Re: mksysb failed

[KOG]

Hi Folks,

Hi Folks

I am trying to create mksysb onto a new tape on the new pseries server. For some reason it failed and I am not clear with what the error messages is telling me. I hope you can shed some light?

Error Msgs

Creating tape boot image
0301-161 bosboot : dd failed to copy /dev/rmt0.1

0301-165 bosboot : WARNING bosboot failed - do not attempt to boot device.

0512-016 mksysb : Attempt to create a bootable tape failed : bosboot -d /dev/rmt0.1 -a failed with return code 48.

I have changed the block size and the command ran ok.

chdev -l rmt0 -a block_size=512
done without any probs and I could rewind it back too.

What should I do?


Thanking you all in advance,

Regards

K

 

[mjlrock]

Did you use mksysb from the command line or through smitty? I know when I tried from the command line it failed but I don't recall the error message.

 

[KOG]

Hello mjlrock

I ran the commands through smitty.

I ran lslpp commands to check if I have the filesets for tape.rte, this is what I got.

# lslpp -l devices.scsi.tape.rte
Fileset Level State Description
----------------------------------------------------------------------------
Path: /usr/lib/objrepos
devices.scsi.tape.rte 4.3.3.10 COMMITTED SCSI Tape Device Software

so what could be the problem?

Regards

K

 

[mjlrock]

Sorry KOG I don't know what the problem could be when run from within smitty. Maybe a possibility is are the tape heads clean? have you looked at the type of data that did get from the failed part and compare it with the data from the one when you change the parameters. Otherwise right now I can't think of anything else.

 

[aixosok]

You might want to check out the free space in /tmp file system. mksysb uses workspace on /tmp to create the tape boot image before dumping it to tape. Use the command

bosboot -q -a -d /dev/ipldevice
to determint the file space required in /tmp

You may want to refresh the boot image in case it has been corrpted somehow. use the command
bosboot -a -d /dev/ipldevice
to rebuild the boot image files then retry the mksysb

I also suggest getting Sysback from IBM. It is a very usefull tool for backing up and moving just about anything from one system or volume group to another. It also provides more information in any error messages. It is well worth the cost for the ability it provices.

 

[KOG]

Hi aixosok

Many thanks for your response, I run the commands when I return to office on Monday morning.

Regards

K

 

[zyrag]

try to run the command
#tctl -f /dev/rmtx status
then, post the result here.

 

[victorv]

hi,
usually to make a system backup, the best is to leave
the block size at default (1024) : to be sure, remove
tape (rmdev or smit) and cfgmgr (or smit device configure
devices added after ipl).

The mksysb command uses current block size to save
all files but before, to write boot image, the preocedure
changes it the block size setting it to a fixed value
and then resets it at current value. To save all files
on tape, is better if you have 1024 value, becouse
this has the best performances (time to wait).

( give the command "which mksysb" and have a look inside
procedure)

Other values set in properties (as 0 or 512) must be used
for compatibility with other systems (sun,hp,ecc) using
tar to export files, but after have made these operations,
is better if you re-set block size to default value.

bye

 

[KOG]

Hello Victorv

Here's the output of tctl -f /dev/rmt0

# tctl -f /dev/rmt0 status
rmt0 Available 10-60-00-5,0 SCSI 8mm Tape Drive
attribute value description user_settable

mode yes Use DEVICE BUFFERS during writes True
block_size 512 BLOCK size (0=variable length) True
extfm no Use EXTENDED file marks True
ret no RETENSION on tape change or reset True
density_set_1 39 DENSITY setting #1 True
density_set_2 39 DENSITY setting #2 True
compress yes Use data COMPRESSION True
size_in_mb 20000 Size in Megabytes False
ret_error no RETURN error on tape change or reset True

Is this ok?

Regards

K

 

[KOG]

Hi

I ran two commands as suggested by aixosok

bosboot -q -a -d /dev/ipldevice and
bosboot - a -d /dev/ipldevice, both ran ok.

Filesystem KB required
/tmp 9144

which I do have enough space on /tmp.

# df -k
Filesystem 1024-blocks Free %Used Iused %Iused Mounted on
/dev/hd4 65536 58756 11% 927 3% /
/dev/hd2 393216 72544 82% 13692 14% /usr
/dev/hd9var 65536 62124 6% 148 1% /var
/dev/hd3 131072 118072 10% 161 1% /tmp

Then ran smitty mksysb and it failed again !

Command: failed stdout: yes stderr: no

Before command completion, additional instructions may appear below.


Creating information file (/image.data) for rootvg.

Creating tape boot image...
0301-161 bosboot: dd failed to copy /dev/rmt0.1.

0301-165 bosboot: WARNING! bosboot failed - do not attempt to boot device.

0512-016 mksysb: Attempt to create a bootable tape failed:
bosboot -d /dev/rmt0.1 -a failed with return code 48.
Any idea why?

Regards

Katherine

 

[victorv]

hi,

the problem seems to be when writing well created
boot-image on tape; are you able to write on tape ?

try .

- put a tape on 8mm device ( the rule to make tape writeble
are opposit of 4mm or music-cartidge: you can write when
windows is open )

- enter command
find /tmp | backup -ivf /dev/rmt0

- verify using
restore -Tvf /dev/rmt0

- try also
dd if=/tmp/an_existent.file of=/dev/rmt0

see

http://publib.boulder.ibm.com/doc_link/en_US/a_doc_lib/cmds/aixcmds1/bosboot.htm

and try:

1)To create a boot image on the default boot logical volume on the fixed disk from which the system is booted, enter:
bosboot -a

2) To create a bootable image called /tmp/tape.bootimage for a tape device, enter:
bosboot -ad /dev/rmt0 -b /tmp/tape.bootimage

3)To copy a given tape boot image to a tape device, enter:
bosboot -w /tmp/tape.bootimage -d rmt0


bye

 

[KOG]

HI

It looks like the tape driver that is the source of problem, for some reason it is failing to write boot image file to it. It is a new pseries 6E1 server which was only purchased few weeks ago ! The tape driver can take 20/40gb tape and the one I inserted is 5gb so I wonder if that is the problem?

# bosboot -a

bosboot: Boot image is 7269 512 byte blocks.
# bosboot -ad /dev/rmt0 -b /tmp/tape.bootimage

bosboot: Boot image is 8111 512 byte blocks.
# bosboot -w /tmp/tape.bootimage -d rmt0

bosboot: Boot image is 8110 512 byte blocks.

0301-161 bosboot: dd failed to copy /dev/rmt0.
#

thanking you all in advance.

Regards

K

 

[victorv]

hi,

sorry for my insistance: are you able,
indipententely from boot-image, to write to tape ?

- enter command
find /tmp | backup -ivf /dev/rmt0

- verify using
restore -Tvf /dev/rmt0

- try also
dd if=/tmp/an_existent.file of=/dev/rmt0

bye

 

[KOG]

Hi victorv

Pls excuse my aix knowledge it is quiet limit, I have changed the tapes twice and received the same error message. One of the tape is brand new.

# find /tmp | backup -ivf /dev/rmt0
Mount volume 1 on /dev/rmt0.
Press Enter to continue.
Backing up to /dev/rmt0.
Cluster 51200 bytes (100 blocks).
Volume 1 on /dev/rmt0
a 0 /tmp
a 0 /tmp/lost+found
a 0 /tmp/.strload.mutex
backup: /tmp/.PMDV1 socket is not backed up.
backup: 0511-432 A write error occurred: A system call received a parameter that
is not valid.
backup: 0511-079 The backup media may be damaged.
Use a different disk to run the backup.

How do I rule out hardware problems? The server is brand new so I am assuming all should be ok. If I have tried two different tapes and received the same error messages, then does it mean the tape driver is faulty?

Thanking you all in advance.

Katherine

 

[victorv]

hi,

please submit us the ouput of commands:

lsdev -C -c tape
df
oslevel -r
instfix -i | grep "AIX"

bye

 

[KOG]

Hi victorv

Here's the outputs

# lsdev -C -c tape
rmt0 Available 10-60-00-5,0 SCSI 8mm Tape Drive
# df
Filesystem 512-blocks Free %Used Iused %Iused Mounted on
/dev/hd4 131072 117504 11% 927 3% /
/dev/hd2 786432 145088 82% 13692 14% /usr
/dev/hd9var 131072 124248 6% 148 1% /var
/dev/hd3 262144 228024 14% 162 1% /tmp
/dev/hd1 131072 126760 4% 24 1% /home
/dev/lv00 32768000 19545672 41% 28846 1% /opt
/dev/lv02 13107200 9276432 30% 26 1% /oracledata1
/dev/lv03 13107200 10233616 22% 26 1% /oracledata2
/dev/lv04 13107200 11684400 11% 21 1% /oracledata3
/dev/lv05 13107200 11646056 12% 20 1% /oracledata4
/dev/lv10 13107200 9570376 27% 26 1% /oracledev1
/dev/lv11 13107200 9481576 28% 25 1% /oracledev2
/dev/lv12 13107200 8739448 34% 28 1% /oracledev3
/dev/lv15 32768000 23252664 30% 40 1% /backup

# oslevel -r
4320-02
# instfix -i|grep "AIX"
All filesets for 4.3.0.0_AIX_ML were found.
All filesets for 4.3.1.0_AIX_ML were found.
All filesets for 4.3.2.0_AIX_ML were found.
All filesets for 4320-02_AIX_ML were found.
All filesets for 4.3.3.0_AIX_ML were found.
Not all filesets for 4330-01_AIX_ML were found.
Not all filesets for 4330-02_AIX_ML were found.
Not all filesets for 4330-03_AIX_ML were found.
Not all filesets for 4330-04_AIX_ML were found.
Not all filesets for 4330-05_AIX_ML were found.
Not all filesets for 4330-06_AIX_ML were found.
Not all filesets for 4330-07_AIX_ML were found.
Not all filesets for 4330-08_AIX_ML were found.
Not all filesets for 4330-09_AIX_ML were found.

I did complete installation of AIX 4.3.3 so how come there are some filesets missing?

Regards

Katherine

 

[victorv]

hi,

A) could you also give output of

ls -l /dev/rmt*

B) Maintenance level

Whithout inspect which filesets are missing for a
specific maintenance level

instfix -i -k 4330-09_AIX_ML -c | grep ":-:"

insert last Maintenance level you get ( 09 or 11 if you can)

restart the system .


The message

"The backup media may be damaged"

is normal when a backup fails, for any reason.


There is some of strange:
the error code 48 (for errno variable) refers to IP address error, and your backup stops with a

/tmp/.PMDV1 socket is not backed up.

After you ML is ok, do a

rmdev -l rmt0 -d
cfgmgr
find /home | backup -ivf /dev/rmt0

bye

 

[victorv]

hi,
always me.

The 8mm tape as 1/4" cartridge, are almost strange.

see

http://publib.boulder.ibm.com/doc_link/en_US/a_doc_lib/files/aixfiles/rmt.htm

The point 3 says:

... The 8mm tape drive uses a minimum block size of 1024 bytes. ...

I don't know if you know the maning of the strange table
in this article:

In your device configuration put your attention at density
setting #1 : change only this value, trying between possibilities and use as special file

/dev/rmt0

So you will use

Rewind on Close = Yes
Retension on Open = No
Density Setting #1

If you use /dev/rmt0.4 you will use

Rewind on Close = Yes
Retension on Open = No
Density Setting #2


(Seeing the table, you have 8 possibility done from
the combination of 3 binary values:

Rewind on Close : Yes or No
Retension on Open : Yes or No
Density Setting : #1 or #2

Probably, indipendentely from sw pacthes, is this your problem.

bye

 

[KOG]

Many thanks for all your input and help, I will do it later this evening or tomw evening as I have some developers using the server so it will be awkward interrupting their work.

Cheers

Katherine

 

[kclute]

Hi,

" The tape driver can take 20/40gb tape and the one I inserted is 5gb so I wonder if that is the problem? "

I am assuming that the tape drive is the 20/40 drive. If it is and you are using a 5 gig tape your backups will not work correctly. The 5gb tapes and the 20/40 gig tapes are not compatible with each other. Try getting one of the 20/40 gig tapes and trying it again with that. It will say "Advanced Metal Evaportated" on the tape with a reorder number of 59H2678. You may have one that came with the new system and drive. Try that and see if you get the same error.

-kirk