On 2024/11/17 10:57, Daniel Jakots wrote:
> On Fri, 15 Nov 2024 09:34:47 +0000, Stuart Henderson
> <s...@spacehopper.org> wrote:
>
> > I'd put the package name first for quicker reading, and mention
> > that it's just server:
> >
> > <h3 id="r20241120">2024/11/20 - [packages] minio server: Removal of
> > Gateway and filesystem modes</h3>
>
> sure, modified locally
>
> > > +<p>
> > > +The last MinIO version supporting the MinIO is provided under the
> > > package +<tt>minio-old</tt>.
> >
> > If you want to change the package stem you also need to deal with
> > making sure the packages conflict. Also it's a bit different to other
> > examples where we have multiple versions of a port in the tree.
> >
> > Attached diff is a simpler approach keeping the same pkgstem, giving
> > this:
>
> Yes, the is-branch is a good idea!
>
> > > 1. should minio and minio-old conflict? To migrate to the SNSD mode,
> > > only the old version is enough, you don't need to have both.
> > > Otherwise if we don't set a conflict marker I'll rename the
> > > minio.rc to minio-old.rc I guess
> >
> > Yes I think so.
>
> This is handled by the is-branch, isn't it?
That's for a different purpose
is-branch
Annotate the few rare ports where several branches are
present in the ports tree (such as autoconf), to help
pkg_info(1) produce stem%branch annotations when needed.
> Can't install minio-0.20241107 because of conflicts (minio-0.20221024p3)
Packages with the same stem (minio-...) conflict by default - for things
like autoconf where they _don't_ conflict we have an extra "@option
no-default-conflict" annotation.
So just using the same stem is enough for the conflict to happen
automatically.
> > If you want to make sure that users do actually change it from the
> > default, you could check that in the rc.d script,
> >
> > getcap -f /etc/login.conf.d/minio:/etc/login.conf -c setenv minio
>
> Indeed a
> rc_pre() {
> getcap -f /etc/login.conf.d/minio:/etc/login.conf -c setenv minio
> }
> works as intended, however I'm afraid users will struggle to find out
> why it doesn't start (we all know how much pkg/README are actually read
> haha).
How about this?
rc_pre() {
if ! getcap -f /etc/login.conf.d/minio:/etc/login.conf \
-c setenv minio | grep -q ^=.*MINIO_ROOT_PASSWORD; then
echo "MINIO_ROOT_PASSWORD not set; refer to pkg-readme" >&2
return 1
fi
}
"rcctl start minio" will just print "failed", but hopefully people know
about "rcctl -d start ..."
> > And by providing the file, you can make sure it has sane permissions.
>
> So how about this?
>
> /usr/ports/net/minio$ cat server/pkg/minio.login
> minio:\
> :openfiles-cur=4096:\
> :openfiles-max=8192:\
> :tc=daemon:
I'm happy with that .login file
> /usr/ports/net/minio$ cat server/pkg/README
> +-------------------------------------------------------------------------------
> | Running ${PKGSTEM} on OpenBSD
> +-------------------------------------------------------------------------------
>
> minio's root access
> ===================
>
> By default, minio server use minioadmin:minioadmin as the user:password for
> the
> root access. The only way to change the default user and/or password is
> through the MINIO_ROOT_USER and MINIO_ROOT_PASSWORD environment variables. On
> OpenBSD, this can be achieved through ${SYSCONFDIR}/login.conf.d/minio:
That should specifically use /etc not ${SYSCONFDIR}. It is hardcoded in
libc so in the unlikely event anyone manages to get things to work with a
modified SYSCONFDIR in the first place, the login.conf.d files must be
exactly in /etc/login.conf.d.
Otherwise lgtm.
> minio:\
> :openfiles-cur=4096:\
> :openfiles-max=8192:\
> :setenv=MINIO_ROOT_USER=root,MINIO_ROOT_PASSWORD=yourstrongpassword:\
> :tc=daemon:
> At this point, I believe a diff is too big to be useful so instead, I
> attached the whole directory (/usr/ports/net$ tar cvzf minio.tgz minio)
OK to import minio-old unlinked for now so we can cut the size of
this :)
And the rest is nearly there.
