Buscar este blog

martes, 3 de marzo de 2015

AIX y Power 8 y Linux

Power 8 ha introducido numerosos cambios en el panorama actual... Entre ellos, incluir Linux Little Endian (el mismo que hay actualmente en Intel) en maquinas Power.

Veremos cambios significativos en el panorama a este respecto? Hacia mucho que no escribiamos ningun en el Blog ;-)

martes, 18 de junio de 2013

Renombrar Discos, Renombrar a hdisk0 rootvg

In a perfect world, 99.9% of AIX administrators would prefer their systems to look like this:

# lspv | grep rootvg
hdisk0         00c342c68dfcbdfb                    rootvg          active

However, in reality, 99.9% of AIX administrators live with systems that look something like this:

# lspv | grep rootvg
hdisk39         00c342c68dfcbdfb                    rootvg          active

And 99.9% of them don’t have time to tidy up their systems so that rootvg resides on hdisk0.

Most of them have much bigger fish to fry, such as performance, virtualisation, automation, security, project delivery, TPS reports, etc!

If they did have time, they could use the mirrorvg and rendev commands to ‘bring order to the Universe’.

1. Add a new disk to the system.
# lspv
hdisk39         00c342c68dfcbdfb                    rootvg          active
hdisk40         00c342c6161c6b47                    None

2. Rename the new disk to hdisk0.
# rendev -l hdisk40 -n hdisk0
hdisk0

3. Add the disk to rootvg and mirror to it.
# extendvg rootvg hdisk0
# mirrorvg rootvg hdisk0
0516-1804 chvg: The quorum change takes effect immediately.
0516-1126 mirrorvg: rootvg successfully mirrored, user should perform
        bosboot of system to initialize boot records.  Then, user must modify
        bootlist to include:  hdisk0 hdisk39.

4. Create a boot image on hdisk0.
# bosboot –a –d /dev/hdisk0
bosboot: Boot image is 49180 512 byte blocks.
# ipl_varyon -i
[S 3670248 14942228 07/24/12-10:18:11:104 ipl_varyon.c 1270] ipl_varyon -i
PVNAME          BOOT DEVICE     PVID                    VOLUME GROUP ID
hdisk39         YES             00c342c68dfcbdfb0000000000000000        00c342c600004c00
hdisk0          YES             00c342c6161c6b470000000000000000        00c342c600004c00
[E 3670248 0:087 ipl_varyon.c 1410] ipl_varyon: exited with rc=0

5. Check and update the bootlist. Ensure hdisk0 is in the bootlist.
# bootlist –m normal hdisk0
# bootlist –m normal –o
hdisk0 blv=hd5 pathid=0
hdisk0 blv=hd5 pathid=1

6. Unmirror rootvg on hdisk39. Remove hdisk39 from rootvg.
# unmirrorvg rootvg hdisk39
0516-1246 rmlvcopy: If hd5 is the boot logical volume, please run 'chpv -c <diskname>'
        as root user to clear the boot record and avoid a potential boot
        off an old boot image that may reside on the disk from which this
        logical volume is moved/removed.
0516-1804 chvg: The quorum change takes effect immediately.
0516-1144 unmirrorvg: rootvg successfully unmirrored, user should perform
        bosboot of system to reinitialize boot records.  Then, user must modify
        bootlist to just include:  hdisk0.
# lspv –l hdisk39
#
# reducevg rootvg hdisk39
#

7. Order in the Universe has been restored.
# lspv | grep rootvg
hdisk0          00c342c6161c6b47                    rootvg          active

Of course, all of this assumes that the name, hdisk0, is not already in use by a hdisk in another volume group on the system.
Again, if you have the time, then you could perform the following to rectify the situation.

1. datavg resides on hdisk0.
# lspv | grep datavg
hdisk0          00c342c6161c6b47                    datavg          active

2. Unmount all data file systems in this volume group. Varyoff the VG.

# unmount /datafs
# varyoffvg datavg


3. Rename hdisk0 to hdisk99 (or something other than hdisk0).
# rendev -l hdisk0 -n hdisk99
hdisk99

4. Varyon the volume group, datavg. Mount the data file systems in datavg.
# varyonvg datavg
# mount /datafs

5. datavg now resides on (the newly renamed) hdisk99.
# lspv | grep datavg

hdisk99         00c342c6161c6b47                    datavg

martes, 5 de marzo de 2013

Si tenemos un error al clonar discos

                                                    COMMAND STATUS
Command: failed        stdout: yes           stderr: no
Before command completion, additional instructions may appear below.
Calling mkszfile to create new /image.data file.
0301-168 bosboot: The current boot logical volume, /dev/hd5,
        does not exist on /dev/hdisk11.


0301-168 bosboot: The current boot logical volume, /dev/hd5,
        does not exist on /dev/hdisk11.
0505-121 alt_disk_install: Error.
Cleaning up.


F1=Help                      F2=Refresh                    F3=Cancel                     F6=Command
F8=Image                     F9=Shell                      F10=Exit                      /=Find
n=Find Next

# lsvg -p rootvg ==> hdisk0
# lslv -m hd5 ==> hdisk0 on PV1 1st partition
# savebase -v ==>  successful
will remove/recreate hd5
# rmlv hd5
# mklv -y hd5 -t boot -a e rootvg 1 hdisk0
# cd /dev
# rm ipldevice
# ln /dev/rhdisk0 /dev/ipldevice
# bosboot -ad /dev/ipldevice ==>  same error
# bootinfo -B hdisk0 ==>  You get 1
# ln /dev/rhd5 /dev/ipl_blv
# cd /dev - same maj/min numbers
# bosboot -ad /dev/ipldevice ==>  works!
# bosboot -ad hdisk0
alt_disk_install was able to proceed now

miércoles, 12 de diciembre de 2012

Scripting en VIO Server

En este post explico como ejecutar comandos propios de la CLI del VIO desde la shell de AIX (del VIO Server).

Esto permite definir scripts y ejecutarlos para realizar tareas que de otra forma se tendrían que hacer a mano.

Es un tema que yo ya conocía, y es probable que vosotros también, pero como me ha sido bastante útil hace poco lo pongo en el blog por si a alguno le viene bien.

viernes, 10 de agosto de 2012

Comprobar IPde power system de IBM desde frontal


HOW TO FIND IP ADDRESS ON HMC PORTS

You must first put the physical control panel in manual operating mode before you can select or activate certain functions.
To put the physical control panel in manual operating mode, do the following:
1. Use the Increment button to scroll to function 02.

2. 0 2 _ _ _ _ _ _ _ _ _ _ _ _ _ _
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

3. Press Enter to start function 02.

4. Press Enter again to move to the second character on the function 02 menu. The current system operating mode is displayed with a pointer, as shown in the following example:
5. 0 2 _ _ B _ _ N < _ _ _ _ _ _ _
_ _ _ _ _ _ _ _ _ _ _ _ P _ _ _

6. Use the Increment button to scroll through the system operating modes, and select M for manual, as shown in the following example:

7. 0 2 _ _ B _ _ M < _ _ _ _ _ _ _
_ _ _ _ _ _ _ _ _ _ _ _ P _ _ _
8. Press Enter to select the system operating mode.

9. Press Enter again to exit function 02.
The control panel is in manual operating mode.



Function 30: Service processor IP address and port location

This function displays the service processor IP address and port location. This function is available only in the manual operating mode and from power on standby.
Note: If IPv6 is displayed on the control panel, then the service processor's network ports are configured with IPv6 IP addresses. There are not enough characters on the control panel to display the entire address.

So to get to this function , you have to use increment Key to reach 30 and then press enter key to enter the function 30 and reach 00 to get IP address on HMC1 port and 01 to get IP address on HMC2 port.

martes, 29 de mayo de 2012

NIMADM 5.3 a 6.1

Dar de alta un nuevo cliente en NIM:
===========================
Editar /etc/hosts :

En el Nim Master
# vi /etc/hosts
[insertar la linea correspondiente al cliente]

En el clinete
# vi /etc/hosts
[insertar la linea correspondiente al master]

smity nim > Perform NIM Administration Tasks > Manage Machines > Define a Machine >
[Cliente]

Solo Camboiar:

Cable Type        [tp]

Recrear el fichero /etc/niminfo:


# rm /etc/niminfo

# niminit -a name=robin -a master=superman -a master_port=1058


Tareas de la migración por NIM: NIMADM:
================================

Prerrequisitos:
-----------------
Nim Master al máximo nivel de parcheado.
Damos por hecho que en el NIM existe un lppsource y un spot de la version a migrar, generado con smitty bffcreate, desde los dvd originales y posteriormente parcheado.
Comprobar paquetes necesarios en le master
# lslpp -l bos.alt_disk_install.rte   
# nim -o showres 'spotaix61031' | grep bos.alt_disk_install.rte

Procedimiento previo:
----------------------------
Romper mirror en Cliente:
# unmirrorvg roovg.
# reducevg rootvg hdisk1 (migratepv pc si fuera necesario).

Confirmar que todos los lvs de rootvg están open/syncd
#lsvg -l rootvg

Generar un vg en Master, sobre un disco, que sirva de cache. En el servidor NIM. (Tamaño igual que hdisk1 de clinete)
El VG se llamará: nimadmvg, esta creado, pero es necesario que lo elimines y lo vuelvas a crear, para que este vacío:

 # lsvg -l nimadmvg
       nimadmvg:
       LV NAME  TYPE       LPs     PPs     PVs  LV STATE      MOUNT POINT

En el cliente certificar que tenemos el disco libre, para clonar sobre el:

# lspv        
hdisk0 0000273ac30fdcfc rootvg          active        
hdisk1 000273ac30fdd6e  None

Procedimiento de migración:
-----------------------------------
# nohup ./tarea.sh &

# nimadm -j nimadmvg -c aix1 -s spotaix61031 -l lppsourceaix61031 -d "hdisk1" –Y
  
Donde :
nimadmvg – vg en servidor nim vacio a usar como cache.
aix1 – nombre del objeto cliente nim
spotaix61031 – spot a usar
lppsourceaix61031 – lppsource a usar
hdisk1 – Disco alternativo

EL commando viene de:

smitty nimadm
* Target NIM Client [select your client]
* NIM LPP_SOURCE resource [el lppsource es el de aix61tl6 que ha sido actualizado]
* NIM SPOT resource [select the spot]
* Target Disk(s) to install [select the disk(s) you want to migrate to]
DISK CACHE volume group name []

NIM IMAGE_DATA resource []
NIM BOSINST_DATA resource []
NIM EXCLUDE_FILES resource []
NIM INSTALLP_BUNDLE resource []
NIM PRE-MIGRATION SCRIPT resource []

lunes, 28 de mayo de 2012

Error al crear bosboot Migrando AIX 5.3 TL12 a 6.1 TL6

Tras realizar la migración, por el procedimiento de disco alternativo:
#alt_disk_copy -d hdisk1 (-B si no queremos que el ordende arranque quede modificado).

Se puede producir un error, por el cual , migración queda colgada, y resuelve como error al el sector de arranque, detallando que no encuentra el archivo:

/usr/lib/drivers/pci/sissas64_dd

Solución:

#cp /usr/lib/driver/pci/sissas_dd /usr/lib/driver/pci/sissas64_dd

A continuación reanudar la migración, ya en su punto final.

En mi caso para terminar la actualización aplique los parches de TL7 SP2 y esos dos ficheros que deben existir, son reconfigurados, y desaparecen los problemas.

Importante: Este error, era el causante de que tras la migración por NIMAD, nim con disco alternativo, la maquina fuera incapaz de arrancar desde el disco recientemente migrado. Sin reportar ningún error, solo se quedaba colgada la particion en "Starting Kernel". La causa es corupcion de hd5, por eso no hasta no reconstruir:

# ls -la /usr/lib/driver/pci/sissas64_dd

 no es posible realizar un:

#bosboot -ad hdisk1 correctamente.

Espero sirva de ayuda a quienes tengais problemas parecidos.

Saludos.