[factory][15.3] Grub is missing
# cat /usr/local/bin/zypse #!/bin/sh zypper --no-refresh se -s $* | egrep -v 'debug|devel|srcp|openSUSE-20' | egrep 'x86|noarch'| sort # zypse grub | grep -v grub2 | grubby | package | 8.40-bp152.3.18 | x86_64 | OSS | live-grub-stick | package | 0.4.2-bp152.3.15 | noarch | OSS il | grub | package | 0.97-lp151.3.5 | x86_64 | (System Packages) I've noticed a new problem, apparently because of this tighter SLE affiliation: Grub is missing from repos. Isn't this supposed to be a minor release? Is 15.2's Grub going to continue working? Does it depend on some lib that will be upgraded and break it? Use case is I'm a multi- multi-booter, on average > 12 installations per PC for testing on real hardware. Grub is only ever installed on partitions. All legacy MBRs contain DOS/OS2/Windows-compatible boot code. Mixing Grub and Grub2 for this use case is unacceptably unwieldy, an invitation for additional complicated trouble. I use Grub2(-efi) only on systems supporting UEFI, which number far fewer than those that don't. I have ~30 64bit systems, but only 5 supporting UEFI. -- Evolution as taught in public schools, like religion, is based on faith, not on science. Team OS/2 ** Reg. Linux User #211409 ** a11y rocks! Felix Miata *** http://fm.no-ip.com/
On 2021/01/14 09:30, Felix Miata wrote:
I've noticed a new problem: Grub is missing from repos. I hesitate to ask, as I haven't heard of this being the case, but wasn't systemd supposed to handle booting? Sorta doubt it has replaced this yet, but maybe someone else knows (and knows if that's still on the roadmap)?
Also, is lilo also gone? At one point there was a special elilo for efi. Might check if lilo is also missing and if elilo still exists..? If so, that might be another clue that MBR booting isn't being tested as part of the 15.2 boot process. That'd be annoying. Another thing that could happen is that lilo required 32-bit devel tools some time back. Dunno if that's still the case, but if they were left off and legacy/MBR booting is the only thing that used them, I could see someone deciding that MBR-boot would go the route of 32-bit support. Have you tried recompiling the 15.2 source under 15.3? I had to compile my own boot loader at one point in the distant past. Not that such would be a great workaround, but it might give you some clues..?
On 15/01/2021 03.26, L A Walsh wrote:
On 2021/01/14 09:30, Felix Miata wrote:
I've noticed a new problem: Grub is missing from repos. I hesitate to ask, as I haven't heard of this being the case, but wasn't systemd supposed to handle booting? Sorta doubt it has replaced this yet, but maybe someone else knows (and knows if that's still on the roadmap)?
It is possible to boot without grub, but "directly"; however this is not supported on openSUSE, last time I heard.
Also, is lilo also gone?
I think so, gone or going, but some people still use it.
At one point there was a special elilo for efi. Might check if lilo is also missing and if elilo still exists..? If so, that might be another clue that MBR booting isn't being tested as part of the 15.2 boot process. That'd be annoying.
Another thing that could happen is that lilo required 32-bit devel tools some time back. Dunno if that's still the case, but if they were left off and legacy/MBR booting is the only thing that used them, I could see someone deciding that MBR-boot would go the route of 32-bit support.
32 bit libraries and stack are supported on Leap 15.x. What is not supported is a 32 bit kernel and thus the system. 32 bit EFI is not supported, if I'm not mistaken. -- Cheers / Saludos, Carlos E. R. (from 15.2 x86_64 at Telcontar)
On 15.01.21 03:26, L A Walsh wrote:
Another thing that could happen is that lilo required 32-bit devel tools some time back. Dunno if that's still the case, but if they were left off and legacy/MBR booting is the only thing that used them, I could see someone deciding that
MBR-boot works just fine with grub2. -- Stefan Seyfried "For a successful technology, reality must take precedence over public relations, for nature cannot be fooled." -- Richard Feynman
On 1/15/21 4:00 AM, Felix Miata wrote:
# cat /usr/local/bin/zypse #!/bin/sh zypper --no-refresh se -s $* | egrep -v 'debug|devel|srcp|openSUSE-20' | egrep 'x86|noarch'| sort # zypse grub | grep -v grub2 | grubby | package | 8.40-bp152.3.18 | x86_64 | OSS | live-grub-stick | package | 0.4.2-bp152.3.15 | noarch | OSS il | grub | package | 0.97-lp151.3.5 | x86_64 | (System Packages)
I've noticed a new problem, apparently because of this tighter SLE affiliation: Grub is missing from repos. Isn't this supposed to be a minor release? Is 15.2's Grub going to continue working? Does it depend on some lib that will be upgraded and break it?
Unlikely that this is the reason, grub(1) has never come from SLE, its highly likely grub(1) isn't even in SLE-15, this means it has always come from the community. Currently grub doesn't build for tumbleweed so there is a chance it also doesn't build for 15.3, it could have also just been dropped while we try not to drop things in point releases sometimes it happens if no one is willing to make it work. Especially when there are alternatives available. -- Simon Lees (Simotek) http://simotek.net Emergency Update Team keybase.io/simotek SUSE Linux Adelaide Australia, UTC+10:30 GPG Fingerprint: 5B87 DB9D 88DC F606 E489 CEC5 0922 C246 02F0 014B
Simon Lees composed on 2021-01-15 13:35 (UTC+1030):
grub(1) has never come from SLE
According to https://distrowatch.com/table.php?distribution=sle Grub was version 0.94 in SLE 9, 0.97 in SLE 10 & 11, and 2.02 in 12. -- Evolution as taught in public schools, like religion, is based on faith, not on science. Team OS/2 ** Reg. Linux User #211409 ** a11y rocks! Felix Miata *** http://fm.no-ip.com/
On 2021-01-14 T 23:21 -0500 Felix Miata wrote:
Simon Lees composed on 2021-01-15 13:35 (UTC+1030):
grub(1) has never come from SLE
According to https://distrowatch.com/table.php?distribution=sle Grub was version 0.94 in SLE 9, 0.97 in SLE 10 & 11,
Correct.
and 2.02 in 12.
The package is called "grub2" in both, SLE 12 and SLE 15. Thus I'd agree to Simon that any "grub" and not "grub2" package available to openSUSE in recent years was coming from the community / part of openSUSE only. So long - MgE -- Matthias G. Eckermann, Director Product Management Linux Platforms SUSE Software Solutions Germany GmbH - Maxfeldstr. 5 - 90409 Nürnberg (HRB 36809, AG Nürnberg) Geschäftsführer: Felix Imendörffer
Matthias G. Eckermann composed on 2021-01-15 02:59 (UTC-0500):
On 2021-01-14 T 23:21 -0500 Felix Miata wrote:
Simon Lees composed on 2021-01-15 13:35 (UTC+1030):
grub(1) has never come from SLE
According to https://distrowatch.com/table.php?distribution=sle Grub was version 0.94 in SLE 9, 0.97 in SLE 10 & 11,
Correct.
and 2.02 in 12.
The package is called "grub2" in both, SLE 12 and SLE 15. Thus I'd agree to Simon that any "grub" and not "grub2" package available to openSUSE in recent years was coming from the community / part of openSUSE only.
Simon referred to "grub(1)", implying to me a reference to what in the past decade plus many have referred to as Grub Legacy, and he used the word "never", not "recent years", so Matthias' last sentence, to me, is clear as mud. :( -- Evolution as taught in public schools, like religion, is based on faith, not on science. Team OS/2 ** Reg. Linux User #211409 ** a11y rocks! Felix Miata *** http://fm.no-ip.com/
On 2021-01-15 T 04:51 -0500 Felix Miata wrote:
On 2021-01-14 T 23:21 -0500 Felix Miata wrote:
Simon Lees composed on 2021-01-15 13:35 (UTC+1030):
grub(1) has never come from SLE
[...]
Simon referred to "grub(1)", implying to me a reference to what in the past decade plus many have referred to as Grub Legacy, and he used the word "never", not "recent years", so Matthias' last sentence, to me, is clear as mud. :(
Your point about "never" is fair: at a certain point in time "grub(1)" was definitely part of SUSE Linux Enterprise, I agree. Stay healthy! So long - MgE -- Matthias G. Eckermann, Director Product Management Linux Platforms SUSE Software Solutions Germany GmbH - Maxfeldstr. 5 - 90409 Nürnberg (HRB 36809, AG Nürnberg) Geschäftsführer: Felix Imendörffer
participants (6)
-
Carlos E.R. -
Felix Miata -
L A Walsh -
Matthias G. Eckermann -
Simon Lees -
Stefan Seyfried