[SOLVED] VirtualBox complaining after each kernel upgrade

I'm using virtualbox-bin from the AUR (the one with USB support) and after each kernel update this error comes up:
I know it's actually /etc/rc.d/vboxdrv setup
(apparently VirtualBox isn't aware of which distribution it's running on)
This message comes up every once in a while (usually after a kernel upgrade) and I was wondering if I can do something about it... Should I install the DKMS package from the AUR prior to running the above command? Maybe that will shut it up once and for all.
What do you think?
PS: The wiki says to run "vboxbuild" but /usr/bin/vboxbuild doesn't even exist.
Last edited by DSpider (2012-05-15 15:39:13)

Solved it a year later because I got tired of running "sudo /etc/rc.d/vboxdrv setup && sudo modprobe vboxdrv" after every kernel update. Hmph... After reading this and studying the virtualbox-hook package from the AUR, I decided that it's finally time to look into it.
It's easy.
1. Create a "/lib/initcpio/install/vbox" file (or whatever you wanna call it) with the following contents:
# vim: set ft=sh:
build ()
echo "Building VirtualBox modules..."
/etc/rc.d/vboxdrv setup
echo "Ok."
help ()
cat<<HELPEOF
This hook rebuilds VirtualBox modules. It does nothing to the initrd image.
HELPEOF
2. Add it to the "HOOKS" section from /etc/mkinitcpio.conf:
HOOKS="base udev autodetect pata scsi sata usbinput filesystems vbox"
Now the next time the linux and linux-headers packages are updated, the script will kick in and the modules will be rebuilt. I used to have it during boot process and it took forever until it would load X. Haha. I remember I used to go to bathroom just after I turned on the computer... Yeah, it took that long before it was done. Anyway, I asked the author if he can perhaps create a virtualbox-bin-hook package or somehow modify it to also work with virtualbox-bin. We shall see.
Oh, and don't forget to add "vboxdrv" to the "MODULES" section from /etc/rc.conf, if you haven't done so already.

Similar Messages

Acpi_call-git reinstallation after each kernel upgrade.

Hi,
I use acpi_call-git and unfortunately after each kernel upgrade I have to reinstall the application in order to make it work.
Any possible solution?

As you already said, you have to reinstall the package after each kernel upgrade, i do that with acpi_call-git , too.
There is a solution for this called DKMS .
aur package https://aur.archlinux.org/packages.php?ID=54641
Edit: I do not use dkms.
Last edited by teateawhy (2012-08-08 23:59:56)

Is SGEN execution required after a kernel upgrade

Hi experts,
I have done a system kernel upgrade from 700 to 701 version after which all the transactions start compiling as i run them.
I would like to know if SGEN is to be run after a kernel upgrade and if so what option to select as SGEN typically takes long time.
If i start SGEN, and if it takes long can i stop it in between with the intention of not continuing it? would there be a problem?
Please suggest.
Regards,
Sandeep.

Hi,
We are going EHP1 so we upgraded the kernel to 701 level. After that , all transactions are getting compiled so we started SGEN.
I was also of the impression that SGEN need not run after kernel. however, had no other option as transaction compiling was taking long time.
Please check this link http://help.sap.com/saphelp_sm32/helpdata/en/b1/52583c65399965e10000000a114084/frameset.htm
In prerequisites its written :
Applying kernel patches may lead to the invalidation of all loads.
Kindly suggest.
I have stopped the SGEN now.
Due to some constraints, we have to delay the patch upgrade to EHP1. I would like to know whether i can put the old kernel 700 back? and would the SGEN run on kernel 701 would have any problem on that. ?
Is downgrade of kernel possible?
Regards,
Sandeep

Hibernate after a kernel upgrade

Has it ever happened to you that you lost your session because you did a hibernate after a kernel upgrade without rebooting first? I commented on an existing bug upstream, where the outcome was that every distro should handle this by themselves. Later I filled one with arch.
By posting in the forums I want to go more public with it, to see if anyone has had the same problems and if we can come up with some ideas. Mine are in the bug description...
arch bugreport:
http://bugs.archlinux.org/task/17462
upstram bugreport:
http://bugs.freedesktop.org/show_bug.cgi?id=21492

I (and I'm sure, the majority of Arch users) know quite well that a system is unlikely to resume cleanly from hibernation after a kernel upgrade. If anything, the issue here is quickly finishing up whatever else you were doing, and then forgetting the need to reboot. I know I could find myself doing this quite easily.
While its not a huge issue (and I wouldn't really like to see a handful of packages modified in an attempt to support this), having a means to implement this in pacman would be very nice. Specifically, a hooks system, such as Allan has described here, would be ideal. In this particular case, those who wish to prevent hibernation (or, perhaps an alternate solution which would keep the system hibernate-able) would have a simple mechanism with which to implement it.
Of course, while a hook system in pacman would be handy, it would also be quite an undertaking.

Is a reboot necessary after a kernel upgrade? [SOLVED/Answered =)]

Hey folks, longtime arch user (since ~2006), first time registration/post.
Just a dead-simple question: is it necessary for me to reboot my Arch systems after 'pacman -Syu' kernel upgrades? I've always done it out of sheer habit coming from a WIndows Sysadmin world where every darn update needs a reboot.
Thanks in advance.
Last edited by train_wreck (2011-10-22 20:55:46)

Wow, thanks y'all for the fast replies. Sounds like my instincts were right .
I have to say I'm blown away by the reliability of this Linux distro. Like I implied, my day job has for years required me to interact with Windows client & server OSes on a daily basis, and the level of random issues/headaches that crop up in that ecosystem is rather astounding. In contrast, I've had as of yet zero problems with my 2 Arch servers. I've had the same essential configs since the first installations of them years ago, and barring a couple reinstalls due to hard drive failures they've been smooth as silk the whole time. Kudos to that.

Restart is required to mount USB after a kernel upgrade?

I have problem with required restart after kernel upgrade couse can't use USB. This is common in my two machines (laptop & pc). What can I do with this annoying problem. I've tried to google for an answer but can't find anything usefull. Only reasonable info was from stackexchange, but it won't work. I still have all old kernels in /lub/modules
ls /lib/modules/
3.10.10-1-ARCH 3.10.5-1-ARCH 3.11.1-1-ARCH 3.11.2-1-ARCH 3.8.10-1-ARCH 3.8.6-1-ARCH 3.8.8-1-ARCH 3.9.3-1-ARCH 3.9.5-1-ARCH 3.9.7-1-ARCH 3.9.9-1-ARCH
3.10.3-1-ARCH 3.10.9-1-ARCH 3.11.1-2-ARCH 3.11.4-1-ARCH 3.8.11-1-ARCH 3.8.7-1-ARCH 3.9.2-1-ARCH 3.9.4-1-ARCH 3.9.6-1-ARCH 3.9.8-1-ARCH extramodules-3.11-ARCH

Maybe I can't do such obvious thing like loading usb module or it just won't work:).
$ ls /etc/modules-load.d/ -1
enp2s0.conf
loop.conf
usbcore.conf
virtualbox.conf
$ cat /etc/modules-load.d/usbcore.conf
usbcore
$ lsmod | grep usb
usbhid 41434 0
hid 89751 2 hid_generic,usbhid
usbcore 178119 5 ohci_hcd,ohci_pci,ehci_hcd,ehci_pci,usbhid
usb_common 1648 1 usbcore
$ dmesg | tail -n2
[ 142.891160] CPUM: APIC 03 at 00000000fee00000 (mapped at ffffc900145a4000) - ver 0x80050010, lint0=0x10700 lint1=0x10400 pc=0x00400 thmr=0x10000
[ 984.213568] usb 7-6: new high-speed USB device number 2 using ehci-pci
And usbdrive still don't mounted automatically

Clean up after each system upgrade

Hi,
After each upgrade my system increase its disk usage.
I type "pacman -Qet", and there is a lot of packets, but I'm not sure is it safe to delete them.
Which is the best rule to follow about cleaning the system, using pacman?
Last edited by clovenhoof (2010-09-22 20:18:47)

clovenhoof wrote:Is it really I have to know what each packet does, in order to delete unused packets manually?
How will you know if it's needed for anything if you have no idea what that package does?
An example: what does 'git' do and why do I have it installed? Pacman shows it's not required by any package. Ah, I'm accessing some git repositories - that's why I have it installed.
You have only a couple hundred packages installed, so 'pacman -Qi <packagename>' away :-)
'pacman -Sc' should be enough to get rid of old & unneeded packages.
Last edited by karol (2010-09-22 21:01:47)

System less responsive after udev/kernel upgrade

Yeah, another udev/kernel upgrade question. After upgrading udev and the kernel yesterday my terminal is less responsive. For example, scrolling is slow and jerky, and uses an inordanite amount of cpu. ls /usr/bin will cause the cpu usage to spike ~70% which is much higher than normal. I can cause the cpu to go to 100% by scrolling quickly in vim. I do not see this in firefox, however. Scrolling there is responsive and does not use any additional cpu beyond what it took before the update.
I really don't know where to start to troubleshoot this - the only thing I have noticed is that the speedstep-centrino module is not loading any longer.
pmorris@arch:~ $ sudo modprobe speedstep-centrino
FATAL: Error inserting speedstep_centrino (/lib/modules/2.6.15-ARCH/kernel/arch/i386/kernel/cpu/cpufreq/speedstep-centrino.ko): Device or resource busy
This is on an ibm thinkpad T40p. I am still getting cpu scaling, but the system is generally less responsive. It may be taking longer to scale up the cpu frequency on demand. Any help is appreciated. Thanks.

tpowa wrote:perhaps a bug in how udev loads the cpufreq modules. can you tell me which modules you would need? then just blacklist the modules you don't need...
Thanks, this worked. Prior to the upgrade I was using the speedstep-centrino module, and loading it in the modules array in /etc/rc.conf. After the upgrade the acpi-cpufreq module was being autoloaded at boot prior to my trying to load the speedstep-centrino module, hence the "device or resource busy" failure when I tried to load the speedstep module. After blacklisting all modules listed in /lib/modules/2.6.15-ARCH/kernel/arch/i386/kernel/cpu/cpufreq except the speedstep-centrino, things are working fine. The correct module is loaded at boot time.
Is there a cleaner method to achieve this other than blacklisting all unwanted modules?

After pacman kernel upgrade, modules won't load

I can't seem to load any of my modules anymore, so I can't even get my network up (via_rhine). I am posting this from a different computer.
My uname -r says:
2.6.11.7-ARCH
but my last kernel upgrade message in /var/log/pacman.log says:
uprgaded kernel26 (2.6.10.3 -> 2.6.11.3-1)
So, what's up here?

that is kinda odd - usually when you upgrade a kernel, here's what happens:
running kernel 2.6.11.1
modules are in /lib/modules/2.6.11.1/*
upgrade 2.6.11.1->2.6.11.7
dir /lib/modules/2.6.11.1 removed
dir /lib/modules/2.6.11.7 created and filled
(modprobes will fail now)
reboot, init scripts will call "depmod -a" to tell modprobe that modules moved and to cache locations
maybe the depmod didn't get called?

Ath5k probs after new kernel upgrade (/kernel26 2.6.34.3-1)

I just did a pacman -Syu and the new kernel /kernel26 2.6.34.3-1 causes issues with my ath based card. Ath5k is the kernel module I've used, and now I get a bunch of calibration errors in dmesg. Is anyone else having this issue?
I'd rather not go back to madwifi...but I will if I have to...ath5k has been pretty solid since it's come out.

See my post here: https://bbs.archlinux.org/viewtopic.php?id=102751
I had problems but I am using ath9k. I partially solved mine by getting the kernel out of testing. Not sure if the two issues are related.
Good luck!!!

VMWARE not working after latest kernel upgrade [SOLVED]

I've done a kernel update and after rebooting VMWare was not working. That's "standard", I figured, because I have a new kernel.
So, I run the any_any patch (as usual) to re-configure the VM to the new kernel but this time no joy
Is there anyone else having the same problem?
The error seems to be here:
In file included from /tmp/vmware-config0/vmmon-only/common/hostKernel.h:56,
from /tmp/vmware-config0/vmmon-only/common/task.c:24:
include/asm/page.h: In function 'pte_t native_make_pte(long unsigned int)':
include/asm/page.h:112: error: expected primary-expression before ')' token
include/asm/page.h:112: error: expected ';' before '{' token
include/asm/page.h:112: error: expected primary-expression before '.' token
include/asm/page.h:112: error: expected `;' before '}' token
make[2]: *** [/tmp/vmware-config0/vmmon-only/common/task.o] Error 1
make[1]: *** [_module_/tmp/vmware-config0/vmmon-only] Error 2
make[1]: Leaving directory `/usr/src/linux-2.6.22-ARCH'
make: *** [vmmon.ko] Error 2
make: Leaving directory `/tmp/vmware-config0/vmmon-only'
Unable to build the vmmon module.
R.
Note: Running /usr/bin/vmware-config.pl is of no help either (same error).
Last edited by ralvez (2007-07-28 13:58:06)

stonecrest wrote:Yes, and the PKGBUILD works;
You'd rather say "I, myself, me, thinks it works".
Because it does not work for many people, including me.
The modules do build with kernel 2.6.22.X, but the network behaviour using bridged networking is extremely clumsy (used to work fine with 2.6.21.X), and shared folders are a pain in the butt- for example whenever I try to copy a file from the guest to a host shared folder, the guest OS (winXP) comes close to totally freezing, and I have to kill explorer to stop the transaction.
Do you call THAT "works"?
IMO the issues with kernel 2.6.22 won't be addressed before a new VMWare version is out... Until then, yeah... the old version/PKGBUILD works...
I'm close to ditching VMWare in favour of VirtualBox, but a couple of VMWare features (mainly related to low level hardware access) still prevent me from doing so.
Last edited by scarecrow (2007-07-29 13:31:51)

Network keeps failing after recent kernel upgrade

I upgraded my kernel this morning to the newest version of 2.6.17, and as soon as I rebooted the Network started saying FAIL each time. Can anyone help me diagnose the problem?
Side note :
I remembered that I had also modified the inittab to use gdm before I rebooted. Would that have any kind of effect?

I've had that problem as well. It's because the modular kernels load the modules at the same time, and because of that which module is initialized first isn't the same at every boot. So the names eth0, eth1 and eth2 for example may change at next boot. Even if you have only one network card, the ethernet over firewire module might be loaded. I solved it by setting eth0, eth1 and eth2 to dhcp, starting them all at boot and starting network in the background (to prevent wating for the two devices that won't start). To find out what devicenames are used, do a "ifconfig -a"
IIRC there is another solution too - you can search through the forums for it.
ATM I'm running a custom kernel with all the network modules built-in - that way the network device names are the same at every boot.

(SOLVED) Virtualbox 2.2 pervents system upgrade

There doesn't seem to be a way to uninstall VirtualBox 2.2 PUEL which I installed from the Sun site. Now I can't upgrade my system. I get the following error:
bash-3.2# pacman -Syu --ignore virtualbox-modules
:: Synchronizing package databases...
core is up to date
extra is up to date
community is up to date
:: Starting full system upgrade...
:: Replace libxml++2 with extra/libxml++? [Y/n]
resolving dependencies...
warning: dependency cycle detected:
warning: xorg-server will be installed before its nvidia-utils dependency
looking for inter-conflicts...
error: failed to prepare transaction (could not satisfy dependencies)
:: virtualbox-modules: requires kernel26<2.6.29
VirtualBox is pretty much a toy to me, I'd rather just get rid of it if it's going to cause problems with maintaining my system.
Last edited by mooreted (2009-04-11 06:13:42)

wonder wrote:--ignore virtual-modules will not do a thing because there is no update for that package. you have that error because that packages _HAS_ as dependency kernel 2.6.28.
if is a toy remove it and install virtualbox_bin for aur
pacman -Rns virtualbox-ose virtualbox-modules
I haven't found a way to remove it. I guess I can find everything name "VirtualBox" or "virtualbox" and delete them all.

[SOLVED]Pacman complains when trying to upgrade

Hello all.
Trying to run "pacman -Syu" I get the following errors:
error: failed to commit transaction (conflicting files)
kernel26-firmware: /lib/firmware/bnx2x-e1-4.8.53.0.fw exists in filesystem
kernel26-firmware: /lib/firmware/bnx2x-e1h-4.8.53.0.fw exists in filesystem
kernel26-firmware: /lib/firmware/cis/3CCFEM556.cis exists in filesystem
kernel26-firmware: /lib/firmware/cis/3CXEM556.cis exists in filesystem
kernel26-firmware: /lib/firmware/cxgb3/t3fw-7.4.0.bin exists in filesystem
Errors occurred, no packages were upgraded.
I suspected that it had something to do with an old custom kernel I'm not really using anymore, but not sure. Running "pacman -Qo" just gives:
error: No package owns /lib/firmware/bnx2x-e1-4.8.53.0.fw
Anyone had any ideas? Help much appreciated.
Last edited by rune0077 (2009-10-11 17:15:45)

Yup, I just ran it with the force option and everything installed and upgraded fine.
Thanks.

Not booting after latest kernel upgrade x86_64

The new kernel 2.6.23.9 refuses to boot - spits out something about filesystems, says cd rom sata all (hooks) ndiswrapper, forcedeth blah blah, continously call traces happening but not booting. Not even in fallback mode.
I will try to boot thru live CD, but what should I alter? mkinitcpio.
This is on 64 bit only. My 686 desktop does not have this issue (atleast one of them). The other one I am using now. I don't want to upgrade this too and become incommunicado as the other desktop is in the office.
This is happening too often with Arch nowadays. It automatically upgraded the nvidia module and asked me to log out of X and remove the module manually. But I chose to reboot and this is what happened.

Actually it is screwing up at the loading modules stage
it says bug soft lockup cpu#0 stuck for 11s [rc.sysinit:1782]
modules linked in ndiswrapper forcedeth ext3 jbd mbcache sd_mod sr_mod cdrom ehci_hcd usbcore sata_nv ata_generic pata_amd libata
Pid : 1782 comm: rc.sysinit Tainted: P D: 2.6.23-ARCH#1
Then a lot of gibberish about RIP etc. This is continuously looping

[SOLVED] VirtualBox complaining after each kernel upgrade

Similar Messages

Maybe you are looking for