Thanks for the reply! I'm a little confused as to your answer to (b) though. We're using static IPs, though we're not using network-manager or conman. If what you're saying is the case, what automatically did the conversion when we did this from Buster to Beowulf?

EDIT: Oh also: Do you have any idea which Debian repo net-tools would be in? As I noted the repos that Rackspace was using didn't have that. Here's their original sources.list:

/etc/apt/sources.list-ORIG 

deb http://mirror.rackspace.com/debian bookworm main non-free-firmware
deb-src http://mirror.rackspace.com/debian bookworm main non-free-firmware

deb http://mirror.rackspace.com/debian-security bookworm-security main non-free-firmware
deb-src http://mirror.rackspace.com/debian-security bookworm-security main non-free-firmware

# bookworm-updates, to get updates before a point release is made;
# see https://www.debian.org/doc/manuals/debian-reference/ch02.en.html#_updates_and_backports
deb http://mirror.rackspace.com/debian bookworm-updates main non-free-firmware
deb-src http://mirror.rackspace.com/debian bookworm-updates main non-free-firmware

I see that they don't have a backports repo in there. Is that where it would be?

Thanks!
Tom

Thanks! I literally just found that and tried that just before you posted that, and that worked. I was about to post here to ask if that was correct.

Thanks again!
Tom

Yup...Adding the auto declaration in the interfaces file was our issue. Now we should be able to configure all that manually. As far as I know that was the only issue with the migration.

I'm going to mark this as solved. However after we work through other things like getting the rackspace services going I may post back here on that for other's future reference.

Thanks again!
Tom

OMG. This is so screwed I almost don't know where to start. So to be clear as to what I was trying here...I was assuming/hoping that getting the VM to use the same interface names before and after the conversion would get all this to work. What I just went through trying that almost defies description:

So first of all, adding net.ifnames=0 to the kernel parameters (and I did make sure that I actually saw that option in /proc/cmdline) did NOT cause it to use the enX0/enX1 interface names...they were still set to eth0/eth1 after rebooting. After that I stumbled on this which just amazed me: Apparently Debian Bookworm DOES in fact use interface names like enX0(??):

https://yeri.be/bookworm-eth0-enx0/

I found it stunning that I didn't find more on that one. What I did was to add the reserve of that to the interfaces file...that is:

rename eth0=enX0
rename eth1=enX1

So what I did next was to try the conversion over again with a new Debian 12 VM. I made the above change first, and then went through the migration steps.Sure enough, after the reboot, the above DID cause it to use the same interface names as the original VM...that is enX0/enX1. However, that STILL didn't properly configure the /etc/network/interfaces file for some reason. It was left untouched.

I'm totally lost as to what part of this whole process would do that but it's clearly not happening. But stuff gets WAY worse than that:

Next we painstakingly created the the entries in /etc/network/interfaces manually. This is where it gets really strange. While doing "ifup enX0" does in fact bring up the interface properly, stopping and starting /etc/init.d/networking did NOT bring them up. Oddly enough, running "/etc/init.d/networking stop" DOES apparently do an "ifdown", yet doing "/etc/init.d/networking start" apparently does NOT do an ifup on the configured interfaces at all!

That last one is the show-stopper now for sure. Everything seems correct at that point. What on earth would prevent that service from running ifup on the configured interfaces??

Wow this is painful. We really need this server, and I'm desperately hoping I don't end up having to use Debiam as-is with systemd, but we're limited to what Rackspace offers.

ANY suggestions are welcome. This is just beyond painful.

Tom

Thanks for the reply!

As to that "Rackspace's Debian Bookworm VM", that's just the Devian 12 VM that Rackspace offers. As far as I'm aware you can't just install your own OS on servers there, and at least you have to start with one that they offer.

I'm confused as to what you're saying about that change to the grub configuration. First of all, before the migration, the interface names on this VM were enX0 and enX1. I'm not sure why and that looked pretty alien to me frankly. Also, I was under the impression that adding that net.ifnames=1 was expressly to end up with the eth0/eth1 names, and not the other way around(?). This is currently what's in /etc/default/grub:

grep GRUB_CMDLINE_LINUX /etc/default/grub
GRUB_CMDLINE_LINUX="root=LABEL=root vsyscall=emulate"
GRUB_CMDLINE_LINUX="vsyscall=emulate"

Clearly after that reboot I'm ending up with eth0 and eth1, so I'm totally confused as to all that. If you're suggesting that change in order to have them match the existing interface names, then again, the existing ones weren't the enp0s0 names anyway.

I'm also confused as to exactly what you're suggesting about installing sysvinit-core. Are you saying to do that and reboot before anything else...even before changing the /etc/apt/sources.list?

Thanks again for the reply!
Tom

Yes! That worked and got it running! In fact I was able to resolve a similar issue with another executable I had that required libstdc++5:i386.

Thanks a million!
Tom

Thanks! Will give that a try.

Tom

Yea...agreed. I'll likely try to compile it from source using the newer libraries.

EDIT: One important thing to note about this post: No clue what I was thinking, but when I was referring the "ld" command I totally intended to be using "ldd". No wonder I got so confused on that.

Thanks!
Tom

Yea I noticed that. There were some heimdal related packages also required by the same executable. So far I haven't been able to figure out if any others might provide that library.

Tom

This is odd. So I was able to find that package for Debian Bookworm here:

https://packages.debian.org/bookworm/libmcrypt4

I installed the deb file libmcrypt4_2.5.8-7_amd64.deb. It's clearly installed:

dpkg-query -L libmcrypt4
/.
/usr
/usr/lib
/usr/lib/x86_64-linux-gnu
/usr/lib/x86_64-linux-gnu/libmcrypt.so.4.4.8
/usr/share
/usr/share/doc
/usr/share/doc/libmcrypt4
/usr/share/doc/libmcrypt4/changelog.Debian.gz
/usr/share/doc/libmcrypt4/changelog.gz
/usr/share/doc/libmcrypt4/copyright
/usr/lib/x86_64-linux-gnu/libmcrypt.so.4

Yet my executable still says it's missing:

error while loading shared libraries: libhcrypto.so.4: cannot open shared object file: No such file or directory

What do you suppose that's all about?

Thanks!
Tom

On a related topic, I've installed binutils to get the ld command, but ls doesn't work on anything I try. It always fails with an error like this:

ld: cannot use executable file '<filename>' as input to a link

I've read some things about that though I can't make any sense out of it. I've seen suggestions to use "-r" but that doesn't help.

Any clue on that would be appreciated as well.

Thanks!
Tom