
INPUT                  /tmp/backint4maxdb_WDO.in
OUTPUT                 /tmp/backint4maxdb_WDO.out
ERROROUTPUT            /tmp/backint4maxdb_WDO.err
PARAMETERFILE          /sapdb/data/wrk/WDO/initWDO.par
TIMEOUT_SUCCESS        600
TIMEOUT_FAILURE        300
ORIGINAL_RUNDIRECTORY /sapdb/data/wrk/WDO
File initWDO.par:
=============
STAGING AREA:                  /tmp/stageWDO 1 GB
FILES PER BACKINT CALL:        1
BACKINT:                       /usr/openv/netbackup/bin/backint
PARAMETERFILE OF BACKINT:      /sapdb/data/wrk/WDO/initWDO.utl
HISTORY FILE:                  /sapdb/data/wrk/WDO/backint_history_WDO.log
INPUTFILE FOR BACKINT:         /tmp/backint4ora_WDO.in
OUTPUTFILE FOR BACKINT:        /tmp/backint4ora_WDO.out
ERRORFILE FOR BACKINT:         /tmp/backint4ora_WDO.err
MAXIMAL DELAY OF BACKINT CALL: 30
File initWDO.utl:
=============
switch_list /tmp/.switch_WDO.lis
switch_sem /tmp/.switch_WDO.sem
switch_log /tmp/.switch_WDO.log
drives 1
client wdocsz01
server backfire
policy sap_WDO

Lars,
The /tmp/backint4maxdb.err file tells me:
Backint for Oracle was unsuccessful.
(Unfortunately that was all information in this file.)
The backint4ora.in file tells me:
/tmp/stageWDO.0
(looks ok to me, also at the time backint4ora was started the stage file was created with non-zero size)
The backint4ora.out file tells me:
ERROR: incorrect format or option in .utl file (-p)
(this seems as if the -p option is not recognized by the backint4ora executable)
When I run this executable manually as user sdb it shows me the commandline that I should use, including the -p option.
One strange thing I notice is that the order of the options is different from the commandline that is executed.
(in the usage it shows -p as almost last parameter, the command executed has -p in the middle; I will see if I can execute the backint4ora command manually, since I have the stage file available, I have copied it during the process)
The backint4ora.err file is empty.
The commandline of the dbmcli command I execute shows me another interesting error message,
this might also be the source of all problems here.
-24920,ERR_BACKUPOP: backup operation was unsuccessful
Could not get external backup ID's from the backup tool.
I tried to retrieve the external backup id's manually for medium BACKDataWDO, it show me OK without any external backup id's.
I am not sure, but I think this is correct since no successfull backup was created yet with external backup tool.
Can it also be that the parameter "history file" from the initWDO.par file is set wrong?
Currently it is set to "/sapdb/data/wrk/WDO/backint_history_WDO.log" (see above also), but I can imagine
that this should be set to the dbm.knl or dbm.ebf file ??
With kind regards,
Roger Hoogeveen

64-bit driver fails to attach in Sol-9

Hi,
I have a driver which I can compile, install and use succesfully on sparc (Enterprise 450) in Sol 7 and 8 in both 32 and 64-bit kernels (it also runs in 2.6 and in i386 2.6, 7 and 8 32-bit modes but never mind).
Sol-9 32-bit is also OK but with 64-bit, when I run add_drv, it loads (I can see cmn_err() messages from my ini() routine) but it fails to probe or attach. I see fini() traces too but nothing from my probe() or attach. devfsadm returns error 11.
If I do a simple modload, then I can see the init() traces so I assume it has linked OK. modinfo shows the driver is loaded. modunload removes it and I then see the fini() traces.
Any ideas?
Just to cover the bases mentioned elsewhere in this forum:
I am using Forte C 6U2 with -xarch=v9 -D_KERNEL -xO3
I am putting the driver into /platform/sun4u/kernel/drv/sparcv9 and I have an entry in /etc/devlink.tab.
I have tried putting it into /usr/kernel/drv/sparcv9 to no avail
I use the following to load it:
add_drv -m"* 0666 root root" -i'"pci1011,1065"' e8k
I don't have a driver.conf file as it is PCI (and none of the other platforms need it)
There's nothing interesting in /var/adm/messages apart from my cmn_err() traces.
In fact I am using exactly the same methodology and build/install scripts that work so well in Sol-7/8 64-bit.
TIA
Bob

Did you had a look at the "compatible:" property in the "prtconf -pv" output > for your pci device? Does it contain the string 'pci1011,1065' ? Yes, it does - here is the entry from prtconf -pv (I also tried the secondary address EACA,8000 with add_drv - which works on Sol-i386 - but no progress):
Node 0xf0096d10
assigned-addresses: 82001010.00000000.00001000.00000000.00000080.81001014.00000000.00000400.00000000.00000080.c2001018.00000000.00040000.00000000.00040000
reg: 00001000.00000000.00000000.00000000.00000000.02001010.00000000.00000000.00000000.00000080.01001014.00000000.00000000.00000000.00000080.42001018.00000000.00000000.00000000.00040000
compatible: 'pcieaca,8000' + 'pci1011,1065' + 'pciclass,0b4001' + 'cpu'
name: 'cpu'
fast-back-to-back:
devsel-speed: 00000001
class-code: 000b4001
interrupts: 00000001
subsystem-vendor-id: 0000eaca
subsystem-id: 00008000
max-latency: 00000000
min-grant: 00000000
revision-id: 00000004
device-id: 00001065
vendor-id: 00001011
The prtpci script gives:
pci, instance #4 (driver name: pcipsy)
Driver properties:
name='interrupt-priorities' type=int items=6 dev=none
value=0000000e.0000000e.0000000e.0000000e.0000000e.0000000e
name='virtual-dma' type=int items=2 dev=none
value=f0000000.10000000
Register Specifications:
Bus Type=0x1cc, Address=0x4000, Size=0x2000
Bus Type=0x1cc, Address=0x1000000, Size=0x100
Bus Type=0x1cc, Address=0x0, Size=0xd000
cpu (driver name: cpu)
Thanks for the prtpci script - I will study it and maybe I can learn something new about prtconf!
Cheers
Bob

Add_drv fails to attach: probe not called after init

i am trying to add_drv a simple pseudo-device driver. i have basically
copied the pio sample driver code available elsewhere on this website.
i've compiled using forte c5.0, and am trying to load on an ultra 5 running solaris 8.
i have tried loading both 32 bit and 64 bit versions.
as this is a pseudo driver (there is no real device yet... i just want to test the load/unload stuff), i have an xxprobe routine which always returns DDI_PROBE_SUCCESS.
let's say the driver executable is named relidrv, which i copied to /usr/kernel/drv or drv/sparcv9 (depending on 32 vs 64 bit'ness of the build)
i have a simple relidrv.conf which simply contains name="relidrv", parent="pseudo" and instance=0.
then, as root:
add_drv -v relidrv
gives me:
exit status = 0
devfsadm: driver failed to attach: relidrv
exit status = 11
Warning: Driver (relidrv) successfully added to system but failed to attach
i put some cmn_err(CE_WARN,...) calls in init(), probe(), and attach(). the init() call returns 0, all well and good. however, according to my cmn_err output, probe is never called, neither is attach.
suggestions?

in a previous thread by murraystokely, the answer appears. a very big thanks to both mause and murray...
the problem was that the driver itself lived in /usr/kernel/drv/sparcv9, but the conf file should have gone into /usr/kernel/drv.
hint for whoever writes the next revision to the Writing Device Drivers manual... mention this!

SCSI target driver won't attach

I am writing a SCSI target driver for a data acquisition system with Solaris 8 X86. When I try the add_drv command: add_drv ds21 I get the following errors:
devfsadm: driver failed to attach: ds21
Warning: Driver (ds21) successfully added to system but failed to attach
I probably need to include an alias with the add_drv command, but, I have used the prtconf -vp command to list the devices names however, none of the devices tied to the SCSI bus show up on the list. Any suggestions?

I think that you are missing entries in your driver.conf file for the devices that your driver is controlling. See the bst sample driver.
If you have just one scsi bus, it should take the form
name="ds21" class="scsi" target=t lun=0;
where t is the SCSI Target ID
If you have more than one scsi bus, it should take the form
name="sd21" parent="/sbus@1,f8000000/esp@0,8000" target=t lun=0;

My driver is not attached!!!

I'm trying to modify HBA driver source
I compiled my modified source and using add_drv command I tried to install compiled driver.
but the following messages were displayed in monitor
devfsadm : driver failed to attach : testglm
Warning : Driver successfully added to system but failed to attach
driver installed
What is the problem???

Thank you so so much for your answer.
I used command add_drv in add_drv -i 'pci1000,f' testglm
but I couldn't attach my driver.
My testglm steps from original glm source
and glm driver is automatically loaded when system boot
So When I add my testglm driver to system , Is is conflicted with operating glm driver???
The following messages are the result of command prtconf -vp
Node 0xf0068468
bus-range: 00000080.00000080
scsi-initiator-id: 00000007
compatible: 'pci108e,5000' + 'pciclass,060400'
66mhz-capable:
fast-back-to-back:
devsel-speed: 00000001
class-code: 00060400
revision-id: 00000013
device-id: 00005000
vendor-id: 0000108e
interrupt-map: 00800800, ......
interrupt-map-mask: 00fff800.00000000.00000000.00000007
#interrupt-cells: 00000001
slot-names: 0000001c.70636961.20736c6f.74203200.70636961.20736c6f.74203300.70636961.20736c6f.74203400
no-probe-list: '0'
clock-frequency: 01f78a40
model: 'SUNW,simba'
#address-cells: 00000003
#size-cells: 00000002
reg: 00000800.00000000.00000000.00000000.00000000
device_type: 'pci'
name: 'pci'
Node 0xf0088b74
latency-timer: 00000011
assigned-addresses: 81800810.00000000.00000400......
device_type: 'scsi-2'
clock-frequency: 02625a00
reg: 00800800.00000000.00000000......
model: 'Symbios,53C875'
compatible: 'glm'
name: 'scsi'
devsel-speed: 00000001
class-code: 00010000
interrupts: 00000001
max-latency: 00000040
min-grant: 00000011
revision-id: 00000014
device-id: 0000000f
vendor-id: 00001000
Node 0xf008c704
device_type: 'block'
compatible: 'sd'
name: 'disk'
Node 0xf008dd50
device_type: 'byte'
compatible: 'st'
name: 'tape'
Node 0xf008e9f0
latency-timer: 00000011
assigned-addresses: 81800910.00000000.00000800......
device_type: 'scsi-2'
clock-frequency: 02625a00
reg: 00800900.00000000.00000000.......
model: 'Symbios,53C875'
compatible: 'glm'
name: 'scsi'
devsel-speed: 00000001
class-code: 00010000
interrupts: 00000001
max-latency: 00000040
min-grant: 00000011
revision-id: 00000014
device-id: 0000000f
vendor-id: 00001000
Node 0xf0092580
device_type: 'block'
compatible: 'sd'
name: 'disk'
Node 0xf0093bcc
device_type: 'byte'
compatible: 'st'
name: 'tape'

Hard drive failing on Macbook 13" OS 10.7.3 Need to xfer all files to an iMAC 21.5. How to do this?

Hard drive failing on Macbook 13" OS 10.7.3 Need to xfer all files to an iMAC 21.5. How to do this?

Can you boot the MacBook into FireWire Target Disk mode? That turns it (temporarily) into an expensive hard drive. You can then move files.
Another way would be to use an external hard drive, attach it to the MacBook and user software such as Bombich's Carbon Copy Cloner to copy the hard drive (assuming you have sufficient space on the iMac.)
Best of luck. Please keep your data backed up in the future to avoid this. It'll make your life easier.

Pseudo driver: Failed to process interrupt

I wrote a very basic, hardware-less device driver. It is similar to the ramdisk example. I'm using 64-bit Solaris 8. The device name is "pkb" and the file pkb.conf (in /usr/kernel/drv/) has the following line:
name="pkb" parent="pseudo" instance=0;
When I execute add_drv -i "pkb" /usr/kernel/drv/sparcv9/pkb I get the following error message on the console:
unix: WARNING: Failed to process interrupt for pkb0 due to down-rev of nexus driver pseudo0
The device does however get an entry the /dev directory (/dev/pkb -> ../devices/pseudo/pkb@0:pkb)
When I execute an open() and then an ioctl() on this new device, ioctl returns a -1 yet the operation appears to work (I get the value I requested). errno = 3 after ioctl() fails. Subsequent ioctl()s do not work at all and errno = 22;
So, after spending many days on the web AND paying for Sun support, I fail to find a reference to the message "unix: WARNING: Failed to process interrupt ...."
Any ideas?

Hi,
Changes were made in Solaris 8 to the way bus nexus drivers process addition of interrupt handlers on sun4u. Consequently there is a check in the code to ensure the version of the nexus driver being called is at the correct level. If its not, you get the message you report.
The pseudo nexus driver doesnt support interrupts, and the check fails.
(If you like playing with kadb, the message comes from within i_ddi_intr_ctlops, which is called by (i_) ddi_add_intr.)
Have you checked the return status from ddi_add_intr? Normally if this fails you driver should then fail the attach, returning DDI_FAILURE.
To test an interrupt handler, you'll need to write a driver for a piece of hardware.
Its not obvious to me what your ioctl routine is doing. Can you post the code for it?
Hope that helps a bit.
Ralph
SUN Developer Technical Support

I have burned CDs before from 2gb to 7gb using different types of discs. Now I get this error message Apple error The disc can't be burned because communication between the computer and the disc drive failed (error code 0x80020022). What should I do

I have burned CDs before from 2gb to 7gb using different types of discs. Now I get this error message Apple error The disc can't be burned because communication between the computer and the disc drive failed (error code 0x80020022). What should I do

Reset the SMC, as per > Resetting the System Management Controller (SMC)
1. Shut down the computer.
2. Unplug the computer's power cord.
3. Wait fifteen seconds.
4. Attach the computer's power cord.
5. Wait five seconds, then press the power button to turn on the computer.
Then zap the PRAM/NVRAM as per > About NVRAM and PRAM
1. Shut down the computer.
2. Locate the following keys on the keyboard: Command, Option, P, and R. You will need to hold these keys down simultaneously in step 4.
3. Turn on the computer.
4. Press and hold the Command-Option-P-R keys. You must press this key combination before the gray screen appears.
5. Hold the keys down until the computer restarts and you hear the startup sound for the second time.
6. Release the keys.
Then retry the Disc Drive.

Duplication with Veritas netbackup 4.5

Alright. This is driving me crazy. How can you clone a database using Veritas? Right now we are using TSM and are switching to Veritas. With TSM, I can run the duplication from either the target or the auxiliary. With Veritas it states you must run it on the auxiliary. One problem with this is if your target does not have a password file. I can't bounce a production db just to create a temp password file.
Another issue is with the username. I understand you have to restore logged in as the user who created the backup. We have 28 different servers with 28 different oracle users so to do this will be an administrative hastle. With TSM you can pass rman an options file with the username in there.
If there's any documention other than in Veritas' manual under "Redirected Restore" I'd love to see it. Veritas' documentation doesn't use the RMAN duplicate command. It's almost as if they expect the exact same directory structure and user on the aux server.
The error that is killing me is the:
Failed to open for restore
I know the backup is there on tape, I just can't get permissions to it and don't know how.
Thanks

Hi 152124
Once you have the catalog created you can begin to play with RMAN. You could backup the whole database and backup it partially and so on. I recommend you to look for the manuals for it.
Regarding Veritas NetBackup , I have not has experienced with that. I have always worked backing up the database directly to file systems for after move them to tapes.
If you have any doubt during your learning you can ask here in this forum.
Joel Pérez
http://otn.oracle.com/experts

Configure MaxDB with Veritas Netbackup

Hi team,
    When i configure Veritas Netbackup with MaxDB I have the following queries:
1. should i create a staging file (will it get created automatically) ?
2. Will all the log, error & History files get created automatically?
3. Should i use PIPES only to take backup? can't i use FILE backup?
please help
Thanks in advance
Regards,
Pradeep Chandru.

Hi,
I have made the log files to be generated. Now i am getting the following error:
dbm.prt
2006-09-04 16:07:15 0x00002624 INF        283 DBMSrv   command backup_save "test01" DATA RECOVERY
2006-09-04 16:08:22 0x00002624 ERR     -24920 DBMSrv   ERR_BACKUPOP: backup operation was unsuccessful
                    0x00002624 ERR     -24778 DBMSrv   The backup tool failed with 2 as sum of exit codes. The database request was canceled and ended with error -903.
2006-09-04 16:32:50 0x00002624 INF        226 DBMSrv   DBM Server client disconnect: PID 7012 on computer xxxxxxxx.local
dbm.ebl
2006-09-04 16:07:49 The database is working on the request.
    2006-09-04 16:08:09 The database is working on the request.
    2006-09-04 16:08:20 Canceling Utility-task after a timeout of 60 seconds elapsed ... OK.
    2006-09-04 16:08:21 The database has finished work on the request.
    Receiving a reply from the database kernel.
    Got the following reply from db-kernel:
        SQL-Code              :-903
The backup operation has ended.
2006-09-04 16:08:21
Filling reply buffer.
    Have encountered error -24920:
        The backup tool failed with 2 as sum of exit codes. The database request was canceled and ended with error -903.
    Constructed the following reply:
        ERR
        -24920,ERR_BACKUPOP: backup operation was unsuccessful
        The backup tool failed with 2 as sum of exit codes. The database request was canceled and ended with error -903.
Reply buffer filled.
2006-09-04 16:08:22
Cleaning up.
    Copying output of Backint for MaxDB to this file.
dbm.prt
(established at 2006-09-04 15:31:14): PID 11028 on computer xxxxxxxx.local
2006-09-04 15:31:14 0x000026f8 INF        283 DBMSrv   command backup_start BACKt09 DATA
2006-09-04 15:32:20 0x000026f8 ERR     -24920 DBMSrv   ERR_BACKUPOP: backup operation was unsuccessful
                    0x000026f8 ERR     -24778 DBMSrv   The backup tool failed with 2 as sum of exit codes. The database request was canceled and ended with error -903.
2006-09-04 15:32:20 0x000026f8 INF        226 DBMSrv   DBM Server client disconnect: PID 11028 on computer xxxxxx.local
2006-09-04 15:32:53 0x00001900 INF        216 DBMSrv   DBM Server client connection (established at 2006-09-04 15:32:53): PID 7308 on computer xxxxx.local
2006-09-04 15:32:53 0x00001900 INF        283 DBMSrv   command backup_start BACKt09 DATA
2006-09-04 15:33:58 0x00001900 ERR     -24920 DBMSrv   ERR_BACKUPOP: backup operation was unsuccessful
                    0x00001900 ERR     -24778 DBMSrv   The backup tool failed with 2 as sum of exit codes. The database request was canceled and ended with error -903.
2006-09-04 15:33:58 0x00001900 INF        226 DBMSrv   DBM Server client disconnect: PID 7308 on computer xxxxxx.local
Incase any logs are required then please let me know.

Storedge c2 tape autoloader not detected in veritas netbackup 6.5

Hi,
I am facing the below problem
storedge c2 tape autoloader not detected in veritas netbackup 6.5
#modinfo | grep sg
shows sgen driver
#mt -f /dev/rmt/1 status shows the HP LTO media when tape is loaded in the drive
# cfgadm -a
shows bot the drive and the changer
but when i try to detect the same in veritas netbackup, it says no removeable drives found
Any helps would be highly apreciated
thanks in advance

Hello,
Netbackup cannot see the devices if sg driver is not loaded.
Once you installed the veritas netbackup, do the following steps to properly load the sg driver.
To Load the drivers.
1. cd /usr/openv/volmgr/bin/driver //You should change to this directory before doing the following steps
2.     /usr/openv/volmgr/bin/sg.build all -mt 15 -ml 1
3.     cp pr /kernel/drv/sg.conf /kernel/drv/sg.conf.bak
4.     rm -f /kernel/drv/sg.conf
5.     /usr/openv/volmgr/bin/driver/sg.install
6.     modinfo | grep sg
7.     /usr/openv/volmgr/bin/sgscan // To verify
Thanks,
Sal.

My hard drive failed and I can't afford the professional data recovery. Do I have any other options?

I have a 13" Early 2011 Macbook Pro. Last week the hard drive failed (for the second time), and when I took it to the Apple Store I was told it would cost $177 to replace the hard drive and the cables. I intend on going through with this and getting it fixed, however the man I talked to said I should hold on to it a little longer before giving it to them for repair in case I want to recover my data. When I researched how to recover the data, my answers mostly consisted of taking it to a professional to do so. I found a local authorized service provider who said they could do so, however I was told it would cost at least $400 to recover my data (possibly more depending on how much data there is to recover).
I can afford to pay $177 to fix the hard drive. I cannot, however, afford to pay an additional $400+ to get all of my data back. Yes I realize I should have backed my files up onto an external hard drive (I'm not the most computer-savvy person), but honestly, all I really need is my photos. When my hard drive was in the process of failing, I got a flash drive and put a lot of my files onto it so I would at least have some of them when I got my Macbook fixed. At this point, I can't even turn it on, so I am not able to go into iPhoto and get my pictures.
Is there any way I can do this myself? There is no way I can get this done professionally. I would love to be able to all of my data and files back, but if I can just get some that is better than nothing. Any help would be appreciated!

obesity wrote:
It was really empty. Clicked on, nothing.
But no prohibited signs, like these:
Get info on folder, zero kb. Just one user account on each mac.
I did navigate back several days prior to my Mac dying. Still nothing.
Sounds like it may have been excluded, or perhaps damaged.
Just for future reference, it's always a good idea to keep "secondary" backups. See #27 in Time Machine - Frequently Asked Questions for some suggestions.

My 5 year old iMac's hard drive is at the point of failing. I'm trying to save al files on it, but it wont allow backup via Time Machine to an external hard drive. How can I keep from losing all the files and programs if the hard drive fails?

I'm concerned that my iMac, purchased 5 years ago, is about the crash and I can't back up my files and programs to an external hard drive via Time Machine because the computer freezes after only a couple hundred MB and gets nowhere near the required amount.
How can I save these files and programs if the computer hard drive fails or has to be replaced without the backup being able to be done? I need to do something quickly, but I'm not sure what I can or should be doing.

Get an external hard drive, some good ones that I like are the OWC, http://www.macsales.com Mercury Elite Pro drives...they are very good and reasonably priced.
Get the drive, connect via USB or FireWire, then use Disk Utility to format the new drive as Mac OS X Extended (Journaled), and then use Disk Utility to Restore the current drive to the new drive. Restore actually clones the drive so you will have a bootable backup on the external drive.
After you have formatted and named the external drive, click Restore in the main Disk Utility window, drag the name of the old hard drive to the Source box, and drag the name of the new external drive to the Destination box. DU will make an exact clone of the internal drive on the external drive.
That clone will be a complete backup of the operating system and all of your data.
You can test it by restarting the iMac and hold the Option key. That will give you a grey screen showing all bootable drives. Just click the external drive, then click the upward pointing arrow under it and the iMac will boot from that drive.

I have a 27" imac (intel) my hard drive failed and I installed a new one. This is my first time dealing with this so I am having trouble re-installing the OS. Install disc spits out and ? mark with folder appears what am I missing?

Like the title says. My 1 TB drive failed and unfortunately my computer isn't covered on the replacement program even though my hard drive is the one they are claiming to replace. So that is not a solution just to get that out of the way. So I replaced the hard drive. What I am having trouble with is re-installing my OS. I have 10.6. When I load it and restart it just spits it out and I get the question mark folder. What am I doing wrong is there a step I'm missing? I did do a back up to my Time Machine before the crash. Can I plug it into it and reload format? I'm new to this so any help would be appreciated.
Thanks!

Well, let's hope that nothing was jiggled out of place when you were installing the new hard drive. Your optical drive may just be out of alignment or the disc you have is "dirty." Get a lint-free cloth and some rubbing alcohol, then clean the disc as thoroughly as you can starting from the center and working outward in a concentric circlular motion. See if that helps. In particular clean the inner portion well.
If this does not help, then you have a problem. You have opened the computer yourself. That action voids all support from Apple. If you take the computer in for service the store may refuse to service the computer. This is not related to your now expired warranty.

Veritas netbackup sg driver failed to attach

Similar Messages

Maybe you are looking for