4 Jul
2014
4 Jul
'14
9:48 p.m.
Hi,
I’ve updated my VPS to Wheezy. Unfortunately, I can’t get it to boot.
Xen console returns
Booting instance: travelkazoo
Using config file "/etc/xen/travelkazoo.conf".
Error: Boot loader didn't return any data!
I booted into rescue, mounted my root partition, chrooted, mounted /sys, /proc, and
/dev/pts, and tried to figure out where I went wrong.
I am not very familiar with xen - am I supposed to run xm create in /etc/xen? My previous
squeeze system’s backups do not have a travelkazoo.conf file in /etc/xen. xm create
/etc/xen/travelkazoo.conf doesn’t work. xend does not exist, even though I have
xen-linux-system-686-pae installed, which is supposed to include it. So it obviously
cannot run.
xen utils installed is version 4.1, as is xen-hypervisor-4.1-i386 (arch is i386).
update-grub2 works fine, /boot/grub2/grub.cfg exists and looks plausible, I’ve moved
/etc/grub.d/10_linux to 50_linux so 20_linux_xen is first in line once I run update-grub2.
I’ve been searching like mad and can’t seem to find anything else to try.
Can someone please give me a hint what I’m missing, or do I need to reinstall squeeze from
scratch and restore from backup? :(
Here are all relevant packages installed:
ii libc6-xen:i386 2.13-38+deb7u1 i386
Embedded GNU C Library: Shared libraries [Xen version]
ii libxen-4.1 4.1.4-3+deb7u1 i386
Public libs for Xen
ii libxenstore3.0 4.1.4-3+deb7u1 i386
Xenstore communications library for Xen
ii linux-image-2.6-xen-686 2.6.32+29 i386
Linux 2.6 for modern PCs (meta-package), Xen dom0 support
ii linux-image-2.6.32-5-xen-686 2.6.32-48squeeze4 i386
Linux 2.6.32 for modern PCs, Xen dom0 support
rc linux-modules-2.6.18-6-xen-686 2.6.18.dfsg.1-26etch2 i386
Linux 2.6.18 modules on i686
ii xen-hypervisor-4.0-amd64 4.0.1-5.11 i386
The Xen Hypervisor on AMD64
ii xen-hypervisor-4.1-i386 4.1.4-3+deb7u1 i386
Xen Hypervisor on i386
ii xen-linux-system-3.2.0-4-686-pae 3.2.57-3+deb7u2 i386
Xen system with Linux 3.2 on modern PCs (meta-package)
ii xen-linux-system-686-pae 3.2+46 i386
Xen system with Linux for modern PCs (meta-package)
ii xen-system-i386 4.1.4-3+deb7u1 i386
Xen System on i386 (meta-package)
4 Jul
4 Jul
9:59 p.m.
New subject: Xen problem: can't boot after update to Debian Wheezy
Hi John,
On Fri, Jul 04, 2014 at 11:48:37PM +0200, John Morgan Salomon wrote:
Hi,
I’ve updated my VPS to Wheezy. Unfortunately, I can’t get it to boot.
Xen console returns
Booting instance: travelkazoo
Using config file "/etc/xen/travelkazoo.conf".
Error: Boot loader didn't return any data!
This indicates that the /boot/grub/menu.lst file cannot be parsed.
Maybe it does not exist, or contains the wrong things. On wheezy it
would be provided by the package grub-legacy.
I am not very familiar with xen - am I supposed to run
xm create in /etc/xen?
Nope, you don't even need an /etc/xen directory; the reference to
/etc/xen/travelkazoo.conf above is the config file of your VPS on
BitFolk's server.
My previous squeeze system’s backups do not have a
travelkazoo.conf file in /etc/xen. xm create /etc/xen/travelkazoo.conf doesn’t work.
xend does not exist, even though I have xen-linux-system-686-pae installed, which is
supposed to include it. So it obviously cannot run.
A guest virtual machine running under Xen does not require xend,
that's purely for the host (dom0) machine. In fact since Debian
lenny you require no Xen-specific software whatsoever.
For Debian wheezy the correct kernel package is linux-image-686-pae
which will pull in the specific kernel linux-image-3.2.0-4-686-pae.
xen utils installed is version 4.1, as is
xen-hypervisor-4.1-i386 (arch is i386). update-grub2 works fine, /boot/grub2/grub.cfg
exists and looks plausible, I’ve moved /etc/grub.d/10_linux to 50_linux so 20_linux_xen is
first in line once I run update-grub2. I’ve been searching like mad and can’t seem to
find anything else to try.
You should not be sing grub2, so this is probably the problem. You
need a /boot/grub/menu.lst which is provided by grub-legacy.
grub-legacy was probably removed during upgrade. If you install
grub-legacy it will probably want to remove grub2, which is fine.
It sounds like booting into rescue VM, chroot into install and
install grub-legacy will most likely fix things.
Can someone please give me a hint what I’m missing, or
do I need to reinstall squeeze from scratch and restore from backup? :(
Worst case would be to install Debian wheezy from scratch. Notes for
that are here:
https://tools.bitfolk.com/wiki/Self-install
but I don't think it will come to that.
Here are all relevant packages installed:
ii libc6-xen:i386 2.13-38+deb7u1 i386
Embedded GNU C Library: Shared libraries [Xen version]
ii libxen-4.1 4.1.4-3+deb7u1 i386
Public libs for Xen
ii libxenstore3.0 4.1.4-3+deb7u1 i386
Xenstore communications library for Xen
ii linux-image-2.6-xen-686 2.6.32+29 i386
Linux 2.6 for modern PCs (meta-package), Xen dom0 support
ii linux-image-2.6.32-5-xen-686 2.6.32-48squeeze4 i386
Linux 2.6.32 for modern PCs, Xen dom0 support
rc linux-modules-2.6.18-6-xen-686 2.6.18.dfsg.1-26etch2 i386
Linux 2.6.18 modules on i686
ii xen-hypervisor-4.0-amd64 4.0.1-5.11 i386
The Xen Hypervisor on AMD64
ii xen-hypervisor-4.1-i386 4.1.4-3+deb7u1 i386
Xen Hypervisor on i386
ii xen-linux-system-3.2.0-4-686-pae 3.2.57-3+deb7u2 i386
Xen system with Linux 3.2 on modern PCs (meta-package)
ii xen-linux-system-686-pae 3.2+46 i386
Xen system with Linux for modern PCs (meta-package)
ii xen-system-i386 4.1.4-3+deb7u1 i386
Xen System on i386 (meta-package)
None of the above are required. You can remove them all and install
linux-image-686-pae.
Cheers,
Andy
--
http://bitfolk.com/ -- No-nonsense VPS hosting
10:32 p.m.
New subject: Xen problem: can't boot after update to Debian Wheezy
Hi Andy,
thanks for the reply.
OK, so I purged all the xen-related packages and configs, and installed grub-legacy and
linux-image-686-pae. dpkg-query -l |grep xen returns nothing
/boot/grub/menu.lst looks OK, but to be safe I backed it up and generated a new one with
update-grub, and zeroed out /boot/grub2/grub.cfg
still, same problem when I drop back into console, destroy the running instance, and try
to boot.
Using config file "/etc/xen/travelkazoo.conf”.
Error: Boot loader didn't return any data!
Here are the non-default entries in /boot/grub/menu.lst:
title Debian GNU/Linux, kernel 3.2.0-4-686-pae
root
kernel /boot/vmlinuz-3.2.0-4-686-pae root= ro
initrd /boot/initrd.img-3.2.0-4-686-pae
title Debian GNU/Linux, kernel 3.2.0-4-686-pae (single-user mode)
root
kernel /boot/vmlinuz-3.2.0-4-686-pae root= ro single
initrd /boot/initrd.img-3.2.0-4-686-pae
Anything else you can suggest for me to try?
Best,
-John
On 4Jul 2014, at 23:59, Andy Smith <andy(a)bitfolk.com> wrote:
Hi John,
On Fri, Jul 04, 2014 at 11:48:37PM +0200, John Morgan Salomon wrote:
Hi,
I’ve updated my VPS to Wheezy. Unfortunately, I can’t get it to boot.
Xen console returns
Booting instance: travelkazoo
Using config file "/etc/xen/travelkazoo.conf".
Error: Boot loader didn't return any data!
This indicates that the /boot/grub/menu.lst file cannot be parsed.
Maybe it does not exist, or contains the wrong things. On wheezy it
would be provided by the package grub-legacy.
I am not very familiar with xen - am I supposed
to run xm create in /etc/xen?
Nope, you don't even need an /etc/xen directory; the reference to
/etc/xen/travelkazoo.conf above is the config file of your VPS on
BitFolk's server.
My previous squeeze system’s backups do not have
a travelkazoo.conf file in /etc/xen. xm create /etc/xen/travelkazoo.conf doesn’t work.
xend does not exist, even though I have xen-linux-system-686-pae installed, which is
supposed to include it. So it obviously cannot run.
A guest virtual machine running under Xen does not require xend,
that's purely for the host (dom0) machine. In fact since Debian
lenny you require no Xen-specific software whatsoever.
For Debian wheezy the correct kernel package is linux-image-686-pae
which will pull in the specific kernel linux-image-3.2.0-4-686-pae.
xen utils installed is version 4.1, as is
xen-hypervisor-4.1-i386 (arch is i386). update-grub2 works fine, /boot/grub2/grub.cfg
exists and looks plausible, I’ve moved /etc/grub.d/10_linux to 50_linux so 20_linux_xen is
first in line once I run update-grub2. I’ve been searching like mad and can’t seem to
find anything else to try.
You should not be sing grub2, so this is probably the problem. You
need a /boot/grub/menu.lst which is provided by grub-legacy.
grub-legacy was probably removed during upgrade. If you install
grub-legacy it will probably want to remove grub2, which is fine.
It sounds like booting into rescue VM, chroot into install and
install grub-legacy will most likely fix things.
Can someone please give me a hint what I’m
missing, or do I need to reinstall squeeze from scratch and restore from backup? :(
Worst case would be to install Debian wheezy from scratch. Notes for
that are here:
https://tools.bitfolk.com/wiki/Self-install
but I don't think it will come to that.
Here are all relevant packages installed:
ii libc6-xen:i386 2.13-38+deb7u1 i386
Embedded GNU C Library: Shared libraries [Xen version]
ii libxen-4.1 4.1.4-3+deb7u1 i386
Public libs for Xen
ii libxenstore3.0 4.1.4-3+deb7u1 i386
Xenstore communications library for Xen
ii linux-image-2.6-xen-686 2.6.32+29 i386
Linux 2.6 for modern PCs (meta-package), Xen dom0 support
ii linux-image-2.6.32-5-xen-686 2.6.32-48squeeze4 i386
Linux 2.6.32 for modern PCs, Xen dom0 support
rc linux-modules-2.6.18-6-xen-686 2.6.18.dfsg.1-26etch2 i386
Linux 2.6.18 modules on i686
ii xen-hypervisor-4.0-amd64 4.0.1-5.11 i386
The Xen Hypervisor on AMD64
ii xen-hypervisor-4.1-i386 4.1.4-3+deb7u1 i386
Xen Hypervisor on i386
ii xen-linux-system-3.2.0-4-686-pae 3.2.57-3+deb7u2 i386
Xen system with Linux 3.2 on modern PCs (meta-package)
ii xen-linux-system-686-pae 3.2+46 i386
Xen system with Linux for modern PCs (meta-package)
ii xen-system-i386 4.1.4-3+deb7u1 i386
Xen System on i386 (meta-package)
None of the above are required. You can remove them all and install
linux-image-686-pae.
Cheers,
Andy
--
http://bitfolk.com/ -- No-nonsense VPS hosting
_______________________________________________
users mailing list
users(a)lists.bitfolk.com
https://lists.bitfolk.com/mailman/listinfo/users
11:32 p.m.
New subject: Xen problem: can't boot after update to Debian Wheezy
Hi John,
On Sat, Jul 05, 2014 at 12:32:40AM +0200, John Morgan Salomon wrote:
kernel /boot/vmlinuz-3.2.0-4-686-pae root= ro
The "root=" part is not valid; it's specifying nothing for the root
filesystem device. Your old menu.lst file probably the info you
need. There is a line somewhere in the file that starts with:
# kopt=...
That is where you put such options as the "root=" business that will
be present in every entry. update-grub reads that line and creates
the entries from it.
For example, one of my wheezy VMs has this:
# kopt=root=UUID=5658a3af-0344-4bc4-8ba6-4ca2cc6cf429 ro console=hvc0
The "root=" part should be the block device that has your root
filesystem on. Yours will be either /dev/xvda or /dev/xvda1. I
generally recommend using the UUID of the filesystem, that way if
the device name changes then things still work.
From the rescue VM you can see with
$ cat /proc/partitions
whether you have a partitioned disk or not (if there's an xvda1 it's
partitioned). But I can save you some time and tell you that you
don't. Therefore your root filesystem is on /dev/xvda.
You can tell the UUID using:
$ sudo tune2fs -l /dev/xvda | grep UUID
but I can save you some time and tell you:
Filesystem UUID: 8a67d497-9377-48b8-910e-b979c6f405e0
So your kopt line should be:
# kopt=root=UUID=8a67d497-9377-48b8-910e-b979c6f405e0 ro console=hvc0
Once you have edited that into the file
$ sudo update-grub.
I suspect it will then boot.
Cheers,
Andy
--
http://bitfolk.com/ -- No-nonsense VPS hosting
5 Jul
5 Jul
5:54 p.m.
New subject: Xen problem: can't boot after update to Debian Wheezy
Hi Andy,
thanks, that solved it (at least my Grub problems, it’s still not booting but that’s an
issue with a startup script somewhere).
Much appreciated.
Best,
-John
On 5Jul 2014, at 01:32, Andy Smith <andy(a)bitfolk.com> wrote:
Hi John,
On Sat, Jul 05, 2014 at 12:32:40AM +0200, John Morgan Salomon wrote:
kernel /boot/vmlinuz-3.2.0-4-686-pae root= ro
The "root=" part is not valid; it's specifying nothing for the root
filesystem device. Your old menu.lst file probably the info you
need. There is a line somewhere in the file that starts with:
# kopt=...
That is where you put such options as the "root=" business that will
be present in every entry. update-grub reads that line and creates
the entries from it.
For example, one of my wheezy VMs has this:
# kopt=root=UUID=5658a3af-0344-4bc4-8ba6-4ca2cc6cf429 ro console=hvc0
The "root=" part should be the block device that has your root
filesystem on. Yours will be either /dev/xvda or /dev/xvda1. I
generally recommend using the UUID of the filesystem, that way if
the device name changes then things still work.
From the rescue VM you can see with
$ cat /proc/partitions
whether you have a partitioned disk or not (if there's an xvda1 it's
partitioned). But I can save you some time and tell you that you
don't. Therefore your root filesystem is on /dev/xvda.
You can tell the UUID using:
$ sudo tune2fs -l /dev/xvda | grep UUID
but I can save you some time and tell you:
Filesystem UUID: 8a67d497-9377-48b8-910e-b979c6f405e0
So your kopt line should be:
# kopt=root=UUID=8a67d497-9377-48b8-910e-b979c6f405e0 ro console=hvc0
Once you have edited that into the file
$ sudo update-grub.
I suspect it will then boot.
Cheers,
Andy
--
http://bitfolk.com/ -- No-nonsense VPS hosting
_______________________________________________
users mailing list
users(a)lists.bitfolk.com
https://lists.bitfolk.com/mailman/listinfo/users
7:43 p.m.
New subject: Xen problem: can't boot after update to Debian Wheezy
Hi John,
On Sat, Jul 05, 2014 at 07:54:56PM +0200, John Morgan Salomon wrote:
thanks, that solved it (at least my Grub problems,
it’s still not booting but that’s an issue with a startup script somewhere).
Glad to hear you have your grub issues sorted. But what's this other
booting issue?
Perhaps we can help.
Cheers,
Andy
--
The optimum programming team size is 1.
Has
Jurassic Park taught us nothing?
— pfilandr
8:34 p.m.
New subject: Xen problem: can't boot after update to Debian Wheezy
Hi Andy,
sure, that’d be welcome - I wasn’t sure whether this list was for bitfolk/xen-specific
issues, so I thought I’d try and resolve it on my own first.
The issue is that when I try to boot into multiuser/runlevel 3, once the startup scripts
pass a certain point the system hangs. Carriage return/control-C on console seem to
register (I get newlines) but with no effect - so the console connection seems to be ok. I
have to destroy the instance from the console and reboot to recover. There’s no exact
script where it hangs - sometimes S03mysql, sometimes S03lighttp, sometimes S03ssh
I am able to boot into single-user, thanks to your help. I manually get networking
working, as well as start sshd. However, when I ctrl-D out of single user mode, it
quickly hangs.
I suspected it might have something to do with networking, as I had ntpd enabled and get
regular “could not resolve <ntp server>” errors even when the rest of the boot
process hangs - and even once sshd starts, it’s not reachable from outside. I’m not sure
why networking wouldn’t be starting ; /etc/network/interfaces looks ok, and as mentioned,
I’m able to configure it manually in single user.
Syslog at debug level tells me nothing - the last entry I see is usually something about
MySQL table checking, but even disabling MySQL (and other services where it hangs) in
/etc/rc3.d doesn’t make a difference.
I’m close to just reinstalling at this point, but if this sounds at all familiar, any
hints are welcome.
Thank you,
-John
On 5Jul 2014, at 21:43, Andy Smith <andy(a)bitfolk.com> wrote:
Hi John,
On Sat, Jul 05, 2014 at 07:54:56PM +0200, John Morgan Salomon wrote:
thanks, that solved it (at least my Grub
problems, it’s still not booting but that’s an issue with a startup script somewhere).
Glad to hear you have your grub issues sorted. But what's this other
booting issue?
Perhaps we can help.
Cheers,
Andy
--
The optimum programming team size is 1.
Has Jurassic Park taught us nothing?
— pfilandr
_______________________________________________
users mailing list
users(a)lists.bitfolk.com
https://lists.bitfolk.com/mailman/listinfo/users
9:36 p.m.
New subject: Xen problem: can't boot after update to Debian Wheezy
Hi John,
On Sat, Jul 05, 2014 at 10:34:17PM +0200, John Morgan Salomon wrote:
The issue is that when I try to boot into
multiuser/runlevel 3, once the startup scripts pass a certain point the system hangs.
Carriage return/control-C on console seem to register (I get newlines) but with no effect
- so the console connection seems to be ok. I have to destroy the instance from the
console and reboot to recover. There’s no exact script where it hangs - sometimes
S03mysql, sometimes S03lighttp, sometimes S03ssh
All of those are after the networking has been brought up, so I
would expect networking to be working at that point if there weren't
any error messages to the contrary.
Perhaps you just have no getty running on that console.
Could you provide:
- A log of the full boot up process
- The output of this:
grep ^co /etc/inittab
(or the path to your /etc/inittab file if you're doing that from the
rescue VM)
Cheers,
Andy
--
http://bitfolk.com/ -- No-nonsense VPS hosting
10:06 p.m.
New subject: Xen problem: can't boot after update to Debian Wheezy
Hi Andy,
here’s the rub - I also thought networking had started, but I’m getting sendmail errors
about “could not find any active network interfaces” on console after letting it stew for
a while. Iptables is disabled.
As mentioned - I _am_ getting console messages - just up to the point where some of those
S03 services start, then only the occasional error messages.
I’m no longer using the rescue VM, because it’s finally able to find the bootloader
(again, thanks).
the info you asked for:
ganesha:/# grep ^co /etc/inittab
ganesha:/#
I’ve attached the output of dmesg, bootlog, and syslog (I cleared these before rebooting
so they should only show a “bad” boot and the subsequent boot into single user mode after
I destroy the instance from console)
Best,
-John
On 5Jul 2014, at 23:36, Andy Smith <andy(a)bitfolk.com> wrote:
Hi John,
On Sat, Jul 05, 2014 at 10:34:17PM +0200, John Morgan Salomon wrote:
The issue is that when I try to boot into
multiuser/runlevel 3, once the startup scripts pass a certain point the system hangs.
Carriage return/control-C on console seem to register (I get newlines) but with no effect
- so the console connection seems to be ok. I have to destroy the instance from the
console and reboot to recover. There’s no exact script where it hangs - sometimes
S03mysql, sometimes S03lighttp, sometimes S03ssh
All of those are after the networking has been brought up, so I
would expect networking to be working at that point if there weren't
any error messages to the contrary.
Perhaps you just have no getty running on that console.
Could you provide:
- A log of the full boot up process
- The output of this:
grep ^co /etc/inittab
(or the path to your /etc/inittab file if you're doing that from the
rescue VM)
Cheers,
Andy
--
http://bitfolk.com/ -- No-nonsense VPS hosting
_______________________________________________
users mailing list
users(a)lists.bitfolk.com
https://lists.bitfolk.com/mailman/listinfo/users
11:24 p.m.
New subject: Xen problem: can't boot after update to Debian Wheezy
On Sun, Jul 06, 2014 at 12:06:48AM +0200, John Morgan Salomon wrote:
Hi Andy,
here’s the rub - I also thought networking had started, but I’m getting sendmail errors
about “could not find any active network interfaces” on console after letting it stew for
a while. Iptables is disabled.
Hmm, possible then that there is a syntax error in
/etc/network/interfaces or something of that nature.
As mentioned - I _am_ getting console messages - just
up to the point where some of those S03 services start, then only the occasional error
messages.
But if you don't have a login prompt on the console then the console
isn't working. It could be that it never completes the boot, but
later on you say your inittab has no "co" line so you definitely
don't have a working console (insofar as it letting you log in
there).
the info you asked for:
ganesha:/# grep ^co /etc/inittab
ganesha:/#
This indicates lack of a getty running on the console device which
is why you get no login prompt I expect.
Here's what an /etc/inittab file looks like on one of my wheezy VMs,
with comments stripped out:
$ egrep -v '^(#|$)' /etc/inittab
id:2:initdefault:
si::sysinit:/etc/init.d/rcS
~:S:wait:/sbin/sulogin
l0:0:wait:/etc/init.d/rc 0
l1:1:wait:/etc/init.d/rc 1
l2:2:wait:/etc/init.d/rc 2
l3:3:wait:/etc/init.d/rc 3
l4:4:wait:/etc/init.d/rc 4
l5:5:wait:/etc/init.d/rc 5
l6:6:wait:/etc/init.d/rc 6
z6:6:respawn:/sbin/sulogin
ca:12345:ctrlaltdel:/sbin/shutdown -t1 -a -r now
pf::powerwait:/etc/init.d/powerfail start
pn::powerfailnow:/etc/init.d/powerfail now
po::powerokwait:/etc/init.d/powerfail stop
1:2345:respawn:/sbin/getty 38400 tty1
2:23:respawn:/sbin/getty 38400 tty2
3:23:respawn:/sbin/getty 38400 tty3
4:23:respawn:/sbin/getty 38400 tty4
5:23:respawn:/sbin/getty 38400 tty5
6:23:respawn:/sbin/getty 38400 tty6
co:2345:respawn:/sbin/getty --noclear hvc0 9600 linux
Notice the "co" line; you don't have one, probably why no
"login:"
prompt. If you add one like mine¹ and then:
# telinit q
to restart init, a login prompt should appear on the console.
I’ve attached the output of dmesg, bootlog, and syslog
(I cleared these before rebooting so they should only show a “bad” boot and the subsequent
boot into single user mode after I destroy the instance from console)
The actual output on the console would be more interesting. You can
just copy and paste it into a paste bin, e.g. pastie.org. Usual GNU
Screen commands will allow you to scroll all the way back up, e.g.
ctrl-a [ to enter copy mode, pg up then to scroll up through it, pg
dn and finally escape to get out of that.
The content of /etc/network/interfaces might be good as well while
we are here.
Cheers,
Andy
¹ Stock Debian wheezy doesn't have the --noclear flag; I add this to
mine to prevent the boot messages being cleared off the console
when the login prompt appears.
--
http://bitfolk.com/ -- No-nonsense VPS hosting
11:32 p.m.
New subject: Xen problem: can't boot after update to Debian Wheezy
Hiya,
You were spot on about getty - I added that line to inittab, and I have a console login.
Awesome, inching ever-closer to this. Thanks!
I still don’t have network, but at least now I can diagnose it from its booted-up state.
I need to have a look at this tomorrow after the frustration of the last game today - will
report back if I find out what it was.
Best,
-John
On 6Jul 2014, at 01:24, Andy Smith <andy(a)bitfolk.com> wrote:
On Sun, Jul 06, 2014 at 12:06:48AM +0200, John Morgan
Salomon wrote:
Hi Andy,
here’s the rub - I also thought networking had started, but I’m getting sendmail errors
about “could not find any active network interfaces” on console after letting it stew for
a while. Iptables is disabled.
Hmm, possible then that there is a syntax error in
/etc/network/interfaces or something of that nature.
As mentioned - I _am_ getting console messages -
just up to the point where some of those S03 services start, then only the occasional
error messages.
But if you don't have a login prompt on the console then the console
isn't working. It could be that it never completes the boot, but
later on you say your inittab has no "co" line so you definitely
don't have a working console (insofar as it letting you log in
there).
the info you asked for:
ganesha:/# grep ^co /etc/inittab
ganesha:/#
This indicates lack of a getty running on the console device which
is why you get no login prompt I expect.
Here's what an /etc/inittab file looks like on one of my wheezy VMs,
with comments stripped out:
$ egrep -v '^(#|$)' /etc/inittab
id:2:initdefault:
si::sysinit:/etc/init.d/rcS
~:S:wait:/sbin/sulogin
l0:0:wait:/etc/init.d/rc 0
l1:1:wait:/etc/init.d/rc 1
l2:2:wait:/etc/init.d/rc 2
l3:3:wait:/etc/init.d/rc 3
l4:4:wait:/etc/init.d/rc 4
l5:5:wait:/etc/init.d/rc 5
l6:6:wait:/etc/init.d/rc 6
z6:6:respawn:/sbin/sulogin
ca:12345:ctrlaltdel:/sbin/shutdown -t1 -a -r now
pf::powerwait:/etc/init.d/powerfail start
pn::powerfailnow:/etc/init.d/powerfail now
po::powerokwait:/etc/init.d/powerfail stop
1:2345:respawn:/sbin/getty 38400 tty1
2:23:respawn:/sbin/getty 38400 tty2
3:23:respawn:/sbin/getty 38400 tty3
4:23:respawn:/sbin/getty 38400 tty4
5:23:respawn:/sbin/getty 38400 tty5
6:23:respawn:/sbin/getty 38400 tty6
co:2345:respawn:/sbin/getty --noclear hvc0 9600 linux
Notice the "co" line; you don't have one, probably why no
"login:"
prompt. If you add one like mine¹ and then:
# telinit q
to restart init, a login prompt should appear on the console.
I’ve attached the output of dmesg, bootlog, and
syslog (I cleared these before rebooting so they should only show a “bad” boot and the
subsequent boot into single user mode after I destroy the instance from console)
The actual output on the console would be more interesting. You can
just copy and paste it into a paste bin, e.g. pastie.org. Usual GNU
Screen commands will allow you to scroll all the way back up, e.g.
ctrl-a [ to enter copy mode, pg up then to scroll up through it, pg
dn and finally escape to get out of that.
The content of /etc/network/interfaces might be good as well while
we are here.
Cheers,
Andy
¹ Stock Debian wheezy doesn't have the --noclear flag; I add this to
mine to prevent the boot messages being cleared off the console
when the login prompt appears.
--
http://bitfolk.com/ -- No-nonsense VPS hosting
_______________________________________________
users mailing list
users(a)lists.bitfolk.com
https://lists.bitfolk.com/mailman/listinfo/users
11:34 p.m.
New subject: Xen problem: can't boot after update to Debian Wheezy
By the way in the meantime here are the contents of /etc/network/interfaces:
auto lo
iface lo inet loopback
auto eth0
iface eth0 inet static
address 85.119.82.201
netmask 255.255.255.0
gateway 85.119.82.1
-John
On 6Jul 2014, at 01:24, Andy Smith <andy(a)bitfolk.com> wrote:
On Sun, Jul 06, 2014 at 12:06:48AM +0200, John Morgan
Salomon wrote:
Hi Andy,
here’s the rub - I also thought networking had started, but I’m getting sendmail errors
about “could not find any active network interfaces” on console after letting it stew for
a while. Iptables is disabled.
Hmm, possible then that there is a syntax error in
/etc/network/interfaces or something of that nature.
As mentioned - I _am_ getting console messages -
just up to the point where some of those S03 services start, then only the occasional
error messages.
But if you don't have a login prompt on the console then the console
isn't working. It could be that it never completes the boot, but
later on you say your inittab has no "co" line so you definitely
don't have a working console (insofar as it letting you log in
there).
the info you asked for:
ganesha:/# grep ^co /etc/inittab
ganesha:/#
This indicates lack of a getty running on the console device which
is why you get no login prompt I expect.
Here's what an /etc/inittab file looks like on one of my wheezy VMs,
with comments stripped out:
$ egrep -v '^(#|$)' /etc/inittab
id:2:initdefault:
si::sysinit:/etc/init.d/rcS
~:S:wait:/sbin/sulogin
l0:0:wait:/etc/init.d/rc 0
l1:1:wait:/etc/init.d/rc 1
l2:2:wait:/etc/init.d/rc 2
l3:3:wait:/etc/init.d/rc 3
l4:4:wait:/etc/init.d/rc 4
l5:5:wait:/etc/init.d/rc 5
l6:6:wait:/etc/init.d/rc 6
z6:6:respawn:/sbin/sulogin
ca:12345:ctrlaltdel:/sbin/shutdown -t1 -a -r now
pf::powerwait:/etc/init.d/powerfail start
pn::powerfailnow:/etc/init.d/powerfail now
po::powerokwait:/etc/init.d/powerfail stop
1:2345:respawn:/sbin/getty 38400 tty1
2:23:respawn:/sbin/getty 38400 tty2
3:23:respawn:/sbin/getty 38400 tty3
4:23:respawn:/sbin/getty 38400 tty4
5:23:respawn:/sbin/getty 38400 tty5
6:23:respawn:/sbin/getty 38400 tty6
co:2345:respawn:/sbin/getty --noclear hvc0 9600 linux
Notice the "co" line; you don't have one, probably why no
"login:"
prompt. If you add one like mine¹ and then:
# telinit q
to restart init, a login prompt should appear on the console.
I’ve attached the output of dmesg, bootlog, and
syslog (I cleared these before rebooting so they should only show a “bad” boot and the
subsequent boot into single user mode after I destroy the instance from console)
The actual output on the console would be more interesting. You can
just copy and paste it into a paste bin, e.g. pastie.org. Usual GNU
Screen commands will allow you to scroll all the way back up, e.g.
ctrl-a [ to enter copy mode, pg up then to scroll up through it, pg
dn and finally escape to get out of that.
The content of /etc/network/interfaces might be good as well while
we are here.
Cheers,
Andy
¹ Stock Debian wheezy doesn't have the --noclear flag; I add this to
mine to prevent the boot messages being cleared off the console
when the login prompt appears.
--
http://bitfolk.com/ -- No-nonsense VPS hosting
_______________________________________________
users mailing list
users(a)lists.bitfolk.com
https://lists.bitfolk.com/mailman/listinfo/users
6 Jul
6 Jul
12:25 a.m.
New subject: Xen problem: can't boot after update to Debian Wheezy
Hi John,
On Sun, Jul 06, 2014 at 01:34:17AM +0200, John Morgan Salomon wrote:
auto eth0
iface eth0 inet static
address 85.119.82.201
netmask 255.255.255.0
gateway 85.119.82.1
Netmask and gateway are incorrect. Netmask should be 255.255.248.0,
gateway should be 85.119.80.1. This may not be the issue but it
certainly won't be helping.
If that doesn't fix it, being logged in to the console and seeing
what "ifdown eth0" and "ifup eth0" does might get us further.
Cheers,
Andy
--
The optimum programming team size is 1.
Has
Jurassic Park taught us nothing?
— pfilandr
4:12 p.m.
New subject: Xen problem: can't boot after update to Debian Wheezy
Hi Andy,
thanks for pointing that out - I must have had an old config, or it was
overwritten - not sure.
ifupdown was not installed; installing that fixed it persistently across
reboots.
Now I'm just struggling with lighttpd untraceably refusing to start up,
but that's something I'm leaving for another day. If I'd known an
upgrade would cause this many problems....when will you start supporting
FreeBSD? :-)
Best regards, and many many thanks again for your patience and help. I
hope the solutions to all my missteps help someone else at some point.
-John
On 2014-07-06 02:25, Andy Smith wrote:
Hi John,
On Sun, Jul 06, 2014 at 01:34:17AM +0200, John Morgan Salomon wrote:
auto eth0
iface eth0 inet static
address 85.119.82.201
netmask 255.255.255.0
gateway 85.119.82.1
Netmask and gateway are incorrect. Netmask should be 255.255.248.0,
gateway should be 85.119.80.1. This may not be the issue but it
certainly won't be helping.
If that doesn't fix it, being logged in to the console and seeing
what "ifdown eth0" and "ifup eth0" does might get us further.
Cheers,
Andy
_______________________________________________
users mailing list
users(a)lists.bitfolk.com
https://lists.bitfolk.com/mailman/listinfo/users 
4:36 p.m.
New subject: Xen problem: can't boot after update to Debian Wheezy
John, You seem to have had some bad luck there! I did the Squeeze to Wheezy
upgrade and it went so smoothly! Luck of the draw I guess
On 6 July 2014 17:12, John Morgan Salomon <john(a)zog.net> wrote:
Hi Andy,
thanks for pointing that out - I must have had an old config, or it was
overwritten - not sure.
ifupdown was not installed; installing that fixed it persistently across
reboots.
Now I'm just struggling with lighttpd untraceably refusing to start up,
but that's something I'm leaving for another day. If I'd known an upgrade
would cause this many problems....when will you start supporting FreeBSD?
:-)
Best regards, and many many thanks again for your patience and help. I
hope the solutions to all my missteps help someone else at some point.
-John
On 2014-07-06 02:25, Andy Smith wrote:
--
Keith Williams
Farang Can Learn Thai www.farangcanlearnthai.com
Keith's Place www.keiths-place.co.uk
Tailor Made English www.tmenglish.org
West Norfolk RSPCA www.westnorfolkrspca.org.uk
Hi John,
On Sun, Jul 06, 2014 at 01:34:17AM +0200, John Morgan Salomon wrote:
_______________________________________________
users mailing list
users(a)lists.bitfolk.com
https://lists.bitfolk.com/mailman/listinfo/users
auto eth0
iface eth0 inet static
address 85.119.82.201
netmask 255.255.255.0
gateway 85.119.82.1
Netmask and gateway are incorrect. Netmask should be 255.255.248.0,
gateway should be 85.119.80.1. This may not be the issue but it
certainly won't be helping.
If that doesn't fix it, being logged in to the console and seeing
what "ifdown eth0" and "ifup eth0" does might get us further.
Cheers,
Andy
_______________________________________________
users mailing list
users(a)lists.bitfolk.com
https://lists.bitfolk.com/mailman/listinfo/users
4:56 p.m.
New subject: Xen problem: can't boot after update to Debian Wheezy
Yeah, I have _no_ idea what happened there - all the instructions I read
seemed really straightforward.
I am still struggling with boatloads of library incompatibilities and
mismatched dependencies, but at least it's allowing me to dump some of
the cruft I've accumulated.
-John
On 2014-07-06 18:36, Keith Williams wrote:
John, You seem to have had some bad luck there! I did
the Squeeze to
Wheezy upgrade and it went so smoothly! Luck of the draw I guess
On 6 July 2014 17:12, John Morgan Salomon <john(a)zog.net
<mailto:john@zog.net>> wrote:
Hi Andy,
thanks for pointing that out - I must have had an old config, or
it was overwritten - not sure.
ifupdown was not installed; installing that fixed it persistently
across reboots.
Now I'm just struggling with lighttpd untraceably refusing to
start up, but that's something I'm leaving for another day. If
I'd known an upgrade would cause this many problems....when will
you start supporting FreeBSD? :-)
Best regards, and many many thanks again for your patience and
help. I hope the solutions to all my missteps help someone else
at some point.
-John
On 2014-07-06 02:25, Andy Smith wrote:
Hi John,
On Sun, Jul 06, 2014 at 01:34:17AM +0200, John Morgan Salomon wrote:
_______________________________________________
users mailing list
users(a)lists.bitfolk.com <mailto:users@lists.bitfolk.com>
https://lists.bitfolk.com/mailman/listinfo/users
--
Keith Williams
Farang Can Learn Thai www.farangcanlearnthai.com
<http://www.farangcanlearnthai.com>
Keith's Place www.keiths-place.co.uk <http://www.keiths-place.co.uk>
Tailor Made English www.tmenglish.org <http://www.tmenglish.org>
West Norfolk RSPCA www.westnorfolkrspca.org.uk
<http://www.westnorfolkrspca.org.uk>
_______________________________________________
users mailing list
users(a)lists.bitfolk.com
https://lists.bitfolk.com/mailman/listinfo/users auto eth0
iface eth0 inet static
address 85.119.82.201
netmask 255.255.255.0
gateway 85.119.82.1
Netmask and gateway are incorrect. Netmask should be 255.255.248.0,
gateway should be 85.119.80.1. This may not be the issue but it
certainly won't be helping.
If that doesn't fix it, being logged in to the console and seeing
what "ifdown eth0" and "ifup eth0" does might get us further.
Cheers,
Andy
_______________________________________________
users mailing list
users(a)lists.bitfolk.com <mailto:users@lists.bitfolk.com>
https://lists.bitfolk.com/mailman/listinfo/users 
7 Jul
7 Jul
9:04 a.m.
New subject: Xen problem: can't boot after update to Debian Wheezy
John said:
Yeah, I have _no_ idea what happened there - all the
instructions I read
seemed really straightforward.
Did you use script to record what happened, as recommended in the Debian
upgrade instructions?
It is really useful when something goes wrong because you can look back
and go 'Oh, yes, it did say that it was going to uninstall mysql-server...'
Ian
2:16 p.m.
New subject: Xen problem: can't boot after update to Debian Wheezy
Embarrassingly, I missed that bit, but it’s certainly something I will pay attention to
next upgrade time.
The majority of my problems appear to have come from things like deprecated options in
various package config files that, rather than being nicely ignored in new versions, cause
stuff to grind to a screeching halt, or packages not being correctly installed that would
be required for new ways of doing various things. Unfortunately, the documentation and
forum posts I found on most of the issues I fixed weren’t very good at giving specific
pointers, so I’m grateful for the help I got here.
-John
On 7Jul 2014, at 11:04, Ian <ian(a)lovingboth.com> wrote:
John said:
Yeah, I have _no_ idea what happened there - all
the instructions I read
seemed really straightforward.
Did you use script to record what happened, as recommended in the Debian
upgrade instructions?
It is really useful when something goes wrong because you can look back
and go 'Oh, yes, it did say that it was going to uninstall mysql-server...'
Ian
_______________________________________________
users mailing list
users(a)lists.bitfolk.com
https://lists.bitfolk.com/mailman/listinfo/users
2:50 p.m.
New subject: Xen problem: can't boot after update to Debian Wheezy
John said:
The majority of my problems appear to have come from
things like
deprecated options in various package config files that, rather
than being nicely ignored in new versions, cause stuff to grind to
a screeching halt, or packages not being correctly installed that
would be required for new ways of doing various things.
The only big config file change I can remember is Dovecot, where it's
a completely new system.
The other two issues were that mysql-server got removed and needed to
be reinstalled, and although it doesn't seem to be in the Debian
release notes's list of obsolete packages, sysklogd has gone. So if
you use that for syslogd, you need to install inetutils-syslogd instead.
More details in an email here on the 21st May last year.
Ian
3:24 p.m.
New subject: Xen problem: can't boot after update to Debian Wheezy
Hi Ian,
On Mon, Jul 07, 2014 at 03:50:46PM +0100, Ian wrote:
The other two issues were that mysql-server got
removed and needed to
be reinstalled,
Could possibly be the result of having a specific versioned package
like "mysql-server-5.0" installed rather than the generic
"mysql-server"?
and although it doesn't seem to be in the Debian
release notes's
list of obsolete packages, sysklogd has gone. So if you use that
for syslogd, you need to install inetutils-syslogd instead.
Way back in Debian lenny syslogd replaced by rsyslog, no
inetutils-syslogd required, most users should notice no difference:
https://www.debian.org/releases/lenny/i386/release-notes/ch-whats-new#syste…
Those upgrading the same install from etch through lenny, squeeze
and to wheezy could possibly have never switched to rsyslog and may
have been caught by this:
https://wiki.debian.org/NewInWheezy#Upgrade_issues
"rsyslog: Upgrades from systems that did not already have
rsyslog installed may not have any syslog daemon running after
the upgrade. After your upgrade, check whether rsyslog is
installed, and install it if necessary."
Cheers,
Andy
--
"I'm /extremely/ miffed about today's events and in my quest to try to make
you understand the level of my unhappiness, I'm likely to use an awful lot
of what we would call /violent sexual imagery/ and I just wanted to check
that neither of you would be terribly offended by that." — Malcolm Tucker
11:38 p.m.
New subject: Xen problem: can't boot after update to Debian Wheezy
Andy said:
Very probably - see that email referred to - although I'm not sure why
I would have done that (possibly just copying packages from the co-op
I was with previously?)
The other two
issues were that mysql-server got removed and
needed to be reinstalled,
Could possibly be the result of having a specific versioned
package like "mysql-server-5.0" installed rather than the generic
"mysql-server"? Those upgrading the same install from etch through
lenny, squeeze
and to wheezy could possibly have never switched to rsyslog and
may have been caught by this:
Again, it's possible that a list of packages was copied. The previous
config had its origins in Woody before being upgraded over the years
as far as Lenny(?)
Ian
3:34 p.m.
New subject: Xen problem: can't boot after update to Debian Wheezy
Hello,
Just a reminder to those not wishing to undertake a possibly painful
major upgrade: there is an offer to provide a new VPS free for two
weeks in order for you to migrate your services over to:
https://tools.bitfolk.com/wiki/Migrating_to_a_new_VPS
This is of course just an option, not the recommended way to
upgrade.
Cheers,
Andy
--
http://bitfolk.com/ -- No-nonsense VPS hosting
3820
days inactive
3823
days old
21 comments
4 participants
participants (4)
-
Andy Smith -
Ian -
John Morgan Salomon -
Keith Williams