alpine issueshttps://gitlab.alpinelinux.org/groups/alpine/-/issues2021-05-12T11:18:28Zhttps://gitlab.alpinelinux.org/alpine/aports/-/issues/12658Enable CONFIG_COMPAT on aarch64 virt kernel2021-05-12T11:18:28ZKevin DaudtEnable CONFIG_COMPAT on aarch64 virt kernel`CONFIG_COMPAT` is currently not enabled on the aarch64 -virt kernel. Enabling this allows us to use the -virt kernels to host armv7 / armhf docker containers.`CONFIG_COMPAT` is currently not enabled on the aarch64 -virt kernel. Enabling this allows us to use the -virt kernels to host armv7 / armhf docker containers.https://gitlab.alpinelinux.org/alpine/aports/-/issues/12657edge/community/thttpd-2.29-r1: switch to sthttpd-2.27.1 security fork2021-05-12T08:31:52ZTerraTechedge/community/thttpd-2.29-r1: switch to sthttpd-2.27.1 security fork```
sthttpd - a fork of thttpd, a tiny/turbo/throttling HTTP server
version 2.27.0 Oct 3, 2014
sthttpd is a fork of Jef Poskanzer's popular thttpd server. This fork aims to simply maintain the original codebase as bugs...```
sthttpd - a fork of thttpd, a tiny/turbo/throttling HTTP server
version 2.27.0 Oct 3, 2014
sthttpd is a fork of Jef Poskanzer's popular thttpd server. This fork aims to simply maintain the original codebase as bugs or security issues are found.
```
The original version has not been patched for: [CVE-2017-10671](https://www.cvedetails.com/cve/CVE-2017-10671/)
The original version still heavily relies on `strcpy()` which has been overhauled to use bounded `memmove()`: https://github.com/blueness/sthttpd/commit/aa3f36c0bf2aef1ffb17f5188ccf5e8afc13d3dc#diff-fd5839f30c92090795049608bd9f55d8d2dcec8f71e613ecfcd8544a7e4e7540
Also, there is a currently open CVE that has not been fixed yet: [CVE-2021-26843](https://github.com/blueness/sthttpd/issues/14)Francesco ColistaFrancesco Colistahttps://gitlab.alpinelinux.org/alpine/aports/-/issues/12656ruby-irb2021-05-09T16:10:48Zjaqqueruby-irbruby-irb requires ruby-io-console but that explicit dependency is missing.
Steps to reproduce:
1. apk add ruby-irb
2 irb
3. apk add ruby-io-console
4. irbruby-irb requires ruby-io-console but that explicit dependency is missing.
Steps to reproduce:
1. apk add ruby-irb
2 irb
3. apk add ruby-io-console
4. irbKevin DaudtKevin Daudthttps://gitlab.alpinelinux.org/alpine/aports/-/issues/12655ruby-webrick2021-05-09T16:10:47Zjaqqueruby-webrickruby-webrick depends on ruby-etc, but that explicit dependency is missing.
Steps to reproduce:
1. apk add ruby-webrick
2. ruby -e 'require "webrick"'
3. apk add ruby-etc
4. ruby -e 'require "webrick"'ruby-webrick depends on ruby-etc, but that explicit dependency is missing.
Steps to reproduce:
1. apk add ruby-webrick
2. ruby -e 'require "webrick"'
3. apk add ruby-etc
4. ruby -e 'require "webrick"'Kevin DaudtKevin Daudthttps://gitlab.alpinelinux.org/alpine/aports/-/issues/12654community/py3-gevent: 1.5.0 incompatible with py 3.92021-07-09T12:58:12Znepeatcommunity/py3-gevent: 1.5.0 incompatible with py 3.9Found this issue when attempting to use a script that uses gevent with the community version of gevent.
https://github.com/gevent/gevent/issues/1786
`NameError: name 'AF_INET' is not defined` happens with Python 3.9 and gevent 1.5.0, l...Found this issue when attempting to use a script that uses gevent with the community version of gevent.
https://github.com/gevent/gevent/issues/1786
`NameError: name 'AF_INET' is not defined` happens with Python 3.9 and gevent 1.5.0, latest version seems to be 21.1.2.https://gitlab.alpinelinux.org/alpine/aports/-/issues/12653gst-plugins-base missing graphene-dev makedepend, which disables glvideoflip ...2021-08-08T21:23:13ZDaniel Kolesagst-plugins-base missing graphene-dev makedepend, which disables glvideoflip (which breaks recent webkit)right now, the `gst-plugins-base` package is missing the `glvideoflip` element, which needs `graphene-dev` in `makedepends` to build; however, `graphene` is in the `community` repository, while `gst-plugins-base` is in `main`
current we...right now, the `gst-plugins-base` package is missing the `glvideoflip` element, which needs `graphene-dev` in `makedepends` to build; however, `graphene` is in the `community` repository, while `gst-plugins-base` is in `main`
current webkit trunk needs the element in order to be able to play video, which means this will become a problem eventually (we found this while investigating why videos render black in the upcoming 2.32.1 release, which had the patches backported - we'll probably end up reverting them for now)
it'd be best to fix this in `gst-plugins-base` but that would require moving `graphene` to `main`; what should be done here?https://gitlab.alpinelinux.org/alpine/aports/-/issues/12652sharing ZFS data sets via nfs (x86_64)2021-05-05T20:35:25ZJohn Bowerssharing ZFS data sets via nfs (x86_64)This issue mostly has to do with sharing ZFS data sets over nfs.
However, there is a first problem part that makes tracking down the main part more difficult.
If I make a directory to share on an ext4 partition after having created Z...This issue mostly has to do with sharing ZFS data sets over nfs.
However, there is a first problem part that makes tracking down the main part more difficult.
If I make a directory to share on an ext4 partition after having created ZFS data sets and unsuccessfully sharing them, the ext4 share doesn’t work – the first time. It is impossible to mount on a client using e.g.,
sudo mount -t nfs -vvvv host_ip_address:/home/me/host_test /home/me/client_test
This command will hang forever. However, if I try to do this on the host server, e.g.,
sudo mount -t nfs -vvvv localhost:/home/me/host_test /home/me/a_test
it will succeed. Thereafter it will always succeed on the client as well. This behavior survives reboots of the host. End of first problem part.
As documented by Max Mopp and me, it is currently impossible to share a ZFS data set over nfs. Setting sharenfs=on (or any other reasonable set of options) does not work. Neither does the combination of setting sharenfs=off and manually putting the share in /etc/exports.
In both cases,
showmount -e host_ip_address
shows correct the share(s) as does the same command on the host, i.e.,
showmount -e localhost.
However,
sudo mount -t nfs -vvvv host_ip_address:/home/me/host_zfs_share /home/me/client_zfs_share
hangs indefinitely on the client. The same command, on the host, i.e.,
sudo mount -t nfs -vvvv localhost:/home/me/host_zfs_share /home/me/test_zfs_share
also results in a hang.
File/directory permissions in all cases are me:me. Example contents of /etc/exports are:
# /etc/exports
#
# See exports(5) for a description.
# use exportfs -arv to reread
/home/me/host_zfs_share *(rw,sync,no_root_squash,no_subtree_check)
/home/me/host_test *(rw,sync,no_root_squash,no_subtree_check)
I haven’t found that changing the options makes any difference. I successfully use these export options and permissions with an Ubuntu ZoL server.
It seems like there is some missing plumbing between the latest version of ZFS on Linux and nfs in Alpine Linux. I hope that this information is sufficient to give a clue as to where to start looking.https://gitlab.alpinelinux.org/alpine/aports/-/issues/12651community/dante: Symbol not found: bindresvport2021-08-09T19:22:39ZMarten Ringwelskicommunity/dante: Symbol not found: bindresvportWhen running `socksify true` it gives me the following error:
```
May 5 21:35:32 (1620243332.952577) true[11886]: error: symbolfunction(): compile time configuration error? Failed to find "bindresvport" using RTLD_NEXT: Symbol not fou...When running `socksify true` it gives me the following error:
```
May 5 21:35:32 (1620243332.952577) true[11886]: error: symbolfunction(): compile time configuration error? Failed to find "bindresvport" using RTLD_NEXT: Symbol not found: bindresvport
```
This error occurs because musl does not implement `bindresvport`.https://gitlab.alpinelinux.org/alpine/aports/-/issues/12650Community/chromium: When will chromium 89.0.4 reach to main/stable branch?2021-05-05T14:14:05ZAshish DandgawhaleCommunity/chromium: When will chromium 89.0.4 reach to main/stable branch?I'm a member of a software development team of a software company, and we are planning to use chromium 89 in one of our projects based on the docker image of alpine Linux. But since chromium 89 is still in the community branch and the ma...I'm a member of a software development team of a software company, and we are planning to use chromium 89 in one of our projects based on the docker image of alpine Linux. But since chromium 89 is still in the community branch and the main branch has a very old 81 version. I was wondering if there is any chance chromium 89 will be moved to the main branch in near future and if not, is there any stability/security issue if we keep using it from the community branch? We were slightly reluctant to use the package from the community branch.
CC: @ncopahttps://gitlab.alpinelinux.org/alpine/aports/-/issues/12649community/qemu: Qemu fails to start with error "There is no option group 'spi...2022-04-15T22:59:15ZMattias Winthercommunity/qemu: Qemu fails to start with error "There is no option group 'spice'"This bug [https://bugs.launchpad.net/qemu/+bug/1910696](https://bugs.launchpad.net/qemu/+bug/1910696) shows up in Alpine in version 5.2.0. Comment 2 in the bug above summarizes the problem well. Debian has a workaround for the problem he...This bug [https://bugs.launchpad.net/qemu/+bug/1910696](https://bugs.launchpad.net/qemu/+bug/1910696) shows up in Alpine in version 5.2.0. Comment 2 in the bug above summarizes the problem well. Debian has a workaround for the problem here: [https://salsa.debian.org/qemu-team/qemu/-/blob/master/debian/patches/build-most-modules-statically-hack.diff](https://salsa.debian.org/qemu-team/qemu/-/blob/master/debian/patches/build-most-modules-statically-hack.diff)
It would be nice to be able to run LXD on Alpine later than 3.12, but as long as this problem exists, 5.0.0 (in 3.12) seems to be the latest version of qemu that works.https://gitlab.alpinelinux.org/alpine/aports/-/issues/12648collision [edge]: main/findutils and community/mlocate2022-09-01T06:11:40ZTerraTechcollision [edge]: main/findutils and community/mlocate```sh
$ sudo apk add mlocate
fetch http://sjc.edge.kernel.org/alpine/edge/main/x86_64/APKINDEX.tar.gz
fetch http://sjc.edge.kernel.org/alpine/edge/community/x86_64/APKINDEX.tar.gz
(1/1) Installing mlocate (0.26-r7)
Executing mlocate-0.26...```sh
$ sudo apk add mlocate
fetch http://sjc.edge.kernel.org/alpine/edge/main/x86_64/APKINDEX.tar.gz
fetch http://sjc.edge.kernel.org/alpine/edge/community/x86_64/APKINDEX.tar.gz
(1/1) Installing mlocate (0.26-r7)
Executing mlocate-0.26-r7.pre-install
ERROR: mlocate-0.26-r7: trying to overwrite usr/bin/locate owned by findutils-4.8.0-r0.
ERROR: mlocate-0.26-r7: trying to overwrite usr/bin/updatedb owned by findutils-4.8.0-r0.
Executing busybox-1.33.0-r7.trigger
[master ab045b6] committing changes in /etc made by "apk add mlocate"
4 files changed, 36 insertions(+)
create mode 100755 periodic/daily/updatedb
create mode 100644 updatedb.conf
1 error; 1893 MiB in 343 packages
```https://gitlab.alpinelinux.org/alpine/aports/-/issues/12647Run Alpine on Raspberry PI 4B with PXE boot from FreeNAS NFS & TFTP2021-05-03T15:37:11ZEdvard FilistovičRun Alpine on Raspberry PI 4B with PXE boot from FreeNAS NFS & TFTPHello,
I want to run Alpine on Raspberry PI 4B with PXE boot from FreeNAS NFS & TFTP
I have my DHCP Service on my core router. TrueNAS with tftp & nfs on it. I want to run alpine linux on this configuration, and build in a future k8s o...Hello,
I want to run Alpine on Raspberry PI 4B with PXE boot from FreeNAS NFS & TFTP
I have my DHCP Service on my core router. TrueNAS with tftp & nfs on it. I want to run alpine linux on this configuration, and build in a future k8s or k3s cluster. my PI are powered from PoE, and I have 12 devices in my PI rack.
I browse a lot of google, but still can't find a solution. I assume also to have persistence storage on NFS, and some shared storage for kubernetes on NFS.
Edhttps://gitlab.alpinelinux.org/alpine/aports/-/issues/12646edge/main/mc: various timeouts encountered during usage with ncurses2021-05-04T09:44:22ZTerraTechedge/main/mc: various timeouts encountered during usage with ncursesUsing `mc` on Alpine (edge), I've encountered timeouts while doing simple things like `alt-p` and `alt-n` for moving around in command history. There are also timeouts when leaving the built-in editor as well as the viewer, plus other v...Using `mc` on Alpine (edge), I've encountered timeouts while doing simple things like `alt-p` and `alt-n` for moving around in command history. There are also timeouts when leaving the built-in editor as well as the viewer, plus other various normal usages.
The following strace will show the `alt-p` timeout.
```
select(6, [0 5], NULL, NULL, NULL) = 1 (in [0]) <4.021007>
rt_sigaction(SIGINT, {sa_handler=SIG_IGN, sa_mask=[], sa_flags=SA_RESTORER, sa_restorer=0x7f1d6fdfac4b}, NULL, 8) = 0 <0.000006>
select(6, [0 5], NULL, NULL, NULL) = 1 (in [0]) <0.000005>
read(0, "\33", 1) = 1 <0.000005>
poll([{fd=0, events=POLLIN}], 1, 200) = 1 ([{fd=0, revents=POLLIN}]) <0.000004>
read(0, "p", 1) = 1 <0.000003>
# TIMEOUT: 1.001324 #######################################################################
select(1, [0], NULL, NULL, {tv_sec=1, tv_usec=0}) = 0 (Timeout) <1.001324>
###########################################################################################
select(4, [3], NULL, NULL, {tv_sec=0, tv_usec=0}) = 0 (Timeout) <0.000021>
select(1, [0], NULL, NULL, {tv_sec=0, tv_usec=0}) = 0 (Timeout) <0.000006>
select(1, [0], NULL, NULL, {tv_sec=0, tv_usec=0}) = 0 (Timeout) <0.000003>
select(4, [3], NULL, NULL, {tv_sec=0, tv_usec=0}) = 0 (Timeout) <0.000003>
rt_sigaction(SIGTSTP, {sa_handler=SIG_IGN, sa_mask=[], sa_flags=SA_RESTORER|SA_RESTART, sa_restorer=0x7f1d6fdfac4b}, {sa_handler=0x7f1d6fd7dece, sa_mask=[], sa_flags=SA_RESTORER|SA_RESTART, sa_restorer=0x7f1d6fdfac4b}, 8) = 0 <0.000003>
poll([{fd=0, events=POLLIN}], 1, 0) = 0 (Timeout) <0.000004>
poll([{fd=0, events=POLLIN}], 1, 0) = 0 (Timeout) <0.000003>
write(1, "\33[58;64Hhecksum "..., 41) = 41 <0.000014>
rt_sigaction(SIGTSTP, {sa_handler=0x7f1d6fd7dece, sa_mask=[], sa_flags=SA_RESTORER|SA_RESTART, sa_restorer=0x7f1d6fdfac4b}, NULL, 8) = 0 <0.000003>
rt_sigaction(SIGTSTP, {sa_handler=SIG_IGN, sa_mask=[], sa_flags=SA_RESTORER|SA_RESTART, sa_restorer=0x7f1d6fdfac4b}, {sa_handler=0x7f1d6fd7dece, sa_mask=[], sa_flags=SA_RESTORER|SA_RESTART, sa_restorer=0x7f1d6fdfac4b}, 8) = 0 <0.000003>
poll([{fd=0, events=POLLIN}], 1, 0) = 0 (Timeout) <0.000003>
poll([{fd=0, events=POLLIN}], 1, 0) = 0 (Timeout) <0.000003>
rt_sigaction(SIGTSTP, {sa_handler=0x7f1d6fd7dece, sa_mask=[], sa_flags=SA_RESTORER|SA_RESTART, sa_restorer=0x7f1d6fdfac4b}, NULL, 8) = 0 <0.000003>
select(4, [3], NULL, NULL, {tv_sec=0, tv_usec=0}) = 0 (Timeout) <0.000004>
rt_sigaction(SIGINT, {sa_handler=0x5608184a53bc, sa_mask=[], sa_flags=SA_RESTORER, sa_restorer=0x7f1d6fdfac4b}, NULL, 8) = 0 <0.000003>
select(6, [0 5], NULL, NULL, NULL <detached ...>
```
I've used mc for years on Gentoo and I've always linked it against the `slang` libraries instead of `ncurses`. I felt like this was a `ncurses` issue so I copied out main/mc into my testing/mc build arena. After building and installing the new `mc + slang` version, all of the annoying timeout delays were fixed.
Here is the `mc + slang` strace.
```
select(7, [3 6], NULL, NULL, NULL) = 1 (in [3]) <3.765956>
rt_sigaction(SIGINT, {sa_handler=SIG_IGN, sa_mask=[], sa_flags=SA_RESTORER, sa_restorer=0x7f0b5cd1ec4b}, NULL, 8) = 0 <0.000005>
select(7, [3 6], NULL, NULL, NULL) = 1 (in [3]) <0.000014>
select(4, [3], NULL, NULL, {tv_sec=10, tv_usec=0}) = 1 (in [3], left {tv_sec=9, tv_usec=999999}) <0.000005>
read(3, "\33", 1) = 1 <0.000005>
# TIMEOUT: 0.000069 #######################################################################
select(4, [3], NULL, NULL, {tv_sec=1, tv_usec=0}) = 1 (in [3], left {tv_sec=0, tv_usec=999999}) <0.000069>
###########################################################################################
select(4, [3], NULL, NULL, {tv_sec=0, tv_usec=0}) = 1 (in [3], left {tv_sec=0, tv_usec=0}) <0.000004>
select(4, [3], NULL, NULL, {tv_sec=10, tv_usec=0}) = 1 (in [3], left {tv_sec=9, tv_usec=999999}) <0.000004>
read(3, "p", 1) = 1 <0.000004>
select(5, [4], NULL, NULL, {tv_sec=0, tv_usec=0}) = 0 (Timeout) <0.000004>
select(4, [3], NULL, NULL, {tv_sec=0, tv_usec=0}) = 0 (Timeout) <0.000003>
select(4, [3], NULL, NULL, {tv_sec=0, tv_usec=0}) = 0 (Timeout) <0.000003>
select(5, [4], NULL, NULL, {tv_sec=0, tv_usec=0}) = 0 (Timeout) <0.000003>
write(1, "\33[58;56Habuild checksum\33[87C\33[58"..., 36) = 36 <0.000008>
select(5, [4], NULL, NULL, {tv_sec=0, tv_usec=0}) = 0 (Timeout) <0.000003>
rt_sigaction(SIGINT, {sa_handler=0x55acf26aff43, sa_mask=[], sa_flags=SA_RESTORER, sa_restorer=0x7f0b5cd1ec4b}, NULL, 8) = 0 <0.000003>
select(7, [3 6], NULL, NULL, NULL <detached ...>
```
<DELAY>
ncurses: 1.001324
slang : 0.000069
So far, `mc + slang` is incredibly fast and responsive.
As an aside, the same behavior is also in Alpine Linux v3.13.
Below is the patch I used to switch `mc` from using `ncurses` ==> `slang`.
```patch
--- APKBUILD 2021-04-28 14:40:27.006666699 -0700
+++ /home/fqdevel/devel/alpine/fq/testing/mc/APKBUILD 2021-05-03 04:04:20.126666592 -0700
@@ -2,13 +2,14 @@
# Maintainer: Natanael Copa <ncopa@alpinelinux.org>
pkgname=mc
pkgver=4.8.26
-pkgrel=0
+pkgrel=1
pkgdesc="Filemanager/shell that emulates Norton Commander"
url="https://www.midnight-commander.org"
arch="all"
license="GPL-3.0-or-later"
subpackages="$pkgname-doc $pkgname-lang"
-makedepends="e2fsprogs-dev glib-dev libssh2-dev ncurses-dev perl"
+depends="slang"
+makedepends="e2fsprogs-dev glib-dev libssh2-dev slang-dev perl"
source="https://www.midnight-commander.org/downloads/mc-$pkgver.tar.xz
fix-cchar_t.patch
alpine_syntax.patch"
@@ -28,7 +29,7 @@
--enable-vfs-sftp \
--with-internal-edit \
--with-mmap \
- --with-screen=ncurses \
+ --with-screen=slang \
--with-subshell \
--without-gpm-mouse \
--without-included-gettext \
```
If it is approved to move from `ncurses` to `slang`, would you like for me to create a merge request?
Thank you for your consideration.https://gitlab.alpinelinux.org/alpine/infra/aports-qa-bot/-/issues/16Automatically mark status:mr-build-broken when pipeline fails2021-05-08T07:40:41ZLeoAutomatically mark status:mr-build-broken when pipeline failsWhen a pipeline (or job, whichever is higher level) fails we should add this label and remove when a pipeline suceed if it is presentWhen a pipeline (or job, whichever is higher level) fails we should add this label and remove when a pipeline suceed if it is presenthttps://gitlab.alpinelinux.org/alpine/security/secdb/-/issues/6Missing CVE for a "fixed version" of heimdal package in release 3.62022-01-29T20:05:34ZtomerMissing CVE for a "fixed version" of heimdal package in release 3.6Hey @ncopa @Leo!
Seems like there is an issue with the secdb data for alpine 3.6 for the package heimdal https://secdb.alpinelinux.org/v3.6/main.json for version 7.4.0-r0, potentially after the changes to use the golang script instead of...Hey @ncopa @Leo!
Seems like there is an issue with the secdb data for alpine 3.6 for the package heimdal https://secdb.alpinelinux.org/v3.6/main.json for version 7.4.0-r0, potentially after the changes to use the golang script instead of the lua one .
The issue seem to be coming from the aports data, you can see on the master branch in aports there is a CVE assigned to the fixed version `7.4.0-r0` https://gitlab.alpinelinux.org/alpine/aports/-/blob/master/main/heimdal/APKBUILD while for branch 3.6-stable, the CVE is missing and the fixed version does exist https://gitlab.alpinelinux.org/alpine/aports/-/blob/3.6-stable/main/heimdal/APKBUILD
Should the CVE be added for the 3.6 release as well, or the fixed version should be removed?https://gitlab.alpinelinux.org/alpine/abuild/-/issues/10030abuild-keygen set -e change breaks $USER fallback2021-05-14T09:59:42ZMartijn Braamabuild-keygen set -e change breaks $USER fallbackIn commit 754270e4607d31f7ffed502411e616251cb49460 a set -e line is added to abort early on errors, but it has the side effect of making it impossible to reach the $USER fallback at https://gitlab.alpinelinux.org/alpine/abuild/-/blob/mas...In commit 754270e4607d31f7ffed502411e616251cb49460 a set -e line is added to abort early on errors, but it has the side effect of making it impossible to reach the $USER fallback at https://gitlab.alpinelinux.org/alpine/abuild/-/blob/master/abuild-keygen.in#L33 because it would fail on the git command
This fallback was used by postmarketOS ofcourse :Dhttps://gitlab.alpinelinux.org/alpine/infra/aports-qa-bot/-/issues/15Feature: use separate token for admin tasks.2023-08-29T15:36:05ZKevin DaudtFeature: use separate token for admin tasks.To improve security, I would like to limit the access that the bot has.
Right now, it uses algitbot, which has full admin access. Some admin level access is required to be able to read user e-mails, so my proposal is:
* Use a read-only...To improve security, I would like to limit the access that the bot has.
Right now, it uses algitbot, which has full admin access. Some admin level access is required to be able to read user e-mails, so my proposal is:
* Use a read-only admin token for automaintainer (and other tasks that require it)
* Use a project access token for interacting with the project
The project access token is associated with a bot user. I've already created one, and renamed it to @aports-qa-bot (though we could rename it to something different as well).
Not sure if this will cause issues though (ie, when interacting with forks).https://gitlab.alpinelinux.org/alpine/alpine-conf/-/issues/10474[Installation-Wiki FIXME-2 (data disk mode)] allow selecting new data disk f...2021-05-20T10:32:08ZSPO[Installation-Wiki FIXME-2 (data disk mode)] allow selecting new data disk for configs[ Alpine's "Installation" page covers the Alpinelinux disk modes. (https://wiki.alpinelinux.org/wiki/Installation)
However, it still has to refer to two manual `setup-alpine` "FIXME" workarounds, to cover the basic installation options....[ Alpine's "Installation" page covers the Alpinelinux disk modes. (https://wiki.alpinelinux.org/wiki/Installation)
However, it still has to refer to two manual `setup-alpine` "FIXME" workarounds, to cover the basic installation options. ]
=> In `setup-alpine`, after chhosing the "data disk mode", please allow selecting the just configured data disk, to also store configs.
----
From the wiki:
[FIXME-2: Setup-alpine can not yet configure to store lbu configs to the "data disk" after selecting to use one.
It's still necessary to
* first select to save configs to "none" in setup-alpine (the new data partition is not listed), and
* to manually edit /etc/lbu/lbu.conf to set e.g. LBU_MEDIA=sdXY,
* execute a corresponding echo "/dev/sdXY /media/sdXY vfat rw 0 0" >> /etc/fstab afterwards, and
* save the config with lbu commit to have the partition (here, dubbed as sdXY) mounted when booting.
]https://gitlab.alpinelinux.org/alpine/alpine-conf/-/issues/10473[Installation-Wiki FIXME-1 (diskless mode)] config/package cache on *internal...2021-05-20T10:42:44ZSPO[Installation-Wiki FIXME-1 (diskless mode)] config/package cache on *internal* disk[ Alpine's "Installation" page covers the Alpinelinux disk modes. (https://wiki.alpinelinux.org/wiki/Installation) However, it still has to refer to two manual `setup-alpine` "FIXME" workarounds, to cover the basic installation options. ...[ Alpine's "Installation" page covers the Alpinelinux disk modes. (https://wiki.alpinelinux.org/wiki/Installation) However, it still has to refer to two manual `setup-alpine` "FIXME" workarounds, to cover the basic installation options. ]
=> In `setup-alpine`, please also show internal partitions in the selection for local configs and package cache in the "diskless" operation mode.
---
From the wiki:
[FIXME-1: Storing local configs and the package cache on an internal disk still requires some manual steps to have the partition listed, i.e. making a /etc/fstab entry, mountpoint, and mount *before* running setup-alpine. And requires to manually commit this configuration to disk afterwards.]
The manual workaround steps are maintained at:
https://wiki.alpinelinux.org/wiki/Alpine_local_backup#Saving_and_loading_ISO_image_customizationshttps://gitlab.alpinelinux.org/alpine/infra/aports-qa-bot/-/issues/14Feature: ping mentor team for MRs from first-time contributors2021-05-29T13:59:44ZKevin DaudtFeature: ping mentor team for MRs from first-time contributorsNot sure if it's easy to extract, but gitlab shows when an MR is from first-time contributors. It would be nice if it would automatically ping `@teams/mentors` in that case.Not sure if it's easy to extract, but gitlab shows when an MR is from first-time contributors. It would be nice if it would automatically ping `@teams/mentors` in that case.