[LITMUS^RT] HELP:NFS+KVM Tutorial

Mon Jan 20 18:08:30 CET 2014

Again, when you see this type of error, it’s because your kernel expects to find NFS as a module.  A module is like a shared library (i.e., *.so) for the OS.  Modules are good for general use, but are a pain to work with during Litmus development.  We suggest that you avoid them.

Module files are denoted with a “.ko” suffix and are located in /lib/modules/<kernel name>/*.  That is, modules are stored on the filesystem.  In this case, your kernel is looking for the NFS module for your specific kernel in the qcow2.img file.  Of course, there are no modules in your image file because you compiled the kernel within your host environment, not the guest environment.  To get NFS working, you need to compile NFS support DIRECTLY INTO the kernel—analogous to a static library (i.e., *.a).

To do so:
1) Bring up your compilation configuration with “make menuconfig”
2) Select “File systems”
3) Navigate towards the bottom of the list and select “Network File Systems”.  Hit space-bar to ensure that there is a “[*]” next to the option.
4) With “Network File Systems” selected, hit enter/return to select more options.
5) Enable the NFS features that you need (probably just “NFS client subsystem”, but you can select the others to, just to be safe).  Keep hitting space-bar until the item is selected with “<*>”.  The “*” means that it will be compiled into the kernel.  “<M>” means that the feature will be compiled as a module.  Don’t compile as a module!!!!  <*> good.  <M> bad.

These instructions should guide you in the right direction, but I don’t have time or ability to replicate your particular system configuration to test them.  You may have to spend some more time on your own, perhaps searching google for answers, to finally get things to work.  There is no shortage of Linux kernel tutorials online.  You can always try googling a particular error code or message.  I’m sorry that we don’t have a complete turn-key configuration for Litmus.  Unfortunately, Linux doesn’t really either (at least for x86), so it’s all but impossible for us to provide one.

-Glenn

On Jan 20, 2014, at 11:45 AM, Sanjib Das <cnt.sanjib at googlemail.com> wrote:

> Hi,
> here is the last few lines before the login prompt come up using this command 
> "qemu-system-x86_64 -enable-kvm -cpu host -smp 2 -hda /home/sdas/images/ubuntu.
> backing.qcow2.img -m 1024 -name "ubuntu-qemu-cjk" -nographic -kernel "/home/sdas/litmus/litmus-rt/arch/x86_64/boot/bzImage" -append "console=ttyS0,115200 root=/dev/hda1" -gdb tcp::12345 -net nic -net user,hostfwd=::2222-:22
> "
> 
> OUTPUT::
> 
> modprobe: chdir(3.10.5-litmus2012.2+): No such file or directory
> Begin: Loading essential drivers ... done.
> Begin: Running /scripts/init-premount ... done.
> Begin: Mounting root file system ... Begin: Running /scripts/local-top ... modprobe: chdir(3.10.5-litmus2012.2+): No such file or directory
> done.
> Begin: Running /scripts/local-premount ... done.
> modprobe: chdir(3.10.5-litmus2012.2+): No such file or directory
> [   13.709380] EXT4-fs (hda1): mounted filesystem with ordered data mode. Opts: (null)
> Begin: Running /scripts/local-bottom ... done.
> done.
> Begin: Running /scripts/init-bottom ... done.
> 
> Last login: Mon Jan 20 10:59:16 EST 2014 on ttyS0
> Welcome to Ubuntu 11.10 (GNU/Linux 3.10.5-litmus2012.2+ x86_64)
> 
>  * Documentation:  https://help.ubuntu.com/
> Your Ubuntu release is not supported anymore.
> For upgrade information, please visit:
> http://www.ubuntu.com/releaseendoflife
> 
> New release '12.04.3 LTS' available.
> Run 'do-release-upgrade' to upgrade to it.
> 
> Thanks in advance 
> Sanjib
> 
> 
> 
> On Mon, Jan 20, 2014 at 5:36 PM, Sanjib Das <cnt.sanjib at googlemail.com> wrote:
> Dear Andea and Glenn,
> 
> Thank you very much.
> Now I am able to make an nfs and also share data. but only with the following command 
> 'qemu-system-x86_64 -enable-kvm -cpu host -smp 2 -hda /home/sdas/images/ubuntu.backing.qcow2.img -m 1024 -name "ubuntu-qemu-cjk" -nographic -net nic -net user,hostfwd=::2222-:22'
> 
> But it use this command "qemu-system-x86_64 -enable-kvm -cpu host -smp 2 -hda /home/sdas/images/ubuntu.backing.qcow2.img -m 1024 -name "ubuntu-qemu-cjk" -nographic -kernel "/home/sdas/litmus/litmus-rt/arch/x86_64/boot/bzImage" -append "console=ttyS0,115200 root=/dev/hda1" -gdb tcp::12345 -net nic -net user,hostfwd=::2222-:22"
> 
> 'modprobe nfs' returns error 'Can't read module /lib/modules........................'
> 
> I guessed I have to use -initrd option in qemu-system-x86_63 command, make bzImage is not creating any initrd or initramfs file.
> 
> Any suggestion with cordially accepted. 
> 
> Thanks in advance
> 
> Sanjib
> 
> On Mon, Jan 20, 2014 at 2:48 PM, Andrea Bastoni <bastoni at sprg.uniroma2.it> wrote:
> 
> On Jan 20, 2014 2:40 PM, "Glenn Elliott" <gelliott at cs.unc.edu> wrote:
> >
> > Comments inline.
> >
> > On Jan 20, 2014, at 2:17 AM, Sanjib Das <cnt.sanjib at googlemail.com> wrote:
> >
> >> Hi Glenn,
> >>
> >> ssh -p 2222 realtime at localhost Command requires a password, where I gave 'realtime' and is not working.
> >> sdas at debian:~$ ssh -p 2222 realtime at localhost
> >> realtime at localhost's password:
> >
> >
> > Well, at least it’s working.  It looks like we’ve lost any documentation of the passwords.
> 
> Hi all,
> 
> Perhaps silly question, but did someone already try root root?
> 
> I believe I initially set up our qemu like that, but they've been reworked quite a bit since that.
> 
> Thanks,
> Andrea
> 
> > However, you say that you can log in when -vga is working.  So do this:
> > 1) Boot with -vga.  This will dump you into a root-privelged console.
> > 2) Do “passwd”
> > 3) Pick any password you want for the root user.  (You can also reset the password of the “realtime” user if you wish.)
> > 4) Shutdown
> >
> >> FYI:"
> >>
> >> kvm -smp 2 -m 512 -boot c -vga std -net nic -net user,hostfwd=tcp::10022-:22 -kernel path_to_kernel/arch/x86/boot/bzImage -append root=/dev/hda1 -hda kvm_image.img
> >>
> >> "
> >> also ask for user-name and password, whats exactly i don't know. But if I use -nographic  instead of  -vga std it doesn't ask for pass.
> >
> >
> > 5) Now relaunch with “-nographic”.  Doing “ssh -p 2222 root at localhost” should now work using the password you set in step #3 above.
> >
> >> And also
> >>
> >> sdas at debian:~$    sudo chmod u+s /usr/local/libexec/qemu-bridge-helper
> >> chmod: cannot access `/usr/local/libexec/qemu-bridge-helper': No such file or directory
> >> sdas at debian:~$
> >>
> >> It seems to be a bug.
> >
> >
> > Not a bug.  You just don’t have qemu-bridge-helper installed at /usr/local/libexec/.  On my 12.x version of Ubuntu, it’s at /usr/lib/qemu-bridge-helper (“> find /usr -name “*qemu*” is a brute force way of searching for a file).  I’m not sure about Debian.  I found an old bug report that indicates Debian (at least in 2012) may not have it installed (http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=691138).
> >
> >> And nfsd output
> >> sdas at debian:~$ sudo tail -f /var/log/messages | grep -i 'nfs'
> >> Jan 20 07:41:24 debian kernel: [ 2768.813543] nfsd: last server has exited, flushing export cache
> >> Jan 20 07:41:26 debian kernel: [ 2770.506707] NFSD: Using /var/lib/nfs/v4recovery as the NFSv4 state recovery directory
> >> Jan 20 07:41:26 debian kernel: [ 2770.506741] NFSD: starting 90-second grace period
> >>
> >>
> >> Thanks in advance
> >> Sanjib
> >>
> >>
> >>
> >>
> >>
> >>
> >> On Mon, Jan 20, 2014 at 1:19 AM, Glenn Elliott <gelliott at cs.unc.edu> wrote:
> >>>
> >>>
> >>> On Jan 19, 2014, at 7:07 PM, Sanjib Das <cnt.sanjib at googlemail.com> wrote:
> >>>
> >>>> Dear Concern,
> >>>>
> >>>> Can any one give me some suggestion about file sharing. Cause I am following the link https://wiki.litmus-rt.org/litmus/LiblitmusViaNFS but still not succeeded.
> >>>>
> >>>> I am trying to do this with the following command 'qemu-system-x86_64 -enable-kvm -cpu host -smp 2 -hda /home/sdas/images/ubuntu.backing.qcow2.img -m 1024 -name "ubuntu-qemu-cjk" -nographic -kernel "/home/sdas/litmus/litmus-rt/arch/x86_64/boot/bzImage" -append "console=ttyS0,115200 root=/dev/hda1" -gdb tcp::12345 -net nic -net user,hostfwd=::2222-:22' 
> >>>>
> >>>> And as a result eth0 in side the guest is configured automatically 
> >>>>
> >>>> From guest to host  'ping 10.0.2.2' is working
> >>>> "root at ubuntu-qemu:~# ping 10.0.2.2
> >>>> PING 10.0.2.2 (10.0.2.2) 56(84) bytes of data.
> >>>> 64 bytes from 10.0.2.2: icmp_req=1 ttl=255 time=2.28 ms
> >>>> 64 bytes from 10.0.2.2: icmp_req=2 ttl=255 time=0.515 ms
> >>>> ^C
> >>>> --- 10.0.2.2 ping statistics ---
> >>>> 2 packets transmitted, 2 received, 0% packet loss, time 1001ms
> >>>> rtt min/avg/max/mdev = 0.515/1.398/2.281/0.883 ms
> >>>> root at ubuntu-qemu:~# 
> >>>> "
> >>>> But Host to guest 'ping 10.0.2.15' not working
> >>>>
> >>>> "sdas at debian:~$ ping 10.0.2.15
> >>>> PING 10.0.2.15 (10.0.2.15) 56(84) bytes of data.
> >>>> ^C
> >>>> --- 10.0.2.15 ping statistics ---
> >>>> 2 packets transmitted, 0 received, 100% packet loss, time 1006ms
> >>>>
> >>>> sdas at debian:~$ 
> >>>> "
> >>>>
> >>>> ip route guest :
> >>>> root at ubuntu-qemu:~# ip route
> >>>> default via 10.0.2.2 dev eth0  metric 100 
> >>>> 10.0.2.0/24 dev eth0  proto kernel  scope link  src 10.0.2.15 
> >>>> root at ubuntu-qemu:~# 
> >>>>
> >>>>
> >>>> ip route host:
> >>>>
> >>>> sdas at debian:~$ ip route
> >>>> default via 192.168.0.1 dev eth0 
> >>>> 169.254.0.0/16 dev eth0  scope link  metric 1000 
> >>>> 192.168.0.0/24 dev eth0  proto kernel  scope link  src 192.168.0.101 
> >>>> sdas at debian:~$ 
> >>>>
> >>>> my host is debian 7 wheezy.
> >>>>
> >>>> Any quick suggestion/ alternate tutorial / link/  will very helpful.
> >>>
> >>>
> >>>
> >>> I don’t believe that direct network access from the host to the guest is necessary to get NFS to work.  With respect to host->guest networking, you can ssh into the guest OS by doing “ssh -p 2222 realtime at localhost”.  If you want to set up a virtual network, I believe these instructions will be useful: http://wiki.qemu.org/Features/HelperNetworking
> >>>
> >>> Anyway, assuming that your NFS server on the host is working and your guest can ping the host, I believe your guest OS should be able to mount the host’s network drive.  You’ve already shown that the guest can ping the host.  Are you sure NFS is working on the host?  What do the NFS logs say?
> >>>
> >>> -Glenn
> >>>
> >>>
> >>> _______________________________________________
> >>> litmus-dev mailing list
> >>> litmus-dev at lists.litmus-rt.org
> >>> https://lists.litmus-rt.org/listinfo/litmus-dev
> >>>
> >>
> >> _______________________________________________
> >> litmus-dev mailing list
> >> litmus-dev at lists.litmus-rt.org
> >> https://lists.litmus-rt.org/listinfo/litmus-dev
> >
> >
> >
> > _______________________________________________
> > litmus-dev mailing list
> > litmus-dev at lists.litmus-rt.org
> > https://lists.litmus-rt.org/listinfo/litmus-dev
> >
> 
> _______________________________________________
> litmus-dev mailing list
> litmus-dev at lists.litmus-rt.org
> https://lists.litmus-rt.org/listinfo/litmus-dev
> 
> 
> 
> _______________________________________________
> litmus-dev mailing list
> litmus-dev at lists.litmus-rt.org
> https://lists.litmus-rt.org/listinfo/litmus-dev

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.litmus-rt.org/pipermail/litmus-dev/attachments/20140120/47785e8f/attachment.html>