David Moffatt | 1 Jun 18:44 2009

Re: [uml-user] get process CPU kernel time / user time

Massimiliano,

I am running 2.6.24 with a fairly generic kernel config and it works
fine for me.  However, I am not using the SKAS patch on the host.

~ # time grep -r fred /var > /dev/null     
Command terminated by signal 2
real    1m 6.91s
user    0m 0.17s
sys     0m 0.01s

--David.

-----Original Message-----
From: maxd <at> inwind.it [mailto:maxd <at> inwind.it] 
Sent: Sunday, May 31, 2009 3:11 AM
To: user-mode-linux-user
Subject: [uml-user] get process CPU kernel time / user time

Hi guys,
is there a way to get kernel time and user time of a process running
inside user mode linux? I have noticed with the "time" command that the
kernel time is always 0. Moreover, with pidstat I get wierd values for
CPU % utilization (values larger than 100 %). 
I am running a kernel 2.6.27-7, in skas mode.
Thanks a lot for any help you will be able to provide.
Cheers,
Massimiliano

------------------------------------------------------------------------
(Continue reading)

maxd@inwind.it | 2 Jun 12:40 2009
Picon

Re: [uml-user] get process CPU kernel time / user time

Hi David,
thanks for the reply. I played a little bit more with user mode linux and I have got non zero sys times.
However, I still get very wierd values with pidstat: the CPU usage can reach values even larger than 1000,
while with "top" everything looks fine. I googled a little bit, but no one seems to have this problem. 
Does anyone have any idea about the possible cause? 
Cheers,
Massimiliano

---------- Initial Header -----------

>From      : "David Moffatt" David.Moffatt <at> access-company.com
To          : maxd <at> inwind.it,"user-mode-linux-user" user-mode-linux-user <at> lists.sourceforge.net
Cc          : 
Date      : Mon, 1 Jun 2009 09:44:50 -0700
Subject : RE: [uml-user] get process CPU kernel time / user time

> Massimiliano,
> 
> I am running 2.6.24 with a fairly generic kernel config and it works
> fine for me.  However, I am not using the SKAS patch on the host.
> 
> ~ # time grep -r fred /var > /dev/null     
> Command terminated by signal 2
> real    1m 6.91s
> user    0m 0.17s
> sys     0m 0.01s
> 
> --David.
> 
> 
(Continue reading)

Mark Lidd | 8 Jun 02:33 2009
Picon

[uml-user] uml jail does not run when "auto-started"

I have been using UML for while with good results, but this lastest 
issue has got me stumped.  I put together a UML jail with a perl 
wrapper.  It starts running as root and calls compartment to do the 
chroot, change user, and run the UML.  It runs just fine from the 
command line, but it does not run from a crontab or if I connect it to a 
"button" or rc.local.  I checked the environment but have been careful 
not to use anything from the enviroment.  The UML is basically fedore 9 
(I think).  the UML kernel is 2.6.26, gcc 4.3.0 20080428 redhat 4.3.0-8. 
running in SKAS0.  The host is 2.6.27.19-170.2.35.fc8  I fc10 also 
exhibits the same startup issues.

The UML bootup messages stops at VFS: mounted root (ext2 filesystem) 
readonly.  The next message "Welcome to Fedora" never appears.

The UML process seems to hang using 98% cpu time (from top).

Some memory parameters are: virt 256, res 9532, SHR 9332, mem .3%.

A normal run (from the command line) is virt 256M, resm 43, shr 43, mem 
1.2%.  The script is suid and I do change to the root user (real & eff) 
inside of the perl script.

I also tried running it from  rc.local, but no luck.

Any ideas?

thanks,

Mark

(Continue reading)

shah khadafi | 8 Jun 04:37 2009
Picon

[uml-user] failed whwn online ocfs2 in UML

i try to build OCFS2 cluster in UML.

i use ocfs2-tools-1.2.2 package to this project. i was setting like following in chapter 12 USER MODE LINUX book.

i use kernel 2.6.24 and 2.6.25 as sources kernel UML

1. start build OCFS2 as modules kernel,

    linux-2.6.24# make modules_install INSTALL_MOD_PATH=../rootfs

INSTALL crypto/crypto_hash.ko

INSTALL crypto/hmac.ko

INSTALL crypto/sha1_generic.ko

INSTALL drivers/block/nbd.ko

INSTALL drivers/net/dummy.ko

INSTALL drivers/net/ppp_generic.ko

INSTALL drivers/net/slhc.ko

INSTALL drivers/net/slip.ko

INSTALL fs/autofs/autofs.ko

INSTALL fs/autofs4/autofs4.ko

INSTALL fs/binfmt_misc.ko

INSTALL fs/configfs/configfs.ko

INSTALL fs/dlm/dlm.ko

INSTALL fs/gfs2/gfs2.ko

INSTALL fs/gfs2/locking/dlm/lock_dlm.ko

INSTALL fs/gfs2/locking/nolock/lock_nolock.ko

INSTALL fs/ocfs2/cluster/ocfs2_nodemanager.ko

INSTALL fs/ocfs2/dlm/ocfs2_dlm.ko

INSTALL fs/ocfs2/dlm/ocfs2_dlmfs.ko

INSTALL fs/ocfs2/ocfs2.ko

INSTALL net/sctp/sctp.ko

DEPMOD 2.6.24


    same process i build OCFS2 as modules kernel to source kernel 2.6.25.
    i run the 2 kernel UML like this
  
    ./linux ubd0=FestyFawn-i386-rootfs root=/dev/ubda eth0=tuntap,uml-conn0 mem=128M

    ./linux ubd0=../FestyFawn-i386-rootfs root=/dev/ubda eth0=tuntap,uml-conn1 mem=128M
   
2. configure ocfs2-tools-1.2.2 package in UML
    before i run this step, i was upgrade my root filesystem to Ubuntu 8.04 HardyHeron and then i configure ocfs2-tools-1.2.2 package
   
    root <at> none:/usr/local/ocfs2-tools-1.2.2# ./configure
   
3. i start ocfs2-tools-1.2.2 package with load necessary modules. 
    root <at> none:/usr/local/ocfs2-tools-1.2.2#./vendor/common/o2cb.init load

  Loading module “configfs”: OK
  Mounting configfs filesystem at /sys/kernel/config: OK
  Loading module “ocfs2_nodemanager”: OK
  Loading module “ocfs2_dlm”: OK
  Loading module “ocfs2_dlmfs”: OK

  Mounting ocfs2_dlmfs filesystem at /dlm: OK

4. i copy file cluster.conf to /etc/ocfs2/.
    root <at> none:/usr/local/ocfs2-tools-1.2.2#cat /etc/ocfs/cluster.conf
    node:
            ip_port = 7777
            ip_address = 192.168.0.112
            number = 0
            name = node0
            cluster = ocfs2

    node:
            ip_port = 7777
            ip_address = 192.168.0.116
            number = 0
            name = node1
            cluster = ocfs2    

    cluster:
           node_count = 0
           name = ocfs2

5. i setting hostname and ip address to
    UML0# hostname node0
    UML1# hostname node1
    then i restart both UMLs
 
6. then i run command on host system to setup filesystem to shared device.
   
    host_system$ dd if=/dev/zero of=ocfs seek=$[ 100 * 1024 ] bs=1K count=1
    host_system# uml_mconsole dxhcF8 config ubdbc=ocfs 
   
    where dxhcF8 is umid
    in UML file /dev seen like this
          root <at> node0:# ls -l /dev/ubd*
           /ubda /ubdb /urandom

7. after all process above ,  i have  problem when i try to online OCFS2
    root <at> node0:/usr/local/ocfs2-tools-1.2.2#./vendor/common/o2cb.init online ocfs2
    Starting O2CB cluster ocfs2: Failed
    ./vendor/common/o2cb.init: line 629: o2cb_ctl: command not found
    Stopping O2CB cluster ocfs2: Failed
    ./vendor/common/o2cb.init: line 652: o2cb_ctl: command not found

i was check the file o2cb.init

what happen about that failed ?? maybe i was missing one step to finish setup OCFS2 in UML.

any one can give me advice and explain about failed to online OCFS2 cluster.

thanks.



       





------------------------------------------------------------------------------
OpenSolaris 2009.06 is a cutting edge operating system for enterprises 
looking to deploy the next generation of Solaris that includes the latest 
innovations from Sun and the OpenSource community. Download a copy and 
enjoy capabilities such as Networking, Storage and Virtualization. 
Go to: http://p.sf.net/sfu/opensolaris-get
_______________________________________________
User-mode-linux-user mailing list
User-mode-linux-user <at> lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-user
Jeff Dike | 8 Jun 05:41 2009

Re: [uml-user] uml jail does not run when "auto-started"

On Sun, Jun 07, 2009 at 08:33:53PM -0400, Mark Lidd wrote:
> I have been using UML for while with good results, but this lastest 
> issue has got me stumped.  I put together a UML jail with a perl 
> wrapper.  It starts running as root and calls compartment to do the 
> chroot, change user, and run the UML.  It runs just fine from the 
> command line, but it does not run from a crontab or if I connect it to a 
> "button" or rc.local.

I bet it's confused about its stdin/stdout.  Normal practice is to run it
inside a screen session.

				Jeff

------------------------------------------------------------------------------
OpenSolaris 2009.06 is a cutting edge operating system for enterprises 
looking to deploy the next generation of Solaris that includes the latest 
innovations from Sun and the OpenSource community. Download a copy and 
enjoy capabilities such as Networking, Storage and Virtualization. 
Go to: http://p.sf.net/sfu/opensolaris-get
shah khadafi | 11 Jun 17:54 2009
Picon

[uml-user] how to on line OCFS2 tools on UML

i try to build OCFS2 cluster in UML.

i use ocfs2-tools-1.2.2 package to this project. i was setting like following in chapter 12 USER MODE LINUX book.

i use kernel 2.6.24 and 2.6.25 as sources kernel UML

1. start build OCFS2 as modules kernel,

    linux-2.6.24# make modules_install INSTALL_MOD_PATH=../rootfs

INSTALL crypto/crypto_hash.ko

INSTALL crypto/hmac.ko

INSTALL crypto/sha1_generic.ko

INSTALL drivers/block/nbd.ko

INSTALL drivers/net/dummy.ko

INSTALL drivers/net/ppp_generic.ko

INSTALL drivers/net/slhc.ko

INSTALL drivers/net/slip.ko

INSTALL fs/autofs/autofs.ko

INSTALL fs/autofs4/autofs4.ko

INSTALL fs/binfmt_misc.ko

INSTALL fs/configfs/configfs.ko

INSTALL fs/dlm/dlm.ko

INSTALL fs/gfs2/gfs2.ko

INSTALL fs/gfs2/locking/dlm/lock_dlm.ko

INSTALL fs/gfs2/locking/nolock/lock_nolock.ko

INSTALL fs/ocfs2/cluster/ocfs2_nodemanager.ko

INSTALL fs/ocfs2/dlm/ocfs2_dlm.ko

INSTALL fs/ocfs2/dlm/ocfs2_dlmfs.ko

INSTALL fs/ocfs2/ocfs2.ko

INSTALL net/sctp/sctp.ko

DEPMOD 2.6.24


    same process i build OCFS2 as modules kernel to source kernel 2.6.25.
    i run the 2 kernel UML like this
  
    ./linux ubd0=FestyFawn-i386-rootfs root=/dev/ubda eth0=tuntap,uml-conn0 mem=128M

    ./linux ubd0=../FestyFawn-i386-rootfs root=/dev/ubda eth0=tuntap,uml-conn1 mem=128M
   
2. configure ocfs2-tools-1.2.2 package in UML
    before i run this step, i was upgrade my root filesystem to Ubuntu 8.04 HardyHeron and then i configure ocfs2-tools-1.2.2 package
   
    root <at> none:/usr/local/ocfs2-
tools-1.2.2# ./configure
   
3. i start ocfs2-tools-1.2.2 package with load necessary modules. 
    root <at> none:/usr/local/ocfs2-tools-1.2.2#./vendor/common/o2cb.init load

  Loading module “configfs”: OK
  Mounting configfs filesystem at /sys/kernel/config: OK
  Loading module “ocfs2_nodemanager”: OK
  Loading module “ocfs2_dlm”: OK
  Loading module “ocfs2_dlmfs”: OK

  Mounting ocfs2_dlmfs filesystem at /dlm: OK

4. i copy file cluster.conf to /etc/ocfs2/.
    root <at> none:/usr/local/ocfs2-tools-1.2.2#cat /etc/ocfs/cluster.conf
    node:
            ip_port = 7777
            ip_address = 192.168.0.112
            number = 0
            name = node0
            cluster = ocfs2

    node:
            ip_port = 7777
            ip_address = 192.168.0.116
            number = 0
            name = node1
            cluster = ocfs2    

    cluster:
           node_count = 0
           name = ocfs2

5. i setting hostname and ip address to
    UML0# hostname node0
    UML1# hostname node1
    then i restart both UMLs
 
6. then i run command on host system to setup filesystem to shared device.
   
    host_system$ dd if=/dev/zero of=ocfs seek=$[ 100 * 1024 ] bs=1K count=1
    host_system# uml_mconsole dxhcF8 config ubdbc=ocfs 
   
    where dxhcF8 is umid
    in UML file /dev seen like this
          root <at> node0:# ls -l /dev/ubd*
           /ubda /ubdb /urandom

7. after all process above ,  i have  problem when i try to online OCFS2
    root <at> node0:/usr/local/ocfs2-tools-1.2.2#./vendor/common/o2cb.init online ocfs2
    Starting O2CB cluster ocfs2: Failed
    ./vendor/common/o2cb.init: line 629: o2cb_ctl: command not found
    Stopping O2CB cluster ocfs2: Failed
    ./vendor/common/o2cb.init: line 652: o2cb_ctl: command not found

i was check the file o2cb.init

what happen about that failed ?? maybe i was missing one step to finish setup OCFS2 in UML.

any one can give me advice and explain about failed to online OCFS2 cluster.

thanks.
------------------------------------------------------------------------------
Crystal Reports - New Free Runtime and 30 Day Trial
Check out the new simplified licensing option that enables unlimited
royalty-free distribution of the report engine for externally facing 
server and web deployment.
http://p.sf.net/sfu/businessobjects
_______________________________________________
User-mode-linux-user mailing list
User-mode-linux-user <at> lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-user
shah khadafi | 11 Jun 17:57 2009
Picon

[uml-user] failed when try to online ocfs2 on UML

i try to build OCFS2 cluster in UML.

i use ocfs2-tools-1.2.2 package to this project. i was setting like following in chapter 12 USER MODE LINUX book.

i use kernel 2.6.24 and 2.6.25 as sources kernel UML

1. start build OCFS2 as modules kernel,

    linux-2.6.24# make modules_install INSTALL_MOD_PATH=../rootfs

INSTALL crypto/crypto_hash.ko

INSTALL crypto/hmac.ko

INSTALL crypto/sha1_generic.ko

INSTALL drivers/block/nbd.ko

INSTALL drivers/net/dummy.ko

INSTALL drivers/net/ppp_generic.ko

INSTALL drivers/net/slhc.ko

INSTALL drivers/net/slip.ko

INSTALL fs/autofs/autofs.ko

INSTALL fs/autofs4/autofs4.ko

INSTALL fs/binfmt_misc.ko

INSTALL fs/configfs/configfs.ko

INSTALL fs/dlm/dlm.ko

INSTALL fs/gfs2/gfs2.ko

INSTALL fs/gfs2/locking/dlm/lock_dlm.ko

INSTALL fs/gfs2/locking/nolock/lock_nolock.ko

INSTALL fs/ocfs2/cluster/ocfs2_nodemanager.ko

INSTALL fs/ocfs2/dlm/ocfs2_dlm.ko

INSTALL fs/ocfs2/dlm/ocfs2_dlmfs.ko

INSTALL fs/ocfs2/ocfs2.ko

INSTALL net/sctp/sctp.ko

DEPMOD 2.6.24


    same process i build OCFS2 as modules kernel to source kernel 2.6.25.
    i run the 2 kernel UML like this
  
    ./linux ubd0=FestyFawn-i386-rootfs root=/dev/ubda eth0=tuntap,uml-conn0 mem=128M

    ./linux ubd0=../FestyFawn-i386-rootfs root=/dev/ubda eth0=tuntap,uml-conn1 mem=128M
   
2. configure ocfs2-tools-1.2.2 package in UML
    before i run this step, i was upgrade my root filesystem to Ubuntu 8.04 HardyHeron and then i configure ocfs2-tools-1.2.2 package
   
    root <at> none:/usr/local/ocfs2-
tools-1.2.2# ./configure
   
3. i start ocfs2-tools-1.2.2 package with load necessary modules. 
    root <at> none:/usr/local/ocfs2-tools-1.2.2#./vendor/common/o2cb.init load

  Loading module “configfs”: OK
  Mounting configfs filesystem at /sys/kernel/config: OK
  Loading module “ocfs2_nodemanager”: OK
  Loading module “ocfs2_dlm”: OK
  Loading module “ocfs2_dlmfs”: OK

  Mounting ocfs2_dlmfs filesystem at /dlm: OK

4. i copy file cluster.conf to /etc/ocfs2/.
    root <at> none:/usr/local/ocfs2-tools-1.2.2#cat /etc/ocfs/cluster.conf
    node:
            ip_port = 7777
            ip_address = 192.168.0.112
            number = 0
            name = node0
            cluster = ocfs2

    node:
            ip_port = 7777
            ip_address = 192.168.0.116
            number = 0
            name = node1
            cluster = ocfs2    

    cluster:
           node_count = 0
           name = ocfs2

5. i setting hostname and ip address to
    UML0# hostname node0
    UML1# hostname node1
    then i restart both UMLs
 
6. then i run command on host system to setup filesystem to shared device.
   
    host_system$ dd if=/dev/zero of=ocfs seek=$[ 100 * 1024 ] bs=1K count=1
    host_system# uml_mconsole dxhcF8 config ubdbc=ocfs 
   
    where dxhcF8 is umid
    in UML file /dev seen like this
          root <at> node0:# ls -l /dev/ubd*
           /ubda /ubdb /urandom

7. after all process above ,  i have  problem when i try to online OCFS2
    root <at> node0:/usr/local/ocfs2-tools-1.2.2#./vendor/common/o2cb.init online ocfs2
    Starting O2CB cluster ocfs2: Failed
    ./vendor/common/o2cb.init: line 629: o2cb_ctl: command not found
    Stopping O2CB cluster ocfs2: Failed
    ./vendor/common/o2cb.init: line 652: o2cb_ctl: command not found

i was check the file o2cb.init

what happen about that failed ?? maybe i was missing one step to finish setup OCFS2 in UML.

any one can give me advice and explain about failed to online OCFS2 cluster.

thanks.
------------------------------------------------------------------------------
Crystal Reports - New Free Runtime and 30 Day Trial
Check out the new simplified licensing option that enables unlimited
royalty-free distribution of the report engine for externally facing 
server and web deployment.
http://p.sf.net/sfu/businessobjects
_______________________________________________
User-mode-linux-user mailing list
User-mode-linux-user <at> lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-user
shah khadafi | 16 Jun 04:41 2009
Picon

[uml-user] failed online ocfs2 on UML

i try to building OCFS2 cluster on virtual machine. i use UML as virtual machine.

i use ocfs2-tools-1.2.2 package to this project. i was setting the tools OCFS2 like following in chapter 12 USER MODE LINUX book by Jeff Dike.

i use kernel 2.6.24 and 2.6.25 as sources kernel UML. i was planning both kernels to used as nodes cluster.

1. start build OCFS2 as modules kernel to direktory rootfs 2.6.24 and 2.6.25,

<!-- <at> page { margin: 0.79in } P { margin-bottom: 0.08in } A:link { so-language: zxx } -->

  linux-2.6.24#mount Ubuntu-FestyFawn rootfs/ -o loop


    linux-2.6.24# make modules_install INSTALL_MOD_PATH=../rootfs

INSTALL crypto/crypto_hash.ko

INSTALL crypto/hmac.ko

INSTALL crypto/sha1_generic.ko

INSTALL drivers/block/nbd.ko

INSTALL drivers/net/dummy.ko

INSTALL drivers/net/ppp_generic.ko

INSTALL drivers/net/slhc.ko

INSTALL drivers/net/slip.ko

INSTALL fs/autofs/autofs.ko

INSTALL fs/autofs4/autofs4.ko

INSTALL fs/binfmt_misc.ko

INSTALL fs/configfs/configfs.ko

INSTALL fs/dlm/dlm.ko

INSTALL fs/gfs2/gfs2.ko

INSTALL fs/gfs2/locking/dlm/lock_dlm.ko

INSTALL fs/gfs2/locking/nolock/lock_nolock.ko

INSTALL fs/ocfs2/cluster/ocfs2_nodemanager.ko

INSTALL fs/ocfs2/dlm/ocfs2_dlm.ko

INSTALL fs/ocfs2/dlm/ocfs2_dlmfs.ko

INSTALL fs/ocfs2/ocfs2.ko

INSTALL net/sctp/sctp.ko

DEPMOD 2.6.24


   linux-2.6.24#umount ../rootfs   

    same process to source UML kernel 2.6.25.
    i run the 2 kernel UML like this
  
    ./linux ubd0=FestyFawn-i386-rootfs root=/dev/ubda eth0=tuntap,uml-conn0 mem=128M

    ./linux ubd0=../FestyFawn-i386-rootfs root=/dev/ubda eth0=tuntap,uml-conn1 mem=128M
   
    uml-conn0 and uml-conn1 are tunnel interface, before it i was create bridged networking.

2. configure ocfs2-tools-1.2.2 package in UML
    before i run this step, i was upgrade my root filesystem to Ubuntu 8.04 HardyHeron and then i configure ocfs2-tools-1.2.2      package
   
    UML/usr/local/ocfs2-tools-1.2.2# ./configure
   
3. i start ocfs2-tools-1.2.2 package with load necessary modules. 
    root <at> none:/usr/local/ocfs2-tools-1.2.2#./vendor/common/o2cb.init load

  Loading module “configfs”: OK
  Mounting configfs filesystem at /sys/kernel/config: OK
  Loading module “ocfs2_nodemanager”: OK
  Loading module “ocfs2_dlm”: OK
  Loading module “ocfs2_dlmfs”: OK

  Mounting ocfs2_dlmfs filesystem at /dlm: OK

4. i copy file cluster.conf to direktory /etc/ocfs2/.
    root <at> none:/usr/local/ocfs2-tools-1.2.2#cat /etc/ocfs/cluster.conf
    node:
            ip_port = 7777
            ip_address = 192.168.0.112
            number = 0
            name = node0
            cluster = ocfs2

    node:
            ip_port = 7777
            ip_address = 192.168.0.116
            number = 0
            name = node1
            cluster = ocfs2    

    cluster:
           node_count = 0
           name = ocfs2

5. i setting hostname and ip address to
    UML0# hostname node0
    UML1# hostname node1
    then i restart both UMLs
 
6. then i run command on host system to setup filesystem to shared device.
   
    host_system$ dd if=/dev/zero of=ocfs seek=$[ 100 * 1024 ] bs=1K count=1
    host_system# uml_mconsole dxhcF8 config ubdbc=ocfs 
   
    where dxhcF8 is umid
    in UML file /dev seen like this
          root <at> node0:# ls -l /dev/ubd*
           /ubda /ubdb /urandom

7. after all process above ,  i have  problem when i try to online OCFS2
    root <at> node0:/usr/local/ocfs2-tools-1.2.2#./vendor/common/o2cb.init online ocfs2
    Starting O2CB cluster ocfs2: Failed
    ./vendor/common/o2cb.init: line 629: o2cb_ctl: command not found
    Stopping O2CB cluster ocfs2: Failed
    ./vendor/common/o2cb.init: line 652: o2cb_ctl: command not found

i was check the file o2cb.init

what happen about that failed ?? maybe i was missing one step to finish setup OCFS2 in UML ??.

any one can give me advice and explain about failed to online OCFS2 cluster.

thanks.
------------------------------------------------------------------------------
Crystal Reports - New Free Runtime and 30 Day Trial
Check out the new simplified licensing option that enables unlimited
royalty-free distribution of the report engine for externally facing 
server and web deployment.
http://p.sf.net/sfu/businessobjects
_______________________________________________
User-mode-linux-user mailing list
User-mode-linux-user <at> lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-user
shah khadafi | 16 Jun 04:49 2009
Picon

[uml-user] failed online ocfs2 in UML

i try to building OCFS2 cluster on virtual machine. i use UML as virtual machine.

i use ocfs2-tools-1.2.2 package to this project. i was setting the tools OCFS2 like following in chapter 12 USER MODE LINUX book by Jeff Dike.

i use kernel 2.6.24 and 2.6.25 as sources kernel UML. both kernels i use to nodes cluster.

1. start build OCFS2 as modules kernel to direktory rootfs 2.6.24 and 2.6.25,

  linux-2.6.24#mount Ubuntu-FestyFawn rootfs/ -o loop


    linux-2.6.24# make modules_install INSTALL_MOD_PATH=../rootfs

INSTALL crypto/crypto_hash.ko

INSTALL crypto/hmac.ko

INSTALL crypto/sha1_generic.ko

INSTALL drivers/block/nbd.ko

INSTALL drivers/net/dummy.ko

INSTALL drivers/net/ppp_generic.ko

INSTALL drivers/net/slhc.ko

INSTALL drivers/net/slip.ko

INSTALL fs/autofs/autofs.ko

INSTALL fs/autofs4/autofs4.ko

INSTALL fs/binfmt_misc.ko

INSTALL fs/configfs/configfs.ko

INSTALL fs/dlm/dlm.ko

INSTALL fs/gfs2/gfs2.ko

INSTALL fs/gfs2/locking/dlm/lock_dlm.ko

INSTALL fs/gfs2/locking/nolock/lock_nolock.ko

INSTALL fs/ocfs2/cluster/ocfs2_nodemanager.ko

INSTALL fs/ocfs2/dlm/ocfs2_dlm.ko

INSTALL fs/ocfs2/dlm/ocfs2_dlmfs.ko

INSTALL fs/ocfs2/ocfs2.ko

INSTALL net/sctp/sctp.ko

DEPMOD 2.6.24


   linux-2.6.24#umount ../rootfs   

    same process i building OCFS2 as modules kernel to source kernel 2.6.25 too.
    i run the 2 kernel UML like this
  
    ./linux ubd0=FestyFawn-i386-rootfs root=/dev/ubda eth0=tuntap,uml-conn0 mem=128M

    ./linux ubd0=../FestyFawn-i386-rootfs root=/dev/ubda eth0=tuntap,uml-conn1 mem=128M

     uml-conn0 and uml-coon1 are the name tunnel interface, before it i was create bridged networking.
   
2. configure ocfs2-tools-1.2.2 package in UML directory /usr/local.
    before i run this step, i was upgrade my root filesystem to Ubuntu 8.04 HardyHeron.
    then i configure ocfs2-tools-1.2.2 package
   
    UML/usr/local/ocfs2-tools-1.2.2# ./configure
   
3. i start ocfs2-tools-1.2.2 package with load necessary modules. 
    root <at> none:/usr/local/ocfs2-tools-1.2.2#./vendor/common/o2cb.init load

  Loading module “configfs”: OK
  Mounting configfs filesystem at /sys/kernel/config: OK
  Loading module “ocfs2_nodemanager”: OK
  Loading module “ocfs2_dlm”: OK
  Loading module “ocfs2_dlmfs”: OK

  Mounting ocfs2_dlmfs filesystem at /dlm: OK

4. i copy file cluster.conf to /etc/ocfs2/.
    root <at> none:/usr/local/ocfs2-tools-1.2.2#cat /etc/ocfs/cluster.conf
    node:
            ip_port = 7777
            ip_address = 192.168.0.112
            number = 0
            name = node0
            cluster = ocfs2

    node:
            ip_port = 7777
            ip_address = 192.168.0.116
            number = 0
            name = node1
            cluster = ocfs2    

    cluster:
           node_count = 0
           name = ocfs2

5. i setting hostname and ip address to
    UML0# hostname node0
    UML1# hostname node1
    then i restart both UMLs

   where node0 is UML kernel 2.6.24, node1 is UML kernel 2.6.25. 

6. then i run command on host system to setup filesystem to shared device.
   
    host_system$ dd if=/dev/zero of=ocfs seek=$[ 100 * 1024 ] bs=1K count=1
    host_system# uml_mconsole dxhcF8 config ubdbc=ocfs 
   
    where dxhcF8 is umid
    in UML file /dev seen like this
    root <at> node0:# ls -l /dev/ubd*
         /ubda /ubdb /urandom

7. after all process above ,  i have some problem when i try to online OCFS2
    root <at> node0:/usr/local/ocfs2-tools-1.2.2#./vendor/common/o2cb.init online ocfs2
    Starting O2CB cluster ocfs2: Failed
    ./vendor/common/o2cb.init: line 629: o2cb_ctl: command not found
    Stopping O2CB cluster ocfs2: Failed
    ./vendor/common/o2cb.init: line 652: o2cb_ctl: command not found

i was check the file o2cb.init, what happen about that failed ?? maybe i was missing one step to finish setup OCFS2 in UML ??.

any one can give me suggest and explain about failed to online OCFS2 cluster.

thanks.
------------------------------------------------------------------------------
Crystal Reports - New Free Runtime and 30 Day Trial
Check out the new simplified licensing option that enables unlimited
royalty-free distribution of the report engine for externally facing 
server and web deployment.
http://p.sf.net/sfu/businessobjects
_______________________________________________
User-mode-linux-user mailing list
User-mode-linux-user <at> lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-user
Gordon Russell | 18 Jun 17:44 2009
Picon

[uml-user] Compiling user mode linux 2.6.30 in fedora 11

I notice that a few people on the web, plus myself, have found it
impossible to compile user mode linux with fedora 11.
It gets pretty near the end, before dying with:

  CC      init/version.o
  LD      init/built-in.o
  LD      .tmp_vmlinux1
/usr/bin/ld:arch/um/kernel/vmlinux.lds:1: ignoring invalid character
`#' in expression
/usr/bin/ld:arch/um/kernel/vmlinux.lds:1: syntax error
collect2: ld returned 1 exit status

vmlinux.lds has I believe a problem at the beginning, with lines like:
# 1 "arch/um/kernel/vmlinux.lds.S"
# 1 "<built-in>"
# 1 "<command-line>"
# 1 "./include/linux/autoconf.h" 1
# 1 "<command-line>" 2
# 1 "arch/um/kernel/vmlinux.lds.S"

I have tried a few crazy things without much success. gcc is 4.4, ld
is 2.19.51.0.2-17.fc11. It is a 32 bit architecture.
Any thoughts? Same config file seems to work with this kernel on
another of my machines which is still on fedora 10.

Thanks
Gordon.

------------------------------------------------------------------------------
Crystal Reports - New Free Runtime and 30 Day Trial
Check out the new simplified licensing option that enables unlimited
royalty-free distribution of the report engine for externally facing 
server and web deployment.
http://p.sf.net/sfu/businessobjects

Gmane