Sorry to come back to this, but I'm now trying to netboot Ubuntu
22.04.2, having found the above mentioned work-around for 20.04.
Something has changed: I can no longer netboot from NFSv4 by replacing
/usr/lib/klibc/bin/nfsmount with an NFSv4-friendly hack.
I am now getting the same error messages
Status changed to 'Confirmed' because the bug affects multiple users.
** Changed in: klibc (Ubuntu)
Status: New => Confirmed
--
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to klibc in Ubuntu.
https://bugs.launchpad.net/b
The upstream bug is in Debian 11. The current version of RaspiOS, which
seems to support booting from NFSv4, is a Debian 11 derivative. That
means that RaspiOS has fixed it for Debian 11. Is it possible to copy
the RaspiOS solution?
--
You received this bug notification because you are a member o
Debian 11 is also broken if an intird is being used but Raspberry Pi OS
is *not* using an initrd hence not using nfsmount.
--
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to klibc in Ubuntu.
https://bugs.launchpad.net/bugs/195471
See https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=409272#14 for a
work-around which replaces nfsmount in the initrd with /sbin/mount.nfs.
But that probably pulls in dependencies and blows up the size of the
initrd.
--
You received this bug notification because you are a member of Ubuntu
Touch
Thanks, I’ll look at that.
I note that this issue is now a confirmed bug. What does that mean in
practical terms? Will it be dealt with this week, this month, this
year…?
I need to be able to boot from NFSv4. If RaspiOS can do that it might
make sense for me to move after 15 years from Ubuntu to
It simply means that the ticket has been acknowledged as a bug/issue,
nothing more. As you might have noticed it's been reported in Debian in
2007, so it doesn't seem to be that important. I know that doesn't help
you but the fact that there's a work-around makes it even less
important. I'll take a
Thanks Juerg, you’ve been very helpful.
Presumably there is no quick fix for this because of the klibc/glibc
thing, so it’s an architectural rather than a programming issue. And it
needs to be fixed in Debian rather than in Ubuntu.
I’d offer to help but I arrived here after drilling down from a
c
Please do test the work-around. If indeed NFSv4 solves your overlayfs
problem then that would be another reason for fixing nfsmount.
We can carry Ubuntu specific fixes if for some reason Debian doesn't
want them but the preferred way is obviously to fix it in Debian so that
it tickles down natural
Have you considered informing klibc upstream? (Granted, save for hpa
that’s basically the Debian maintainers, but they don’t necessarily get
Launchpad bugreports either.)
--
You received this bug notification because you are a member of Ubuntu
Touch seeded packages, which is subscribed to klibc i
Upstream is aware since 2007 [1] :-) But apparently lacking NFSv4
support is not painful enough to get some traction. I'm currently
digging through the code.
[1] https://lists.zytor.com/archives/klibc/2007-February/002062.html
--
You received this bug notification because you are a member of Ubu
Done it! I've managed to boot my RPi4 with Ubuntu Jammy off an NFSv4
share using the workaround from https://bugs.debian.org/cgi-
bin/bugreport.cgi?bug=409272#14 .
I created the file in the attachment and compiled it with klcc. If
anyone else wants to use it they'll have to edit the file, put thei
Whilst trying to understand the problem I looked at the source of
nfsmount in klibc. Nfsmount has a switch statement in which it checks
the value of nfsvers and only proceeds if it is 2 or 3. Given that my
version of nfsmount, compiled with klcc, can indeed mount using NFSv4 I
suspect that in libc'
I will test the work-around, but I won’t get round to it before the end
of the week.
> On 9 Aug 2022, at 13:55, Juerg Haefliger <1954...@bugs.launchpad.net> wrote:
>
> Please do test the work-around. If indeed NFSv4 solves your overlayfs
> problem then that would be another reason for fixing nfs
Good point: since this is the first bug I have ever actively pursued,
this is one of possibly many avenues I should follow.
> On 9 Aug 2022, at 19:30, Thorsten Glaser <1954...@bugs.launchpad.net> wrote:
>
> Have you considered informing klibc upstream? (Granted, save for hpa
> that’s basically t
15 matches
Mail list logo
