[PATCH 00/10] *** SUBJECT HERE ***

[PATCH 00/10] *** SUBJECT HERE ***

[Harald Hoyer]

Some important things I changed today.

Harald Hoyer (10):
  add binutiles requirement to specfile (because of nm)
  output everything to /dev/kmesg and add dmesg for the emergency_shell
  Defer mount to the real mount loop
  remove 50plymouth-pre0.7 module
  add firmware packages to be required by the dracut-generic package
  add rm to be installed for initqueue
  fail iscsiroot, if iscsistart fails
  put back the nfs mount in the udev event
  initqueue now loops until /dev/root exists or root is mounted
  add "rdshell" command line argument

 dracut.8                                           |    3 +
 dracut.spec                                        |    8 ++-
 modules.d/40network/net-genrules.sh                |    8 +-
 modules.d/50plymouth-pre0.7/63-luks.rules          |   11 --
 modules.d/50plymouth-pre0.7/check                  |    3 -
 modules.d/50plymouth-pre0.7/cryptroot-ask.sh       |    4 -
 modules.d/50plymouth-pre0.7/install                |    8 --
 modules.d/50plymouth-pre0.7/plymouth-newroot.sh    |    4 -
 .../50plymouth-pre0.7/plymouth-populate-initrd     |   25 -----
 modules.d/50plymouth-pre0.7/plymouth-start.sh      |   31 ------
 modules.d/50plymouth/63-luks.rules                 |    2 +-
 modules.d/90crypt/70-luks.rules                    |    2 +-
 modules.d/90dmraid/61-dmraid.rules                 |   12 ++
 modules.d/90dmraid/dmraid.sh                       |    3 -
 modules.d/90dmraid/dmraid_scan                     |    8 ++
 modules.d/90dmraid/install                         |    3 +-
 modules.d/95debug/install                          |    2 +-
 modules.d/95iscsi/iscsiroot                        |    4 +-
 modules.d/95nbd/nbdroot                            |    7 +-
 modules.d/95nfs/nfsroot                            |   22 ++--
 modules.d/95rootfs-block/block-genrules.sh         |   12 ++-
 modules.d/95rootfs-block/install                   |    1 +
 modules.d/99base/init                              |  111 ++++++++++++--------
 modules.d/99base/initqueue                         |   11 ++-
 modules.d/99base/install                           |    3 +-
 test/TEST-10-RAID/test.sh                          |    4 +-
 test/TEST-20-NFS/test.sh                           |    2 +-
 test/TEST-30-ISCSI/test.sh                         |    4 +-
 test/TEST-40-NBD/test.sh                           |    4 +-
 29 files changed, 153 insertions(+), 169 deletions(-)
 delete mode 100644 modules.d/50plymouth-pre0.7/63-luks.rules
 delete mode 100755 modules.d/50plymouth-pre0.7/check
 delete mode 100755 modules.d/50plymouth-pre0.7/cryptroot-ask.sh
 delete mode 100755 modules.d/50plymouth-pre0.7/install
 delete mode 100755 modules.d/50plymouth-pre0.7/plymouth-newroot.sh
 delete mode 100755 modules.d/50plymouth-pre0.7/plymouth-populate-initrd
 delete mode 100755 modules.d/50plymouth-pre0.7/plymouth-start.sh
 create mode 100644 modules.d/90dmraid/61-dmraid.rules
 delete mode 100644 modules.d/90dmraid/dmraid.sh
 create mode 100755 modules.d/90dmraid/dmraid_scan

--
To unsubscribe from this list: send the line "unsubscribe initramfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 01/10] add binutiles requirement to specfile (because of nm)

[Harald Hoyer]

---
 dracut.spec |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

diff --git a/dracut.spec b/dracut.spec
index 7b7a224..3e33819 100644
--- a/dracut.spec
+++ b/dracut.spec
@@ -27,6 +27,7 @@ Requires: module-init-tools
 Requires: cpio
 Requires: coreutils
 Requires: findutils
+Requires: binutils
 Requires: grep
 Requires: mktemp
 Requires: mount
-- 
1.6.2.5

--
To unsubscribe from this list: send the line "unsubscribe initramfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 02/10] output everything to /dev/kmesg and add dmesg for the emergency_shell

[Harald Hoyer]

---
 modules.d/99base/init    |    9 ++++++++-
 modules.d/99base/install |    2 +-
 2 files changed, 9 insertions(+), 2 deletions(-)

diff --git a/modules.d/99base/init b/modules.d/99base/init
index 701706d..b49ba21 100755
--- a/modules.d/99base/init
+++ b/modules.d/99base/init
@@ -7,6 +7,8 @@
 
 emergency_shell()
 {
+    exec >/dev/console 2>&1 </dev/console
+    getarg rdinitdebug || dmesg
     source_all emergency
     echo ; echo
     echo "Bug in initramfs /init detected. Dropping to a shell. Good luck!"
@@ -64,7 +66,12 @@ if [ ! -c /dev/ptmx ]; then
     fi
 fi
 
-getarg rdinitdebug && set -x
+if getarg rdinitdebug; then
+    set -x
+else
+    exec >/dev/kmsg 2>&1
+fi
+
 mkdir /dev/shm
 mkdir /dev/pts
 mount -t devpts -o gid=5,mode=620 /dev/pts /dev/pts >/dev/null 2>&1
diff --git a/modules.d/99base/install b/modules.d/99base/install
index ff15200..4ecba9c 100755
--- a/modules.d/99base/install
+++ b/modules.d/99base/install
@@ -1,5 +1,5 @@
 #!/bin/bash
-dracut_install mount mknod mkdir modprobe pidof sleep chroot sed ls flock cp mv
+dracut_install mount mknod mkdir modprobe pidof sleep chroot sed ls flock cp mv dmesg 
 if [ ! -e "${initdir}/bin/sh" ]; then
     dracut_install bash
     (ln -s bash "${initdir}/bin/sh" || :)
-- 
1.6.2.5

--
To unsubscribe from this list: send the line "unsubscribe initramfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 03/10] Defer mount to the real mount loop

[Harald Hoyer]

Udev rules set a /dev/root symlink to the real root and add
a mount script to /mount/. This enables the proper use of pre-mount
scripts and prevents mount being killed by a udev timeout.
---
 modules.d/40network/net-genrules.sh        |    8 ++++----
 modules.d/95debug/install                  |    2 +-
 modules.d/95iscsi/iscsiroot                |    2 +-
 modules.d/95nbd/nbdroot                    |    7 ++++---
 modules.d/95nfs/nfsroot                    |   23 +++++++++++++----------
 modules.d/95rootfs-block/block-genrules.sh |    8 ++++----
 modules.d/99base/init                      |    9 ++++++++-
 7 files changed, 35 insertions(+), 24 deletions(-)

diff --git a/modules.d/40network/net-genrules.sh b/modules.d/40network/net-genrules.sh
index f843c53..dbd0474 100755
--- a/modules.d/40network/net-genrules.sh
+++ b/modules.d/40network/net-genrules.sh
@@ -19,20 +19,20 @@ fix_bootif() {
     BOOTIF=$(getarg 'BOOTIF=')
     if [ -n "$BOOTIF" ] ; then
 	BOOTIF=$(fix_bootif "$BOOTIF")
-	printf 'ACTION=="add", SUBSYSTEM=="net", ATTR{address}=="%s", RUN+="/sbin/initqueue /sbin/ifup $env{INTERFACE}"\n' "$BOOTIF"
+	printf 'ACTION=="add", SUBSYSTEM=="net", ATTR{address}=="%s", RUN+="/sbin/ifup $env{INTERFACE}"\n' "$BOOTIF"
 
     # If we have to handle multiple interfaces, handle only them.
     elif [ -n "$IFACES" ] ; then
 	for iface in $IFACES ; do
-	    printf 'ACTION=="add", SUBSYSTEM=="net", KERNEL=="%s", RUN+="/sbin/initqueue /sbin/ifup $env{INTERFACE}"\n' "$iface"
+	    printf 'ACTION=="add", SUBSYSTEM=="net", KERNEL=="%s", RUN+="/sbin/ifup $env{INTERFACE}"\n' "$iface"
 	done
 
     # Default: We don't know the interface to use, handle all
     else
-	printf 'ACTION=="add", SUBSYSTEM=="net", RUN+="/sbin/initqueue /sbin/ifup $env{INTERFACE}"\n'
+	printf 'ACTION=="add", SUBSYSTEM=="net", RUN+="/sbin/ifup $env{INTERFACE}"\n'
     fi
 
     # Udev event 'online' only gets fired from ifup/dhclient-script.
     # No special rules required
-    printf 'ACTION=="online", SUBSYSTEM=="net", RUN+="/sbin/initqueue /sbin/netroot $env{INTERFACE}"\n'
+    printf 'ACTION=="online", SUBSYSTEM=="net", RUN+="/sbin/netroot $env{INTERFACE}"\n'
 } > /etc/udev/rules.d/60-net.rules
diff --git a/modules.d/95debug/install b/modules.d/95debug/install
index 33a724d..6d4f089 100755
--- a/modules.d/95debug/install
+++ b/modules.d/95debug/install
@@ -1,3 +1,3 @@
 #!/bin/bash
-dracut_install ln ps grep more dmesg cat rm strace free showmount
+dracut_install ln ps grep more cat rm strace free showmount 
 dracut_install ping netstat rpcinfo
diff --git a/modules.d/95iscsi/iscsiroot b/modules.d/95iscsi/iscsiroot
index 8ee7cbc..316d8f4 100755
--- a/modules.d/95iscsi/iscsiroot
+++ b/modules.d/95iscsi/iscsiroot
@@ -122,7 +122,7 @@ echo "InitiatorName='$iscsi_initiator'" > /dev/.initiatorname.iscsi
 
 # FIXME $iscsi_protocol??
 
-echo "iscsi_lun=$iscsi_lun . /bin/mount-lun.sh " > /mount/iscsi.sh
+echo "iscsi_lun=$iscsi_lun . /bin/mount-lun.sh " > /mount/01-$$-iscsi.sh
 
 iscsistart -i $iscsi_initiator -t $iscsi_target_name	\
     -g $iscsi_target_group -a $iscsi_target_ip	\
diff --git a/modules.d/95nbd/nbdroot b/modules.d/95nbd/nbdroot
index ad2e906..ecac211 100755
--- a/modules.d/95nbd/nbdroot
+++ b/modules.d/95nbd/nbdroot
@@ -93,9 +93,10 @@ done
 # If we didn't get a root= on the command line, then we need to
 # add the udev rules for mounting the nbd0 device
 if [ ! -e /etc/udev/rules.d/99-mount.rules ]; then
-    printf 'KERNEL=="%s", RUN+="/sbin/initqueue /bin/mount -t %s -o %s %s %s"\n' \
-	nbd0 "$nbdfstype" "$fsopts" /dev/nbd0 "$NEWROOT" \
-		> /etc/udev/rules.d/99-mount.rules
+    printf 'KERNEL=="nbd0", SYMLINK+="root"\n'> /etc/udev/rules.d/99-mount.rules
+    printf '/bin/mount -t %s -o %s %s %s\n' \
+	   "$nbdfstype" "$fsopts" /dev/nbd0 "$NEWROOT" \
+	> /mount/01-$$-nbd.sh
 fi
 
 nbd-client $preopts "$nbdserver" "$nbdport" /dev/nbd0 $opts || exit 1
diff --git a/modules.d/95nfs/nfsroot b/modules.d/95nfs/nfsroot
index 0da8ee3..5f5b08f 100755
--- a/modules.d/95nfs/nfsroot
+++ b/modules.d/95nfs/nfsroot
@@ -142,14 +142,17 @@ if [ "$nfs" = "nfs4" ]; then
     [ -z "$(pidof rpc.idmapd)" ] && rpc.idmapd
 
     # XXX Should we loop here?
-    exec mount -t nfs4 -o$options${nfslock+,$nfslock} \
-	$server:$path $NEWROOT
-fi
-
-# NFSv{2,3} doesn't support using locks as it requires a helper to transfer
-# the rpcbind state to the new root
-[ "$nfslock" = "lock" ] && \
-    warn "Locks unsupported on NFSv{2,3}, using nolock" 1>&2
+    echo mount -t nfs4 -o$options${nfslock+,$nfslock} \
+	$server:$path $NEWROOT > /mount/01-$$-nfs4.sh
+    [ -e /dev/root ] || >/dev/root
+else
+    # NFSv{2,3} doesn't support using locks as it requires a helper to transfer
+    # the rpcbind state to the new root
+    [ "$nfslock" = "lock" ] && \
+        warn "Locks unsupported on NFSv{2,3}, using nolock" 1>&2
 
-# XXX Should we loop here?
-exec mount -t nfs -o$options${options:+,}nolock $server:$path $NEWROOT
+    # XXX Should we loop here?
+    echo mount -t nfs -o$options${options:+,}nolock $server:$path $NEWROOT \
+        > /mount/01-$$-nfs.sh
+    [ -e /dev/root ] || >/dev/root
+fi
diff --git a/modules.d/95rootfs-block/block-genrules.sh b/modules.d/95rootfs-block/block-genrules.sh
index 2706dbb..5d1a669 100644
--- a/modules.d/95rootfs-block/block-genrules.sh
+++ b/modules.d/95rootfs-block/block-genrules.sh
@@ -1,8 +1,8 @@
 if [ "${root%%:*}" = "block" ]; then
     (
-    printf 'KERNEL=="%s", RUN+="/sbin/initqueue /bin/mount -t %s -o %s %s %s"\n' \
-	${root#block:/dev/} "$fstype" "$rflags" "${root#block:}" "$NEWROOT"
-    printf 'SYMLINK=="%s", RUN+="/sbin/initqueue /bin/mount -t %s -o %s %s %s"\n' \
-	${root#block:/dev/} "$fstype" "$rflags" "${root#block:}" "$NEWROOT"
+    printf 'KERNEL=="%s", SYMLINK+="root"\n' \
+	${root#block:/dev/} 
+    printf 'SYMLINK=="%s", SYMLINK+="root"\n' \
+	${root#block:/dev/} 
     ) >> /etc/udev/rules.d/99-mount.rules
 fi
diff --git a/modules.d/99base/init b/modules.d/99base/init
index b49ba21..bb20220 100755
--- a/modules.d/99base/init
+++ b/modules.d/99base/init
@@ -22,8 +22,11 @@ do_initqueue()
         # bail out, if we have mounted the root filesystem
 	[ -d "$NEWROOT/proc" ] && break;
 
+        # check if root can be mounted
+        [ -e /dev/root ] && break;
+
 	if [ $UDEVVERSION -ge 143 ]; then
-	    udevadm settle --exit-if-exists=/initqueue/work
+	    udevadm settle --exit-if-exists=/initqueue/work --exit-if-exists=/dev/root
 	else
 	    udevadm settle --timeout=30
 	fi
@@ -33,8 +36,12 @@ do_initqueue()
 	for job in /initqueue/*.job; do
 	    . $job
 	    rm -f $job
+
             # bail out, if we have mounted the root filesystem
 	    [ -d "$NEWROOT/proc" ] && break;
+
+            # check if root can be mounted
+	    [ -e /dev/root ] && break;
 	done
     done
 }
-- 
1.6.2.5

--
To unsubscribe from this list: send the line "unsubscribe initramfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 04/10] remove 50plymouth-pre0.7 module

[Harald Hoyer]

---
 modules.d/50plymouth-pre0.7/63-luks.rules          |   11 -------
 modules.d/50plymouth-pre0.7/check                  |    3 --
 modules.d/50plymouth-pre0.7/cryptroot-ask.sh       |    4 --
 modules.d/50plymouth-pre0.7/install                |    8 -----
 modules.d/50plymouth-pre0.7/plymouth-newroot.sh    |    4 --
 .../50plymouth-pre0.7/plymouth-populate-initrd     |   25 ----------------
 modules.d/50plymouth-pre0.7/plymouth-start.sh      |   31 --------------------
 7 files changed, 0 insertions(+), 86 deletions(-)
 delete mode 100644 modules.d/50plymouth-pre0.7/63-luks.rules
 delete mode 100755 modules.d/50plymouth-pre0.7/check
 delete mode 100755 modules.d/50plymouth-pre0.7/cryptroot-ask.sh
 delete mode 100755 modules.d/50plymouth-pre0.7/install
 delete mode 100755 modules.d/50plymouth-pre0.7/plymouth-newroot.sh
 delete mode 100755 modules.d/50plymouth-pre0.7/plymouth-populate-initrd
 delete mode 100755 modules.d/50plymouth-pre0.7/plymouth-start.sh

diff --git a/modules.d/50plymouth-pre0.7/63-luks.rules b/modules.d/50plymouth-pre0.7/63-luks.rules
deleted file mode 100644
index 711e865..0000000
--- a/modules.d/50plymouth-pre0.7/63-luks.rules
+++ /dev/null
@@ -1,11 +0,0 @@
-# hacky rules to try to try unlocking dm-crypt devs
-#
-# Copyright 2008, Red Hat, Inc.
-# Jeremy Katz <katzj-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
-
-
-SUBSYSTEM!="block", GOTO="luks_end"
-ACTION!="add|change", GOTO="luks_end"
-ENV{ID_FS_TYPE}=="crypto_LUKS", RUN+="cryptroot-ask $env{DEVNAME} luks-$env{ID_FS_UUID}"
-
-LABEL="luks_end"
diff --git a/modules.d/50plymouth-pre0.7/check b/modules.d/50plymouth-pre0.7/check
deleted file mode 100755
index 018c0e9..0000000
--- a/modules.d/50plymouth-pre0.7/check
+++ /dev/null
@@ -1,3 +0,0 @@
-#!/bin/bash
-[[ $1 = -d ]] && echo crypt
-[[ -x /sbin/plymouthd && -x /bin/plymouth && ! /usr/sbin/plymouth-set-default-theme ]]
diff --git a/modules.d/50plymouth-pre0.7/cryptroot-ask.sh b/modules.d/50plymouth-pre0.7/cryptroot-ask.sh
deleted file mode 100755
index fadb502..0000000
--- a/modules.d/50plymouth-pre0.7/cryptroot-ask.sh
+++ /dev/null
@@ -1,4 +0,0 @@
-#!/bin/sh
-
-[ -b /dev/mapper/$2 ] || exec /bin/plymouth ask-for-password --command="/sbin/cryptsetup luksOpen -T1 $1 $2"
-
diff --git a/modules.d/50plymouth-pre0.7/install b/modules.d/50plymouth-pre0.7/install
deleted file mode 100755
index 4d3f7dc..0000000
--- a/modules.d/50plymouth-pre0.7/install
+++ /dev/null
@@ -1,8 +0,0 @@
-#!/bin/bash
-. "$moddir"/plymouth-populate-initrd
-inst_hook pre-udev 10 "$moddir"/plymouth-start.sh
-inst_hook pre-pivot 90 "$moddir"/plymouth-newroot.sh
-inst_rules "$moddir/63-luks.rules"
-inst "$moddir"/cryptroot-ask.sh /lib/udev/cryptroot-ask
-inst cryptsetup
-
diff --git a/modules.d/50plymouth-pre0.7/plymouth-newroot.sh b/modules.d/50plymouth-pre0.7/plymouth-newroot.sh
deleted file mode 100755
index c6f8c6d..0000000
--- a/modules.d/50plymouth-pre0.7/plymouth-newroot.sh
+++ /dev/null
@@ -1,4 +0,0 @@
-#!/bin/sh
-
-[ -x /bin/plymouth ] && /bin/plymouth --newroot=$NEWROOT
-
diff --git a/modules.d/50plymouth-pre0.7/plymouth-populate-initrd b/modules.d/50plymouth-pre0.7/plymouth-populate-initrd
deleted file mode 100755
index 1802b25..0000000
--- a/modules.d/50plymouth-pre0.7/plymouth-populate-initrd
+++ /dev/null
@@ -1,25 +0,0 @@
-#!/bin/bash
-if ldd /sbin/plymouthd |grep -q lib64; then
-    LIBDIR="/usr/lib64"
-else
-    LIBDIR="/usr/lib"
-fi
-PLYMOUTH_LOGO_FILE="/usr/share/pixmaps/system-logo-white.png"
-PLYMOUTH_PLUGIN_NAME=$(plymouth-set-default-plugin)
-
-inst /sbin/plymouthd /bin/plymouthd
-dracut_install /bin/plymouth "${LIBDIR}/plymouth/text.so" \
-    "${LIBDIR}/plymouth/details.so" "${PLYMOUTH_LOGO_FILE}" \
-    /etc/system-release "${LIBDIR}/plymouth/${PLYMOUTH_PLUGIN_NAME:=text}.so"
-mkdir -p "${initdir}/usr/share/plymouth"
-
-[[ -f ${LIBDIR}/plymouth/default.so ]] && inst ${LIBDIR}/plymouth/default.so
-
-if [[ -d /usr/share/plymouth/${PLYMOUTH_PLUGIN_NAME} ]]; then
-    for x in "/usr/share/plymouth/${PLYMOUTH_PLUGIN_NAME}"/* ; do
-        [[ -f "$x" ]] || break
-        inst $x
-    done
-fi
-
-# vim:ts=8:sw=4:sts=4:et
diff --git a/modules.d/50plymouth-pre0.7/plymouth-start.sh b/modules.d/50plymouth-pre0.7/plymouth-start.sh
deleted file mode 100755
index 310042d..0000000
--- a/modules.d/50plymouth-pre0.7/plymouth-start.sh
+++ /dev/null
@@ -1,31 +0,0 @@
-#!/bin/sh
-
-[ -c /dev/null ] || mknod /dev/null c 1 3
-mknod /dev/zero c 1 5
-mknod /dev/systty c 4 0
-mknod /dev/tty c 5 0
-[ -c /dev/console ] || mknod /dev/console c 5 1
-[ -c /dev/ptmx ] || mknod /dev/ptmx c 5 2
-mknod /dev/fb c 29 0
-mknod /dev/tty0 c 4 0
-mknod /dev/tty1 c 4 1
-mknod /dev/tty2 c 4 2
-mknod /dev/tty3 c 4 3
-mknod /dev/tty4 c 4 4
-mknod /dev/tty5 c 4 5
-mknod /dev/tty6 c 4 6
-mknod /dev/tty7 c 4 7
-mknod /dev/tty8 c 4 8
-mknod /dev/tty9 c 4 9
-mknod /dev/tty10 c 4 10
-mknod /dev/tty11 c 4 11
-mknod /dev/tty12 c 4 12
-mknod /dev/ttyS0 c 4 64
-mknod /dev/ttyS1 c 4 65
-mknod /dev/ttyS2 c 4 66
-mknod /dev/ttyS3 c 4 67
-/lib/udev/console_init tty0
-
-[ -x /bin/plymouthd ] && /bin/plymouthd 
-[ -x /bin/plymouth ] && /bin/plymouth --show-splash
-
-- 
1.6.2.5

--
To unsubscribe from this list: send the line "unsubscribe initramfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 05/10] add firmware packages to be required by the dracut-generic package

[Harald Hoyer]

---
 dracut.spec |    7 ++++++-
 1 files changed, 6 insertions(+), 1 deletions(-)

diff --git a/dracut.spec b/dracut.spec
index 3e33819..4fae92b 100644
--- a/dracut.spec
+++ b/dracut.spec
@@ -60,12 +60,17 @@ Requires: rpcbind nfs-utils
 Requires: iscsi-initiator-utils
 Requires: nbd
 Requires: net-tools iproute
+Requires: kernel-firmware
+Requires: ql2100-firmware
+Requires: ql2200-firmware
+Requires: ql23xx-firmware
+Requires: ql2400-firmware
+Requires: ql2500-firmware
 
 %description generic
 This package requires everything which is needed to build a generic
 all purpose initramfs.
 
-
 %prep
 %setup -q -n %{name}-%{version}%{?dashgittag}
 
-- 
1.6.2.5

--
To unsubscribe from this list: send the line "unsubscribe initramfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 06/10] add rm to be installed for initqueue

[Harald Hoyer]

---
 modules.d/99base/install |    3 ++-
 1 files changed, 2 insertions(+), 1 deletions(-)

diff --git a/modules.d/99base/install b/modules.d/99base/install
index 4ecba9c..3f5b29a 100755
--- a/modules.d/99base/install
+++ b/modules.d/99base/install
@@ -1,5 +1,6 @@
 #!/bin/bash
-dracut_install mount mknod mkdir modprobe pidof sleep chroot sed ls flock cp mv dmesg 
+dracut_install mount mknod mkdir modprobe pidof sleep chroot \
+    sed ls flock cp mv dmesg rm
 if [ ! -e "${initdir}/bin/sh" ]; then
     dracut_install bash
     (ln -s bash "${initdir}/bin/sh" || :)
-- 
1.6.2.5

--
To unsubscribe from this list: send the line "unsubscribe initramfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 07/10] fail iscsiroot, if iscsistart fails

[Harald Hoyer]

---
 modules.d/95iscsi/iscsiroot |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/modules.d/95iscsi/iscsiroot b/modules.d/95iscsi/iscsiroot
index 316d8f4..b7e3cf8 100755
--- a/modules.d/95iscsi/iscsiroot
+++ b/modules.d/95iscsi/iscsiroot
@@ -130,7 +130,7 @@ iscsistart -i $iscsi_initiator -t $iscsi_target_name	\
     ${iscsi_username+-u $iscsi_username} \
     ${iscsi_password+-w $iscsi_password} \
     ${iscsi_in_username+-U $iscsi_in_username} \
-    ${iscsi_in_password+-W $iscsi_in_password}
+    ${iscsi_in_password+-W $iscsi_in_password} || exit 1
 
 # now we have a root filesystem somewhere in /dev/sda*
 # let the normal block handler handle root=
-- 
1.6.2.5

--
To unsubscribe from this list: send the line "unsubscribe initramfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 08/10] put back the nfs mount in the udev event

[Harald Hoyer]

it solves the following case:

root=/dev/nfs nfsroot=server:/path

- the server could be reachable on any interface
- any interface can get an IP by dhcp
- only one IP is allowed to mount the root
---
 modules.d/95nfs/nfsroot |   11 +++++------
 1 files changed, 5 insertions(+), 6 deletions(-)

diff --git a/modules.d/95nfs/nfsroot b/modules.d/95nfs/nfsroot
index 5f5b08f..0add977 100755
--- a/modules.d/95nfs/nfsroot
+++ b/modules.d/95nfs/nfsroot
@@ -142,9 +142,9 @@ if [ "$nfs" = "nfs4" ]; then
     [ -z "$(pidof rpc.idmapd)" ] && rpc.idmapd
 
     # XXX Should we loop here?
-    echo mount -t nfs4 -o$options${nfslock+,$nfslock} \
-	$server:$path $NEWROOT > /mount/01-$$-nfs4.sh
-    [ -e /dev/root ] || >/dev/root
+    mount -t nfs4 -o$options${nfslock+,$nfslock} \
+	$server:$path $NEWROOT \
+	&& { [ -e /dev/root ] || >/dev/root ; }
 else
     # NFSv{2,3} doesn't support using locks as it requires a helper to transfer
     # the rpcbind state to the new root
@@ -152,7 +152,6 @@ else
         warn "Locks unsupported on NFSv{2,3}, using nolock" 1>&2
 
     # XXX Should we loop here?
-    echo mount -t nfs -o$options${options:+,}nolock $server:$path $NEWROOT \
-        > /mount/01-$$-nfs.sh
-    [ -e /dev/root ] || >/dev/root
+    mount -t nfs -o$options${options:+,}nolock $server:$path $NEWROOT \
+	&& { [ -e /dev/root ] || >/dev/root ; }
 fi
-- 
1.6.2.5

--
To unsubscribe from this list: send the line "unsubscribe initramfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 09/10] initqueue now loops until /dev/root exists or root is mounted

[Harald Hoyer]

init now has the following points to inject scripts:

/cmdline/*.sh
   scripts for command line parsing

/pre-udev/*.sh
   scripts to run before udev is started

/pre-trigger/*.sh
   scripts to run before the main udev trigger is pulled

/initqueue/*.sh
   runs in parallel to the udev trigger
   Udev events can add scripts here with /sbin/initqueue.
   If /sbin/initqueue is called with the "--onetime" option, the script
   will be removed after it was run.
   If /initqueue/work is created and udev >= 143 then this loop can
   process the jobs in parallel to the udevtrigger.
   If the udev queue is empty and no root device is found or no root
   filesystem was mounted, the user will be dropped to a shell after
   a timeout.
   Scripts can remove themselves from the initqueue by "rm $job".

/pre-mount/*.sh
   scripts to run before the root filesystem is mounted
   NFS is an exception, because it has no device node to be created
   and mounts in the udev events

/mount/*.sh
   scripts to mount the root filesystem
   NFS is an exception, because it has no device node to be created
   and mounts in the udev events
   If the udev queue is empty and no root device is found or no root
   filesystem was mounted, the user will be dropped to a shell after
   a timeout.

/pre-pivot/*.sh
   scripts to run before the real init is executed and the initramfs
   disappears
   All processes started before should be killed here.

The behaviour of the dmraid module demonstrates how to use the new
mechanism. If it detects a device which is part of a raidmember from a
udev rule, it installs a job to scan for dmraid devices, if the udev
queue is empty. After a scan, it removes itsself from the queue.
---
 modules.d/50plymouth/63-luks.rules         |    2 +-
 modules.d/90crypt/70-luks.rules            |    2 +-
 modules.d/90dmraid/61-dmraid.rules         |   12 ++++
 modules.d/90dmraid/dmraid.sh               |    3 -
 modules.d/90dmraid/dmraid_scan             |    8 +++
 modules.d/90dmraid/install                 |    3 +-
 modules.d/95rootfs-block/block-genrules.sh |    4 ++
 modules.d/95rootfs-block/install           |    1 +
 modules.d/99base/init                      |   82 +++++++++++++++-------------
 modules.d/99base/initqueue                 |   11 +++-
 test/TEST-10-RAID/test.sh                  |    2 +-
 test/TEST-30-ISCSI/test.sh                 |    2 +-
 test/TEST-40-NBD/test.sh                   |    2 +-
 13 files changed, 85 insertions(+), 49 deletions(-)
 create mode 100644 modules.d/90dmraid/61-dmraid.rules
 delete mode 100644 modules.d/90dmraid/dmraid.sh
 create mode 100755 modules.d/90dmraid/dmraid_scan

diff --git a/modules.d/50plymouth/63-luks.rules b/modules.d/50plymouth/63-luks.rules
index c3c6627..f55bf39 100644
--- a/modules.d/50plymouth/63-luks.rules
+++ b/modules.d/50plymouth/63-luks.rules
@@ -7,6 +7,6 @@
 SUBSYSTEM!="block", GOTO="luks_end"
 ACTION!="add|change", GOTO="luks_end"
 
-ENV{ID_FS_TYPE}=="crypto_LUKS", RUN+="/sbin/initqueue /sbin/cryptroot-ask $env{DEVNAME} luks-$env{ID_FS_UUID}"
+ENV{ID_FS_TYPE}=="crypto_LUKS", RUN+="/sbin/initqueue --onetime /sbin/cryptroot-ask $env{DEVNAME} luks-$env{ID_FS_UUID}"
 
 LABEL="luks_end"
diff --git a/modules.d/90crypt/70-luks.rules b/modules.d/90crypt/70-luks.rules
index c3c6627..f55bf39 100644
--- a/modules.d/90crypt/70-luks.rules
+++ b/modules.d/90crypt/70-luks.rules
@@ -7,6 +7,6 @@
 SUBSYSTEM!="block", GOTO="luks_end"
 ACTION!="add|change", GOTO="luks_end"
 
-ENV{ID_FS_TYPE}=="crypto_LUKS", RUN+="/sbin/initqueue /sbin/cryptroot-ask $env{DEVNAME} luks-$env{ID_FS_UUID}"
+ENV{ID_FS_TYPE}=="crypto_LUKS", RUN+="/sbin/initqueue --onetime /sbin/cryptroot-ask $env{DEVNAME} luks-$env{ID_FS_UUID}"
 
 LABEL="luks_end"
diff --git a/modules.d/90dmraid/61-dmraid.rules b/modules.d/90dmraid/61-dmraid.rules
new file mode 100644
index 0000000..a8034db
--- /dev/null
+++ b/modules.d/90dmraid/61-dmraid.rules
@@ -0,0 +1,12 @@
+# This file causes block devices with Linux RAID (mdadm) signatures to
+# automatically cause mdadm to be run.
+# See udev(8) for syntax
+
+SUBSYSTEM!="block", GOTO="dm_end"
+ACTION!="add|change", GOTO="dm_end"
+
+ENV{ID_FS_TYPE}=="linux_raid_member", GOTO="dm_end"
+
+ENV{ID_FS_TYPE}=="*_raid_member", RUN+="/sbin/initqueue /sbin/dmraid_scan"
+
+LABEL="dm_end"
diff --git a/modules.d/90dmraid/dmraid.sh b/modules.d/90dmraid/dmraid.sh
deleted file mode 100644
index 4e2d5bf..0000000
--- a/modules.d/90dmraid/dmraid.sh
+++ /dev/null
@@ -1,3 +0,0 @@
-dmraid -ay -Z
-udevadm settle --timeout=30 >/dev/null 2>&1 
-
diff --git a/modules.d/90dmraid/dmraid_scan b/modules.d/90dmraid/dmraid_scan
new file mode 100755
index 0000000..433e5d3
--- /dev/null
+++ b/modules.d/90dmraid/dmraid_scan
@@ -0,0 +1,8 @@
+#!/bin/sh
+
+if udevadm settle --timeout=0 >/dev/null 2>&1; then
+    # run dmraid if udev has settled
+    dmraid -ay -Z
+    [ -e "$job" ] && rm -f "$job"
+fi
+
diff --git a/modules.d/90dmraid/install b/modules.d/90dmraid/install
index 636f1d2..9314c00 100755
--- a/modules.d/90dmraid/install
+++ b/modules.d/90dmraid/install
@@ -1,4 +1,5 @@
 #!/bin/bash
 dracut_install dmraid 
-inst_hook pre-mount 10 "$moddir/dmraid.sh"
+inst "$moddir/dmraid_scan" /sbin/dmraid_scan
 inst_rules 64-md-raid.rules
+inst_rules "$moddir/61-dmraid.rules"
diff --git a/modules.d/95rootfs-block/block-genrules.sh b/modules.d/95rootfs-block/block-genrules.sh
index 5d1a669..10471ea 100644
--- a/modules.d/95rootfs-block/block-genrules.sh
+++ b/modules.d/95rootfs-block/block-genrules.sh
@@ -5,4 +5,8 @@ if [ "${root%%:*}" = "block" ]; then
     printf 'SYMLINK=="%s", SYMLINK+="root"\n' \
 	${root#block:/dev/} 
     ) >> /etc/udev/rules.d/99-mount.rules
+    (
+    printf '[ -e "%s" ] && { ln -s "%s" /dev/root; rm "$job"; }\n' \
+	"${root#block:}" "${root#block:}"
+    ) >> /initqueue/blocksymlink.sh
 fi
diff --git a/modules.d/95rootfs-block/install b/modules.d/95rootfs-block/install
index 618d357..d6c161a 100755
--- a/modules.d/95rootfs-block/install
+++ b/modules.d/95rootfs-block/install
@@ -1,4 +1,5 @@
 #!/bin/sh
+dracut_install ln
 inst_hook cmdline 95 "$moddir/parse-block.sh"
 inst_hook pre-udev 30 "$moddir/block-genrules.sh"
 inst_hook mount 99 "$moddir/mount-root.sh"
diff --git a/modules.d/99base/init b/modules.d/99base/init
index bb20220..f082765 100755
--- a/modules.d/99base/init
+++ b/modules.d/99base/init
@@ -16,36 +16,6 @@ emergency_shell()
     sh -i
 }
 
-do_initqueue()
-{
-    while :; do
-        # bail out, if we have mounted the root filesystem
-	[ -d "$NEWROOT/proc" ] && break;
-
-        # check if root can be mounted
-        [ -e /dev/root ] && break;
-
-	if [ $UDEVVERSION -ge 143 ]; then
-	    udevadm settle --exit-if-exists=/initqueue/work --exit-if-exists=/dev/root
-	else
-	    udevadm settle --timeout=30
-	fi
-	[ -f /initqueue/work ] || break
-	rm /initqueue/work
-	
-	for job in /initqueue/*.job; do
-	    . $job
-	    rm -f $job
-
-            # bail out, if we have mounted the root filesystem
-	    [ -d "$NEWROOT/proc" ] && break;
-
-            # check if root can be mounted
-	    [ -e /dev/root ] && break;
-	done
-    done
-}
-
 export PATH=/sbin:/bin:/usr/sbin:/usr/bin
 export TERM=linux
 NEWROOT="/sysroot"
@@ -116,7 +86,47 @@ source_all pre-trigger
 # then the rest
 udevadm trigger $udevtriggeropts  >/dev/null 2>&1
 
-do_initqueue
+i=0
+while :; do
+    # bail out, if we have mounted the root filesystem
+    [ -d "$NEWROOT/proc" ] && break;
+
+    # check if root can be mounted
+    [ -e /dev/root ] && break;
+
+    if [ $UDEVVERSION -ge 143 ]; then
+        udevadm settle --exit-if-exists=/initqueue/work --exit-if-exists=/dev/root 
+    else
+        udevadm settle --timeout=30
+    fi
+    unset queuetriggered
+    if [ -f /initqueue/work ]; then
+        rm /initqueue/work
+	queuetriggered="1"
+    fi
+    	
+    for job in /initqueue/*.sh; do
+	[ -e "$job" ] || break
+        job=$job . $job
+
+        # bail out, if we have mounted the root filesystem
+        [ -d "$NEWROOT/proc" ] && break;
+        # check if root can be mounted
+        [ -e /dev/root ] && break;
+    done
+
+    [ -n "$queuetriggered" ] && continue
+
+    if udevadm settle --timeout=0 >/dev/null 2>&1; then
+        # no more udev jobs
+        sleep 0.5
+        i=$(($i+1))
+        [ $i -gt 20 ] && getarg rdshell \
+            && { flock -s 9 ; emergency_shell; } 9>/.console_lock
+    fi
+done
+unset job
+unset queuetriggered
 
 # pre-mount happens before we try to mount the root filesystem,
 # and happens once.
@@ -124,23 +134,19 @@ getarg 'rdbreak=pre-mount' && emergency_shell
 source_all pre-mount
 getarg 'rdbreak=mount' && emergency_shell
 
-do_initqueue
-
 # mount scripts actually try to mount the root filesystem, and may
 # be sourced any number of times. As soon as one suceeds, no more are sourced.
 i=0
 while :; do
     [ -d "$NEWROOT/proc" ] && break;
-
     for f in /mount/*.sh; do
        [ -f "$f" ] && . "$f"
-       do_initqueue
-       [ "$ROOTFS_MOUNTED" ] && break;
+       [ -d "$NEWROOT/proc" ] && break;
     done
 
-    sleep 0.5
     i=$(($i+1))
-    { flock -s 9 ; [ $i -gt 20 ] && emergency_shell; } 9>/.console_lock
+    [ $i -gt 20 ] && getarg rdshell \
+        && { flock -s 9 ; emergency_shell; } 9>/.console_lock
 done
 
 # pre pivot scripts are sourced just before we switch over to the new root.
diff --git a/modules.d/99base/initqueue b/modules.d/99base/initqueue
index 0a760b7..a5c6ef6 100755
--- a/modules.d/99base/initqueue
+++ b/modules.d/99base/initqueue
@@ -1,5 +1,12 @@
 #!/bin/sh
 
-echo "$@" > /tmp/$$.job
-mv /tmp/$$.job /initqueue/
+if [ "$1" = "--onetime" ]; then
+    onetime="yes"
+    shift
+fi
+echo "$@" > /tmp/$$.sh
+if [ -n "$onetime" ]; then
+   echo '[ -e "$job" ] && rm "$job"' >> /tmp/$$.sh
+fi
+mv /tmp/$$.sh /initqueue/
 >> /initqueue/work
diff --git a/test/TEST-10-RAID/test.sh b/test/TEST-10-RAID/test.sh
index 23f9aff..700c377 100755
--- a/test/TEST-10-RAID/test.sh
+++ b/test/TEST-10-RAID/test.sh
@@ -38,7 +38,7 @@ test_setup() {
 	initdir=overlay
 	. $basedir/dracut-functions
 	dracut_install sfdisk mke2fs poweroff cp umount 
-	inst_simple ./create-root.sh /pre-mount/01create-root.sh
+	inst_simple ./create-root.sh /initqueue/01create-root.sh
     )
  
     # create an initramfs that will create the target root filesystem.
diff --git a/test/TEST-30-ISCSI/test.sh b/test/TEST-30-ISCSI/test.sh
index 6469950..d9da529 100755
--- a/test/TEST-30-ISCSI/test.sh
+++ b/test/TEST-30-ISCSI/test.sh
@@ -81,7 +81,7 @@ test_setup() {
 	initdir=overlay
 	. $basedir/dracut-functions
 	dracut_install sfdisk mke2fs poweroff cp umount 
-	inst_simple ./create-root.sh /pre-mount/01create-root.sh
+	inst_simple ./create-root.sh /initqueue/01create-root.sh
     )
  
     # create an initramfs that will create the target root filesystem.
diff --git a/test/TEST-40-NBD/test.sh b/test/TEST-40-NBD/test.sh
index 2f54511..07de7fe 100755
--- a/test/TEST-40-NBD/test.sh
+++ b/test/TEST-40-NBD/test.sh
@@ -184,7 +184,7 @@ make_encrypted_root() {
 	initdir=overlay
 	. $basedir/dracut-functions
 	dracut_install mke2fs poweroff cp umount
-	inst_simple ./create-root.sh /pre-mount/01create-root.sh
+	inst_simple ./create-root.sh /initqueue/01create-root.sh
     )
 
     # create an initramfs that will create the target root filesystem.
-- 
1.6.2.5

--
To unsubscribe from this list: send the line "unsubscribe initramfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

[PATCH 10/10] add "rdshell" command line argument

[Harald Hoyer]

Only drop to an interactive shell if "rdshell" is specified on the
command line. This prevents malicious users from gaining easy shell
access to the host system (grub might be secured with a password).
---
 dracut.8                   |    3 +++
 modules.d/99base/init      |   37 ++++++++++++++++++++++---------------
 test/TEST-10-RAID/test.sh  |    2 +-
 test/TEST-20-NFS/test.sh   |    2 +-
 test/TEST-30-ISCSI/test.sh |    2 +-
 test/TEST-40-NBD/test.sh   |    2 +-
 6 files changed, 29 insertions(+), 19 deletions(-)

diff --git a/dracut.8 b/dracut.8
index 4bf29a7..4d6ac29 100644
--- a/dracut.8
+++ b/dracut.8
@@ -167,6 +167,9 @@ do not load kernel module <drivername>
 
 .SH Debug
 .TP
+.B rdshell
+drop to a shell, if the imitramfs fails
+.TP
 .B rdinitdebug
 set -x for the dracut shell
 .TP
diff --git a/modules.d/99base/init b/modules.d/99base/init
index f082765..cc42527 100755
--- a/modules.d/99base/init
+++ b/modules.d/99base/init
@@ -8,19 +8,25 @@
 emergency_shell()
 {
     exec >/dev/console 2>&1 </dev/console
+    echo ; echo
+    echo $@
     getarg rdinitdebug || dmesg
     source_all emergency
-    echo ; echo
-    echo "Bug in initramfs /init detected. Dropping to a shell. Good luck!"
-    echo
-    sh -i
+    if getarg rdshell; then
+        echo ; echo
+        echo "Dropping to a shell. Good luck!"
+        echo
+        sh -i
+    else
+        while :; do sleep 1;done
+    fi
 }
 
 export PATH=/sbin:/bin:/usr/sbin:/usr/bin
 export TERM=linux
 NEWROOT="/sysroot"
 
-trap "emergency_shell" 0
+trap "emergency_shell Unkown signal caught!" 0
 
 . /lib/dracut-lib.sh
 
@@ -56,7 +62,7 @@ mount -t devpts -o gid=5,mode=620 /dev/pts /dev/pts >/dev/null 2>&1
 UDEVVERSION=$(udevadm --version)
 
 # run scriptlets to parse the command line
-getarg 'rdbreak=cmdline' && emergency_shell
+getarg 'rdbreak=cmdline' && emergency_shell "Break before cmdline"
 source_all cmdline
 
 [ -z "$root" ] && die "No or empty root= argument"
@@ -73,7 +79,7 @@ source_all cmdline
 } > /tmp/root.info
 
 # pre-udev scripts run before udev starts, and are run only once.
-getarg 'rdbreak=pre-udev' && emergency_shell
+getarg 'rdbreak=pre-udev' && emergency_shell "Break before pre-udev"
 source_all pre-udev
 
 # start up udev and trigger cold plugs
@@ -121,8 +127,8 @@ while :; do
         # no more udev jobs
         sleep 0.5
         i=$(($i+1))
-        [ $i -gt 20 ] && getarg rdshell \
-            && { flock -s 9 ; emergency_shell; } 9>/.console_lock
+        [ $i -gt 20 ] \
+            && { flock -s 9 ; emergency_shell "No root device found"; } 9>/.console_lock
     fi
 done
 unset job
@@ -130,10 +136,11 @@ unset queuetriggered
 
 # pre-mount happens before we try to mount the root filesystem,
 # and happens once.
-getarg 'rdbreak=pre-mount' && emergency_shell
+getarg 'rdbreak=pre-mount' && emergency_shell "Break pre-mount"
 source_all pre-mount
-getarg 'rdbreak=mount' && emergency_shell
 
+
+getarg 'rdbreak=mount' && emergency_shell "Break mount"
 # mount scripts actually try to mount the root filesystem, and may
 # be sourced any number of times. As soon as one suceeds, no more are sourced.
 i=0
@@ -145,8 +152,8 @@ while :; do
     done
 
     i=$(($i+1))
-    [ $i -gt 20 ] && getarg rdshell \
-        && { flock -s 9 ; emergency_shell; } 9>/.console_lock
+    [ $i -gt 20 ] \
+        && { flock -s 9 ; emergency_shell "Can't mount root filesystem"; } 9>/.console_lock
 done
 
 # pre pivot scripts are sourced just before we switch over to the new root.
@@ -159,8 +166,8 @@ for i in "$(getarg init=)" /sbin/init /etc/init /init /bin/sh; do
     [ -f "$NEWROOT$i" -a -x "$NEWROOT$i" ] && { INIT="$i"; break; }
 done
 [ "$INIT" ] || {
-    echo "Cannot find init! Please check to make sure you passed"
-    echo "a valid root filesystem!  Dropping to a shell."
+    echo "Cannot find init!"
+    echo "Please check to make sure you passed a valid root filesystem!"
     emergency_shell
 }
 
diff --git a/test/TEST-10-RAID/test.sh b/test/TEST-10-RAID/test.sh
index 700c377..fbad51a 100755
--- a/test/TEST-10-RAID/test.sh
+++ b/test/TEST-10-RAID/test.sh
@@ -9,7 +9,7 @@ KVERSION=${KVERSION-$(uname -r)}
 test_run() {
     $testdir/run-qemu -hda root.ext2 -m 256M -nographic \
 	-net none -kernel /boot/vmlinuz-$KVERSION \
-	-append "root=/dev/dracut/root rw quiet console=ttyS0,115200n81 $DEBUGFAIL" \
+	-append "root=/dev/dracut/root rw quiet console=ttyS0,115200n81 rdshell $DEBUGFAIL" \
 	-initrd initramfs.testing
     grep -m 1 -q dracut-root-block-success root.ext2 || return 1
 }
diff --git a/test/TEST-20-NFS/test.sh b/test/TEST-20-NFS/test.sh
index 4410751..1811eca 100755
--- a/test/TEST-20-NFS/test.sh
+++ b/test/TEST-20-NFS/test.sh
@@ -46,7 +46,7 @@ client_test() {
   	-net nic,macaddr=$mac,model=e1000 \
   	-net socket,mcast=230.0.0.1:1234 \
   	-kernel /boot/vmlinuz-$KVERSION \
-  	-append "$cmdline $DEBUGFAIL ro quiet console=ttyS0,115200n81" \
+  	-append "$cmdline $DEBUGFAIL rdshell ro quiet console=ttyS0,115200n81" \
   	-initrd initramfs.testing
 
     if [[ $? -ne 0 ]] || ! grep -m 1 -q nfs-OK client.img; then
diff --git a/test/TEST-30-ISCSI/test.sh b/test/TEST-30-ISCSI/test.sh
index d9da529..440cd16 100755
--- a/test/TEST-30-ISCSI/test.sh
+++ b/test/TEST-30-ISCSI/test.sh
@@ -37,7 +37,7 @@ run_client() {
   	-net nic,macaddr=52:54:00:12:34:00,model=e1000 \
   	-net socket,mcast=230.0.0.1:1235 \
   	-kernel /boot/vmlinuz-$KVERSION \
-	-append "root=dhcp rw quiet console=ttyS0,115200n81 $DEBUGFAIL" \
+	-append "root=dhcp rw quiet console=ttyS0,115200n81 rdshell $DEBUGFAIL" \
   	-initrd initramfs.testing
     grep -m 1 -q iscsi-OK client.img || return 1
 }
diff --git a/test/TEST-40-NBD/test.sh b/test/TEST-40-NBD/test.sh
index 07de7fe..c3d8bfc 100755
--- a/test/TEST-40-NBD/test.sh
+++ b/test/TEST-40-NBD/test.sh
@@ -50,7 +50,7 @@ client_test() {
 	-net nic,macaddr=$mac,model=e1000 \
 	-net socket,mcast=230.0.0.1:1236 \
 	-kernel /boot/vmlinuz-$KVERSION \
-	-append "$cmdline $DEBUGFAIL ro quiet console=ttyS0,115200n81" \
+	-append "$cmdline $DEBUGFAIL rdshell ro quiet console=ttyS0,115200n81" \
 	-initrd initramfs.testing
 
     if [[ $? -ne 0 ]] || ! grep -m 1 -q nbd-OK flag.img; then
-- 
1.6.2.5

--
To unsubscribe from this list: send the line "unsubscribe initramfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 02/10] output everything to /dev/kmesg and add dmesg for the emergency_shell

[David Dillow]

On Fri, 2009-07-03 at 18:45 +0200, Harald Hoyer wrote:

> diff --git a/modules.d/99base/init b/modules.d/99base/init
> index 701706d..b49ba21 100755
> --- a/modules.d/99base/init
> +++ b/modules.d/99base/init
> @@ -7,6 +7,8 @@
>  
>  emergency_shell()
>  {
> +    exec >/dev/console 2>&1 </dev/console
> +    getarg rdinitdebug || dmesg

Ugh, please no. If I want dmesg output in my emergency shell, I can type
'dmesg'. Doing it for me means I may push valuable information from
rdinitdebug out of my scrollback buffer before I've had a chance to
review it and write it down.

Also, this adds a dependency on dmesg to a stripped down install, which
is counter to (my) goal of a smaller initramfs over time.

> -getarg rdinitdebug && set -x
> +if getarg rdinitdebug; then
> +    set -x
> +else
> +    exec >/dev/kmsg 2>&1
> +fi

I'm really not happy about polluting the kernel message source with
dracut messages. I know it is convenient, but in my opinion, dmesg
should only show me kernel messages, not a mixture of kernel and
userspace.

At the very least, we need to add an indication that the message is from
dracut, not the kernel.


--
To unsubscribe from this list: send the line "unsubscribe initramfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 02/10] output everything to /dev/kmesg and add dmesg for the emergency_shell

[Harald Hoyer]

[-- Attachment #1: Type: text/plain, Size: 1286 bytes --]

Am 03.07.2009 20:11, schrieb David Dillow:
> On Fri, 2009-07-03 at 18:45 +0200, Harald Hoyer wrote:
>
>> diff --git a/modules.d/99base/init b/modules.d/99base/init
>> index 701706d..b49ba21 100755
>> --- a/modules.d/99base/init
>> +++ b/modules.d/99base/init
>> @@ -7,6 +7,8 @@
>>
>>   emergency_shell()
>>   {
>> +    exec>/dev/console 2>&1</dev/console
>> +    getarg rdinitdebug || dmesg
>
> Ugh, please no. If I want dmesg output in my emergency shell, I can type
> 'dmesg'. Doing it for me means I may push valuable information from
> rdinitdebug out of my scrollback buffer before I've had a chance to
> review it and write it down.

ok

>
> Also, this adds a dependency on dmesg to a stripped down install, which
> is counter to (my) goal of a smaller initramfs over time.
>
>> -getarg rdinitdebug&&  set -x
>> +if getarg rdinitdebug; then
>> +    set -x
>> +else
>> +    exec>/dev/kmsg 2>&1
>> +fi
>
> I'm really not happy about polluting the kernel message source with
> dracut messages. I know it is convenient, but in my opinion, dmesg
> should only show me kernel messages, not a mixture of kernel and
> userspace.
>
> At the very least, we need to add an indication that the message is from
> dracut, not the kernel.

hmm, ok




[-- Attachment #2: S/MIME Cryptographic Signature --]
[-- Type: application/pkcs7-signature, Size: 3636 bytes --]

Re: [PATCH 03/10] Defer mount to the real mount loop

[Seewer Philippe]

Harald Hoyer wrote:
> Udev rules set a /dev/root symlink to the real root and add
> a mount script to /mount/. This enables the proper use of pre-mount
> scripts and prevents mount being killed by a udev timeout.

This looks look good. Network initialization now done asynchronously and 
in parallel again. Thank you!

Some comments below.

> ---
>  modules.d/40network/net-genrules.sh        |    8 ++++----
>  modules.d/95debug/install                  |    2 +-
>  modules.d/95iscsi/iscsiroot                |    2 +-
>  modules.d/95nbd/nbdroot                    |    7 ++++---
>  modules.d/95nfs/nfsroot                    |   23 +++++++++++++----------
>  modules.d/95rootfs-block/block-genrules.sh |    8 ++++----
>  modules.d/99base/init                      |    9 ++++++++-
>  7 files changed, 35 insertions(+), 24 deletions(-)
> 
> diff --git a/modules.d/40network/net-genrules.sh b/modules.d/40network/net-genrules.sh
> index f843c53..dbd0474 100755
> --- a/modules.d/40network/net-genrules.sh
> +++ b/modules.d/40network/net-genrules.sh
> @@ -19,20 +19,20 @@ fix_bootif() {
>      BOOTIF=$(getarg 'BOOTIF=')
>      if [ -n "$BOOTIF" ] ; then
>  	BOOTIF=$(fix_bootif "$BOOTIF")
> -	printf 'ACTION=="add", SUBSYSTEM=="net", ATTR{address}=="%s", RUN+="/sbin/initqueue /sbin/ifup $env{INTERFACE}"\n' "$BOOTIF"
> +	printf 'ACTION=="add", SUBSYSTEM=="net", ATTR{address}=="%s", RUN+="/sbin/ifup $env{INTERFACE}"\n' "$BOOTIF"
>  
>      # If we have to handle multiple interfaces, handle only them.
>      elif [ -n "$IFACES" ] ; then
>  	for iface in $IFACES ; do
> -	    printf 'ACTION=="add", SUBSYSTEM=="net", KERNEL=="%s", RUN+="/sbin/initqueue /sbin/ifup $env{INTERFACE}"\n' "$iface"
> +	    printf 'ACTION=="add", SUBSYSTEM=="net", KERNEL=="%s", RUN+="/sbin/ifup $env{INTERFACE}"\n' "$iface"
>  	done
>  
>      # Default: We don't know the interface to use, handle all
>      else
> -	printf 'ACTION=="add", SUBSYSTEM=="net", RUN+="/sbin/initqueue /sbin/ifup $env{INTERFACE}"\n'
> +	printf 'ACTION=="add", SUBSYSTEM=="net", RUN+="/sbin/ifup $env{INTERFACE}"\n'
>      fi
>  
>      # Udev event 'online' only gets fired from ifup/dhclient-script.
>      # No special rules required
> -    printf 'ACTION=="online", SUBSYSTEM=="net", RUN+="/sbin/initqueue /sbin/netroot $env{INTERFACE}"\n'
> +    printf 'ACTION=="online", SUBSYSTEM=="net", RUN+="/sbin/netroot $env{INTERFACE}"\n'
>  } > /etc/udev/rules.d/60-net.rules
> diff --git a/modules.d/95debug/install b/modules.d/95debug/install
> index 33a724d..6d4f089 100755
> --- a/modules.d/95debug/install
> +++ b/modules.d/95debug/install
> @@ -1,3 +1,3 @@
>  #!/bin/bash
> -dracut_install ln ps grep more dmesg cat rm strace free showmount
> +dracut_install ln ps grep more cat rm strace free showmount 
>  dracut_install ping netstat rpcinfo
> diff --git a/modules.d/95iscsi/iscsiroot b/modules.d/95iscsi/iscsiroot
> index 8ee7cbc..316d8f4 100755
> --- a/modules.d/95iscsi/iscsiroot
> +++ b/modules.d/95iscsi/iscsiroot
> @@ -122,7 +122,7 @@ echo "InitiatorName='$iscsi_initiator'" > /dev/.initiatorname.iscsi
>  
>  # FIXME $iscsi_protocol??
>  
> -echo "iscsi_lun=$iscsi_lun . /bin/mount-lun.sh " > /mount/iscsi.sh
> +echo "iscsi_lun=$iscsi_lun . /bin/mount-lun.sh " > /mount/01-$$-iscsi.sh

I'd do that after iscsistart. That way the file is only written if 
iscsistart succeeded.

>  
>  iscsistart -i $iscsi_initiator -t $iscsi_target_name	\
>      -g $iscsi_target_group -a $iscsi_target_ip	\
> diff --git a/modules.d/95nbd/nbdroot b/modules.d/95nbd/nbdroot
> index ad2e906..ecac211 100755
> --- a/modules.d/95nbd/nbdroot
> +++ b/modules.d/95nbd/nbdroot
> @@ -93,9 +93,10 @@ done
>  # If we didn't get a root= on the command line, then we need to
>  # add the udev rules for mounting the nbd0 device
>  if [ ! -e /etc/udev/rules.d/99-mount.rules ]; then
> -    printf 'KERNEL=="%s", RUN+="/sbin/initqueue /bin/mount -t %s -o %s %s %s"\n' \
> -	nbd0 "$nbdfstype" "$fsopts" /dev/nbd0 "$NEWROOT" \
> -		> /etc/udev/rules.d/99-mount.rules
> +    printf 'KERNEL=="nbd0", SYMLINK+="root"\n'> /etc/udev/rules.d/99-mount.rules
> +    printf '/bin/mount -t %s -o %s %s %s\n' \
> +	   "$nbdfstype" "$fsopts" /dev/nbd0 "$NEWROOT" \
> +	> /mount/01-$$-nbd.sh
>  fi

Same here, I'd only write a mount-script if nbdclient succeeds.

>  
>  nbd-client $preopts "$nbdserver" "$nbdport" /dev/nbd0 $opts || exit 1
> diff --git a/modules.d/95nfs/nfsroot b/modules.d/95nfs/nfsroot
> index 0da8ee3..5f5b08f 100755
> --- a/modules.d/95nfs/nfsroot
> +++ b/modules.d/95nfs/nfsroot
> @@ -142,14 +142,17 @@ if [ "$nfs" = "nfs4" ]; then
>      [ -z "$(pidof rpc.idmapd)" ] && rpc.idmapd
>  
>      # XXX Should we loop here?
> -    exec mount -t nfs4 -o$options${nfslock+,$nfslock} \
> -	$server:$path $NEWROOT
> -fi
> -
> -# NFSv{2,3} doesn't support using locks as it requires a helper to transfer
> -# the rpcbind state to the new root
> -[ "$nfslock" = "lock" ] && \
> -    warn "Locks unsupported on NFSv{2,3}, using nolock" 1>&2
> +    echo mount -t nfs4 -o$options${nfslock+,$nfslock} \
> +	$server:$path $NEWROOT > /mount/01-$$-nfs4.sh
> +    [ -e /dev/root ] || >/dev/root
> +else
> +    # NFSv{2,3} doesn't support using locks as it requires a helper to transfer
> +    # the rpcbind state to the new root
> +    [ "$nfslock" = "lock" ] && \
> +        warn "Locks unsupported on NFSv{2,3}, using nolock" 1>&2
>  
> -# XXX Should we loop here?
> -exec mount -t nfs -o$options${options:+,}nolock $server:$path $NEWROOT
> +    # XXX Should we loop here?
> +    echo mount -t nfs -o$options${options:+,}nolock $server:$path $NEWROOT \
> +        > /mount/01-$$-nfs.sh
> +    [ -e /dev/root ] || >/dev/root
> +fi
> diff --git a/modules.d/95rootfs-block/block-genrules.sh b/modules.d/95rootfs-block/block-genrules.sh
> index 2706dbb..5d1a669 100644
> --- a/modules.d/95rootfs-block/block-genrules.sh
> +++ b/modules.d/95rootfs-block/block-genrules.sh
> @@ -1,8 +1,8 @@
>  if [ "${root%%:*}" = "block" ]; then
>      (
> -    printf 'KERNEL=="%s", RUN+="/sbin/initqueue /bin/mount -t %s -o %s %s %s"\n' \
> -	${root#block:/dev/} "$fstype" "$rflags" "${root#block:}" "$NEWROOT"
> -    printf 'SYMLINK=="%s", RUN+="/sbin/initqueue /bin/mount -t %s -o %s %s %s"\n' \
> -	${root#block:/dev/} "$fstype" "$rflags" "${root#block:}" "$NEWROOT"
> +    printf 'KERNEL=="%s", SYMLINK+="root"\n' \
> +	${root#block:/dev/} 
> +    printf 'SYMLINK=="%s", SYMLINK+="root"\n' \
> +	${root#block:/dev/} 
>      ) >> /etc/udev/rules.d/99-mount.rules
>  fi
> diff --git a/modules.d/99base/init b/modules.d/99base/init
> index b49ba21..bb20220 100755
> --- a/modules.d/99base/init
> +++ b/modules.d/99base/init
> @@ -22,8 +22,11 @@ do_initqueue()
>          # bail out, if we have mounted the root filesystem
>  	[ -d "$NEWROOT/proc" ] && break;
>  
> +        # check if root can be mounted
> +        [ -e /dev/root ] && break;
> +
>  	if [ $UDEVVERSION -ge 143 ]; then
> -	    udevadm settle --exit-if-exists=/initqueue/work
> +	    udevadm settle --exit-if-exists=/initqueue/work --exit-if-exists=/dev/root
>  	else
>  	    udevadm settle --timeout=30
>  	fi
> @@ -33,8 +36,12 @@ do_initqueue()
>  	for job in /initqueue/*.job; do
>  	    . $job
>  	    rm -f $job
> +
>              # bail out, if we have mounted the root filesystem
>  	    [ -d "$NEWROOT/proc" ] && break;
> +
> +            # check if root can be mounted
> +	    [ -e /dev/root ] && break;
>  	done
>      done
>  }
--
To unsubscribe from this list: send the line "unsubscribe initramfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 10/10] add "rdshell" command line argument

[Jeremy Katz]

On Friday, July 03 2009, Harald Hoyer said:
> Only drop to an interactive shell if "rdshell" is specified on the
> command line. This prevents malicious users from gaining easy shell
> access to the host system (grub might be secured with a password).

I don't have a strong opinion about doing this vs not, but how could
they end up getting easy shell access?  If grub is secured with a
password, they can't change kernel arguments.  If they can change kernel
arguments, they can just add rdshell rather than change the root=
specifier.  

Jeremy
--
To unsubscribe from this list: send the line "unsubscribe initramfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 08/10] put back the nfs mount in the udev event

[Seewer Philippe]

Harald Hoyer wrote:
> it solves the following case:
> 
> root=/dev/nfs nfsroot=server:/path
> 
> - the server could be reachable on any interface
> - any interface can get an IP by dhcp
> - only one IP is allowed to mount the root

Thanks.

For the record, some further explanations:

If netroot is in "guess-the-interface-mode", netroot calls the
root-handler for each interface that has a useable IP configuration. If
the root-handler fails, the interface is deconfigured so that the next
interface netroot tries does not suffer from side-effects.

This is accomplished by depending on mount-handlers having an exit-code
telling us whether mounting (would) succeed or not. For devicefile based 
network mounts, this isn't a problem since we need to call some sort of
client-utility that makes the device nodes available (or not if
contacting the server fails). For network-filesystems like NFS that do
not use devicefiles, the actual mount call is needed.


> ---
>  modules.d/95nfs/nfsroot |   11 +++++------
>  1 files changed, 5 insertions(+), 6 deletions(-)
> 
> diff --git a/modules.d/95nfs/nfsroot b/modules.d/95nfs/nfsroot
> index 5f5b08f..0add977 100755
> --- a/modules.d/95nfs/nfsroot
> +++ b/modules.d/95nfs/nfsroot
> @@ -142,9 +142,9 @@ if [ "$nfs" = "nfs4" ]; then
>      [ -z "$(pidof rpc.idmapd)" ] && rpc.idmapd
>  
>      # XXX Should we loop here?
> -    echo mount -t nfs4 -o$options${nfslock+,$nfslock} \
> -	$server:$path $NEWROOT > /mount/01-$$-nfs4.sh
> -    [ -e /dev/root ] || >/dev/root
> +    mount -t nfs4 -o$options${nfslock+,$nfslock} \
> +	$server:$path $NEWROOT \
> +	&& { [ -e /dev/root ] || >/dev/root ; }
>  else
>      # NFSv{2,3} doesn't support using locks as it requires a helper to transfer
>      # the rpcbind state to the new root
> @@ -152,7 +152,6 @@ else
>          warn "Locks unsupported on NFSv{2,3}, using nolock" 1>&2
>  
>      # XXX Should we loop here?
> -    echo mount -t nfs -o$options${options:+,}nolock $server:$path $NEWROOT \
> -        > /mount/01-$$-nfs.sh
> -    [ -e /dev/root ] || >/dev/root
> +    mount -t nfs -o$options${options:+,}nolock $server:$path $NEWROOT \
> +	&& { [ -e /dev/root ] || >/dev/root ; }
>  fi
--
To unsubscribe from this list: send the line "unsubscribe initramfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 09/10] initqueue now loops until /dev/root exists or root is mounted

[Seewer Philippe]

Harald Hoyer wrote:
> init now has the following points to inject scripts:
> 
> /cmdline/*.sh
>    scripts for command line parsing
> 
> /pre-udev/*.sh
>    scripts to run before udev is started
> 
> /pre-trigger/*.sh
>    scripts to run before the main udev trigger is pulled
> 
> /initqueue/*.sh
>    runs in parallel to the udev trigger
>    Udev events can add scripts here with /sbin/initqueue.
>    If /sbin/initqueue is called with the "--onetime" option, the script
>    will be removed after it was run.
>    If /initqueue/work is created and udev >= 143 then this loop can
>    process the jobs in parallel to the udevtrigger.
>    If the udev queue is empty and no root device is found or no root
>    filesystem was mounted, the user will be dropped to a shell after
>    a timeout.
>    Scripts can remove themselves from the initqueue by "rm $job".

I like this. It introduces a bit more complexity than I'd like, but 
allows us to solves a lot of timeing issues between asynchronous and 
parallel tasks in udev.

> /pre-mount/*.sh
>    scripts to run before the root filesystem is mounted
>    NFS is an exception, because it has no device node to be created
>    and mounts in the udev events

Suggestion: If this description goes into a README or something similar, 
I'd rewrite this as "...network filesystems like NFS that do not use 
devicefiles are an exception,..."

> /mount/*.sh
>    scripts to mount the root filesystem
>    NFS is an exception, because it has no device node to be created
>    and mounts in the udev events
Same here, although I'd just drop it since its already mentioned or 
change it to say "network filesystems like NFS that do not use 
devicefiles should mount directly inside their netroot root-handlers."

>    If the udev queue is empty and no root device is found or no root
>    filesystem was mounted, the user will be dropped to a shell after
>    a timeout.
> 
> /pre-pivot/*.sh
>    scripts to run before the real init is executed and the initramfs
>    disappears
>    All processes started before should be killed here.
> 
> The behaviour of the dmraid module demonstrates how to use the new
> mechanism. If it detects a device which is part of a raidmember from a
> udev rule, it installs a job to scan for dmraid devices, if the udev
> queue is empty. After a scan, it removes itsself from the queue.

[snip]
> diff --git a/modules.d/90dmraid/dmraid_scan b/modules.d/90dmraid/dmraid_scan
> new file mode 100755
> index 0000000..433e5d3
> --- /dev/null
> +++ b/modules.d/90dmraid/dmraid_scan
> @@ -0,0 +1,8 @@
> +#!/bin/sh
> +
> +if udevadm settle --timeout=0 >/dev/null 2>&1; then
> +    # run dmraid if udev has settled
> +    dmraid -ay -Z
> +    [ -e "$job" ] && rm -f "$job"
> +fi

Please do not use 'udevadm settle timeout=0'. Older version of udev, 
especially the one on debian lenny does not support a value of 0.

I suggest to increase that timeout to 1.

[snip]
> diff --git a/modules.d/99base/init b/modules.d/99base/init
> index bb20220..f082765 100755
> --- a/modules.d/99base/init
> +++ b/modules.d/99base/init
> @@ -16,36 +16,6 @@ emergency_shell()
>      sh -i
>  }
>  
> -do_initqueue()
> -{
> -    while :; do
> -        # bail out, if we have mounted the root filesystem
> -	[ -d "$NEWROOT/proc" ] && break;
> -
> -        # check if root can be mounted
> -        [ -e /dev/root ] && break;
> -
> -	if [ $UDEVVERSION -ge 143 ]; then
> -	    udevadm settle --exit-if-exists=/initqueue/work --exit-if-exists=/dev/root
> -	else
> -	    udevadm settle --timeout=30
> -	fi
> -	[ -f /initqueue/work ] || break
> -	rm /initqueue/work
> -	
> -	for job in /initqueue/*.job; do
> -	    . $job
> -	    rm -f $job
> -
> -            # bail out, if we have mounted the root filesystem
> -	    [ -d "$NEWROOT/proc" ] && break;
> -
> -            # check if root can be mounted
> -	    [ -e /dev/root ] && break;
> -	done
> -    done
> -}
> -
>  export PATH=/sbin:/bin:/usr/sbin:/usr/bin
>  export TERM=linux
>  NEWROOT="/sysroot"
> @@ -116,7 +86,47 @@ source_all pre-trigger
>  # then the rest
>  udevadm trigger $udevtriggeropts  >/dev/null 2>&1
>  
> -do_initqueue
> +i=0
> +while :; do
> +    # bail out, if we have mounted the root filesystem
> +    [ -d "$NEWROOT/proc" ] && break;
> +
> +    # check if root can be mounted
> +    [ -e /dev/root ] && break;
> +
> +    if [ $UDEVVERSION -ge 143 ]; then
> +        udevadm settle --exit-if-exists=/initqueue/work --exit-if-exists=/dev/root 
> +    else
> +        udevadm settle --timeout=30
> +    fi
> +    unset queuetriggered
> +    if [ -f /initqueue/work ]; then
> +        rm /initqueue/work
> +	queuetriggered="1"
> +    fi
> +    	
> +    for job in /initqueue/*.sh; do
> +	[ -e "$job" ] || break
> +        job=$job . $job
> +
> +        # bail out, if we have mounted the root filesystem
> +        [ -d "$NEWROOT/proc" ] && break;
> +        # check if root can be mounted
> +        [ -e /dev/root ] && break;
> +    done
> +
> +    [ -n "$queuetriggered" ] && continue
> +
> +    if udevadm settle --timeout=0 >/dev/null 2>&1; then

Same here. Please increase the timeout to 1.

> +        # no more udev jobs
> +        sleep 0.5
> +        i=$(($i+1))
> +        [ $i -gt 20 ] && getarg rdshell \
> +            && { flock -s 9 ; emergency_shell; } 9>/.console_lock

Question: Would it make sense to reset i to 0 after we've opened a 
shell? Or is it the intention that if a shell is opened and closed that 
we drop in again as soon as possible?

Regards,
Philippe
--
To unsubscribe from this list: send the line "unsubscribe initramfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 09/10] initqueue now loops until /dev/root exists or root is mounted

[Harald Hoyer]

On 07/06/2009 10:54 AM, Seewer Philippe wrote:
> Suggestion: If this description goes into a README or something similar,
> I'd rewrite this as "...network filesystems like NFS that do not use
> devicefiles are an exception,..."

ok, will do

>
> Please do not use 'udevadm settle timeout=0'. Older version of udev,
> especially the one on debian lenny does not support a value of 0.
>
> I suggest to increase that timeout to 1.
>

ok, will do

> Question: Would it make sense to reset i to 0 after we've opened a
> shell? Or is it the intention that if a shell is opened and closed that
> we drop in again as soon as possible?

Well, if the user had a shell, he had the chance to find the real root, so I 
don't think 20 more loops will solve the problem.
--
To unsubscribe from this list: send the line "unsubscribe initramfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 03/10] Defer mount to the real mount loop

[Harald Hoyer]

On 07/04/2009 02:29 PM, Seewer Philippe wrote:
>> -echo "iscsi_lun=$iscsi_lun . /bin/mount-lun.sh " > /mount/iscsi.sh
>> +echo "iscsi_lun=$iscsi_lun . /bin/mount-lun.sh " > /mount/01-$$-iscsi.sh
>
> I'd do that after iscsistart. That way the file is only written if
> iscsistart succeeded.

right

>
>>
>> iscsistart -i $iscsi_initiator -t $iscsi_target_name \
>> -g $iscsi_target_group -a $iscsi_target_ip \
>> diff --git a/modules.d/95nbd/nbdroot b/modules.d/95nbd/nbdroot
>> index ad2e906..ecac211 100755
>> --- a/modules.d/95nbd/nbdroot
>> +++ b/modules.d/95nbd/nbdroot
>> @@ -93,9 +93,10 @@ done
>> # If we didn't get a root= on the command line, then we need to
>> # add the udev rules for mounting the nbd0 device
>> if [ ! -e /etc/udev/rules.d/99-mount.rules ]; then
>> - printf 'KERNEL=="%s", RUN+="/sbin/initqueue /bin/mount -t %s -o %s
>> %s %s"\n' \
>> - nbd0 "$nbdfstype" "$fsopts" /dev/nbd0 "$NEWROOT" \
>> - > /etc/udev/rules.d/99-mount.rules
>> + printf 'KERNEL=="nbd0", SYMLINK+="root"\n'>
>> /etc/udev/rules.d/99-mount.rules
>> + printf '/bin/mount -t %s -o %s %s %s\n' \
>> + "$nbdfstype" "$fsopts" /dev/nbd0 "$NEWROOT" \
>> + > /mount/01-$$-nbd.sh
>> fi
>
> Same here, I'd only write a mount-script if nbdclient succeeds.

yep
--
To unsubscribe from this list: send the line "unsubscribe initramfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 10/10] add "rdshell" command line argument

[Harald Hoyer]

On 07/06/2009 03:53 AM, Jeremy Katz wrote:
> On Friday, July 03 2009, Harald Hoyer said:
>> Only drop to an interactive shell if "rdshell" is specified on the
>> command line. This prevents malicious users from gaining easy shell
>> access to the host system (grub might be secured with a password).
>
> I don't have a strong opinion about doing this vs not, but how could
> they end up getting easy shell access?  If grub is secured with a
> password, they can't change kernel arguments.  If they can change kernel
> arguments, they can just add rdshell rather than change the root=
> specifier.
>
> Jeremy

If root is on a network device, they can just unplug the network cable and end 
up with a shell, then they can replug the cable and do whatever they want in the 
shell.
--
To unsubscribe from this list: send the line "unsubscribe initramfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Re: [PATCH 10/10] add "rdshell" command line argument

[Harald Hoyer]

On 07/06/2009 03:53 AM, Jeremy Katz wrote:
> On Friday, July 03 2009, Harald Hoyer said:
>> Only drop to an interactive shell if "rdshell" is specified on the
>> command line. This prevents malicious users from gaining easy shell
>> access to the host system (grub might be secured with a password).
>
> I don't have a strong opinion about doing this vs not, but how could
> they end up getting easy shell access?  If grub is secured with a
> password, they can't change kernel arguments.  If they can change kernel
> arguments, they can just add rdshell rather than change the root=
> specifier.
>
> Jeremy

If root is on a network device, they can just unplug the network cable and end 
up with a shell, then they can replug the cable and do whatever they want in the 
shell.
--
To unsubscribe from this list: send the line "unsubscribe initramfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html