[PATCH 0/5] Netroot updates and multinic test-suite

[PATCH 0/5] Netroot updates and multinic test-suite

[Philippe Seewer]

This is a first start at a multinic testsuite, some
improvements to netroot and a bugfix.

Patches can be pulled from 
git://github.com/seewer/dracut-seewer.git
as well

Philippe Seewer (5):
  Bugfix: Fix BOOTIF argument translation
  Give the kernel some time to up interfaces
  Be a bit more optimistic when setting mtus
  Add DAD do dhclient-script
  Add multinic test suite

 modules.d/40network/dhclient-script |   19 +++-
 modules.d/40network/ifup            |    3 +-
 modules.d/40network/net-genrules.sh |    2 +-
 test/TEST-50-MULTINIC/Makefile      |   10 ++
 test/TEST-50-MULTINIC/client-init   |    9 ++
 test/TEST-50-MULTINIC/dhcpd.conf    |   36 ++++++
 test/TEST-50-MULTINIC/exports       |    1 +
 test/TEST-50-MULTINIC/hard-off.sh   |    2 +
 test/TEST-50-MULTINIC/hosts         |    5 +
 test/TEST-50-MULTINIC/server-init   |   30 +++++
 test/TEST-50-MULTINIC/test.log      |   48 ++++++++
 test/TEST-50-MULTINIC/test.sh       |  225 +++++++++++++++++++++++++++++++++++
 12 files changed, 384 insertions(+), 6 deletions(-)
 create mode 100644 test/TEST-50-MULTINIC/Makefile
 create mode 100755 test/TEST-50-MULTINIC/client-init
 create mode 100644 test/TEST-50-MULTINIC/dhcpd.conf
 create mode 100644 test/TEST-50-MULTINIC/exports
 create mode 100755 test/TEST-50-MULTINIC/hard-off.sh
 create mode 100644 test/TEST-50-MULTINIC/hosts
 create mode 100755 test/TEST-50-MULTINIC/server-init
 create mode 100644 test/TEST-50-MULTINIC/test.log
 create mode 100755 test/TEST-50-MULTINIC/test.sh

--
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 1/5] Bugfix: Fix BOOTIF argument translation

[Philippe Seewer]

fix_bootif() in net-genrules stripped the first part off the
passed string, resulting in invalid mac addresses. fixed
---
 modules.d/40network/net-genrules.sh |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/modules.d/40network/net-genrules.sh b/modules.d/40network/net-genrules.sh
index 3d50435..dbd0474 100755
--- a/modules.d/40network/net-genrules.sh
+++ b/modules.d/40network/net-genrules.sh
@@ -2,7 +2,7 @@
 
 # pxelinux provides macaddr '-' separated, but we need ':'
 fix_bootif() {
-    local macaddr=${1##??-}
+    local macaddr=${1}
     local IFS='-'
     macaddr=$(for i in ${macaddr} ; do echo -n $i:; done)
     macaddr=${macaddr%:}
-- 
1.6.0.4

--
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 2/5] Give the kernel some time to up interfaces

[Philippe Seewer]

It does not really matter if the kernel needs some time or if
the ethernet handshake takes a while. It's a good idea to wait
a second after upping an interface to let things settle.
---
 modules.d/40network/dhclient-script |    2 ++
 modules.d/40network/ifup            |    3 ++-
 2 files changed, 4 insertions(+), 1 deletions(-)

diff --git a/modules.d/40network/dhclient-script b/modules.d/40network/dhclient-script
index 5b36a40..9e7eefd 100755
--- a/modules.d/40network/dhclient-script
+++ b/modules.d/40network/dhclient-script
@@ -52,6 +52,8 @@ netif=$interface
 case $reason in
     PREINIT)
 	ip link set $netif up
+	# Give the kernel some time to up the interface
+	sleep 1
 	;;
     BOUND)
 	setup_interface 
diff --git a/modules.d/40network/ifup b/modules.d/40network/ifup
index 89017bb..174edca 100755
--- a/modules.d/40network/ifup
+++ b/modules.d/40network/ifup
@@ -32,8 +32,9 @@ do_dhcp() {
 
 # Handle static ip configuration
 do_static() {
-{
+    {
 	echo ip link set $netif up 
+	echo sleep 1
 	echo ip addr flush dev $netif
 	echo ip addr add $ip/$mask dev $netif
     } > /tmp/net.$netif.up
-- 
1.6.0.4

--
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 3/5] Be a bit more optimistic when setting mtus

[Philippe Seewer]

The usual approach to setting mtus is to set the interface down,
set the mtu then set the interface back up again. Modern hardware
and/or drivers may support setting this on the fly, so we try
this and fall back to the old behaviour it it doesn't work.

In addition this patch only allows mtus greater than 576, this is
taken from debian/ubuntu dhclient-script.
---
 modules.d/40network/dhclient-script |   16 ++++++++++++----
 1 files changed, 12 insertions(+), 4 deletions(-)

diff --git a/modules.d/40network/dhclient-script b/modules.d/40network/dhclient-script
index 9e7eefd..6df630d 100755
--- a/modules.d/40network/dhclient-script
+++ b/modules.d/40network/dhclient-script
@@ -13,10 +13,18 @@ setup_interface() {
 
     [ -f /tmp/net.$netif.override ] && . /tmp/net.$netif.override
 
-    if [ -n "$mtu" ] ; then
-	echo ip link set $netif down
-	echo ip link set $netif mtu $mtu
-	echo ip link set $netif up
+    # Taken from debian dhclient-script:
+    # The 576 MTU is only used for X.25 and dialup connections
+    # where the admin wants low latency.  Such a low MTU can cause
+    # problems with UDP traffic, among other things.  As such,
+    # disallow MTUs from 576 and below by default, so that broken
+    # MTUs are ignored, but higher stuff is allowed (1492, 1500, etc).
+    if [ -n "$mtu" ] && [ $mtu -gt 576 ] ; then
+	echo "if ! ip link set $netif mtu $mtu ; then"
+	echo "ip link set $netif down"
+	echo "ip link set $netif mtu $mtu"
+	echo "ip link set $netif up"
+	echo "fi"
     fi > /tmp/net.$netif.up
 
     echo ip addr add $ip${mask:+/$mask} ${bcast:+broadcast $bcast} dev $netif >> /tmp/net.$netif.up
-- 
1.6.0.4

--
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 4/5] Add DAD do dhclient-script

[Philippe Seewer]

The manpage for dhclient-script says:

Before actually configuring the address, dhclient-script should
somehow ARP for it and exit with a nonzero status if it receives a
reply.

By using arping in dracut this is very easy, since arping has a
specific option to do just that.
---
 modules.d/40network/dhclient-script |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

diff --git a/modules.d/40network/dhclient-script b/modules.d/40network/dhclient-script
index 6df630d..3fb8da6 100755
--- a/modules.d/40network/dhclient-script
+++ b/modules.d/40network/dhclient-script
@@ -64,6 +64,7 @@ case $reason in
 	sleep 1
 	;;
     BOUND)
+	arping -q -D -c 2 -I $netif $new_ip_address || exit 1
 	setup_interface 
 	set | while read line; do
 	    [ "${line#new_}" = "$line" ] && continue
-- 
1.6.0.4

--
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 5/5] Add multinic test suite

[Philippe Seewer]

This is a start at a test suite for the multinic part of dracut
---
 test/TEST-50-MULTINIC/Makefile    |   10 ++
 test/TEST-50-MULTINIC/client-init |    9 ++
 test/TEST-50-MULTINIC/dhcpd.conf  |   36 ++++++
 test/TEST-50-MULTINIC/exports     |    1 +
 test/TEST-50-MULTINIC/hard-off.sh |    2 +
 test/TEST-50-MULTINIC/hosts       |    5 +
 test/TEST-50-MULTINIC/server-init |   30 +++++
 test/TEST-50-MULTINIC/test.log    |   48 ++++++++
 test/TEST-50-MULTINIC/test.sh     |  225 +++++++++++++++++++++++++++++++++++++
 9 files changed, 366 insertions(+), 0 deletions(-)
 create mode 100644 test/TEST-50-MULTINIC/Makefile
 create mode 100755 test/TEST-50-MULTINIC/client-init
 create mode 100644 test/TEST-50-MULTINIC/dhcpd.conf
 create mode 100644 test/TEST-50-MULTINIC/exports
 create mode 100755 test/TEST-50-MULTINIC/hard-off.sh
 create mode 100644 test/TEST-50-MULTINIC/hosts
 create mode 100755 test/TEST-50-MULTINIC/server-init
 create mode 100644 test/TEST-50-MULTINIC/test.log
 create mode 100755 test/TEST-50-MULTINIC/test.sh

diff --git a/test/TEST-50-MULTINIC/Makefile b/test/TEST-50-MULTINIC/Makefile
new file mode 100644
index 0000000..bc0ddb6
--- /dev/null
+++ b/test/TEST-50-MULTINIC/Makefile
@@ -0,0 +1,10 @@
+all:
+	@make -s --no-print-directory -C ../.. all
+	@basedir=../.. testdir=../ ./test.sh --all
+setup:
+	@make --no-print-directory -C ../.. all
+	@basedir=../.. testdir=../ ./test.sh --setup
+clean:
+	@basedir=../.. testdir=../ ./test.sh --clean
+run:
+	@basedir=../.. testdir=../ ./test.sh --run
diff --git a/test/TEST-50-MULTINIC/client-init b/test/TEST-50-MULTINIC/client-init
new file mode 100755
index 0000000..66290c6
--- /dev/null
+++ b/test/TEST-50-MULTINIC/client-init
@@ -0,0 +1,9 @@
+#!/bin/sh
+exec >/dev/console 2>&1
+export TERM=linux
+export PS1='initramfs-test:\w\$ '
+stty sane
+echo "made it to the rootfs! Powering down."
+echo OK $(ls /dev/.initramfs/*.ifcfg) > /dev/sda
+#sh -i
+poweroff -f
diff --git a/test/TEST-50-MULTINIC/dhcpd.conf b/test/TEST-50-MULTINIC/dhcpd.conf
new file mode 100644
index 0000000..be8dee7
--- /dev/null
+++ b/test/TEST-50-MULTINIC/dhcpd.conf
@@ -0,0 +1,36 @@
+ddns-update-style none;
+
+use-host-decl-names true;
+
+subnet 192.168.50.0 netmask 255.255.255.0 {
+	option subnet-mask      255.255.255.0;
+	option routers          192.168.50.1;
+	next-server             192.168.50.1;
+	server-identifier       192.168.50.1;
+	option domain-name-servers 192.168.50.1;
+	option domain-search "example.com";
+	option domain-name "other.com";
+
+	group {
+		host client-if1 {
+			hardware ethernet 52:54:00:12:34:00;
+			fixed-address 192.168.50.100;
+		}
+	}
+
+	group {
+		host client-if2 {
+			hardware ethernet 52:54:00:12:34:01;
+			fixed-address 192.168.50.101;
+		}
+	}
+
+	group {
+		option root-path "nfs:192.168.50.1:/nfs/client";
+
+		host client-if3 {
+			hardware ethernet 52:54:00:12:34:02;
+			fixed-address 192.168.50.102;
+		}
+	}
+}
diff --git a/test/TEST-50-MULTINIC/exports b/test/TEST-50-MULTINIC/exports
new file mode 100644
index 0000000..82b62f0
--- /dev/null
+++ b/test/TEST-50-MULTINIC/exports
@@ -0,0 +1 @@
+/nfs/client	192.168.50.0/24(ro,insecure,no_subtree_check,no_root_squash)
diff --git a/test/TEST-50-MULTINIC/hard-off.sh b/test/TEST-50-MULTINIC/hard-off.sh
new file mode 100755
index 0000000..85fd849
--- /dev/null
+++ b/test/TEST-50-MULTINIC/hard-off.sh
@@ -0,0 +1,2 @@
+#!/bin/sh
+getarg rdinitdebug || poweroff -f
diff --git a/test/TEST-50-MULTINIC/hosts b/test/TEST-50-MULTINIC/hosts
new file mode 100644
index 0000000..d02a4e9
--- /dev/null
+++ b/test/TEST-50-MULTINIC/hosts
@@ -0,0 +1,5 @@
+127.0.0.1	localhost
+192.168.50.1	server
+192.168.50.100	client-if1
+192.168.50.101	client-if2
+192.168.50.102	client-if3
diff --git a/test/TEST-50-MULTINIC/server-init b/test/TEST-50-MULTINIC/server-init
new file mode 100755
index 0000000..78eab41
--- /dev/null
+++ b/test/TEST-50-MULTINIC/server-init
@@ -0,0 +1,30 @@
+#!/bin/sh 
+export TERM=linux
+export PS1='nfstest-server:\w\$ '
+stty sane
+echo "made it to the rootfs!"
+echo server > /proc/sys/kernel/hostname
+ip addr add 127.0.0.1/8 dev lo
+ip link set lo up
+ip addr add 192.168.50.1/24 dev eth0
+ip link set eth0 up
+modprobe sunrpc
+mount -t rpc_pipefs sunrpc /var/lib/nfs/rpc_pipefs
+[ -x /sbin/portmap ] && portmap
+[ -x /sbin/rpcbind ] && rpcbind
+modprobe nfsd
+mount -t nfsd nfsd /proc/fs/nfsd
+exportfs -r
+rpc.nfsd
+rpc.mountd
+rpc.idmapd
+exportfs -r
+>/var/lib/dhcpd/dhcpd.leases
+chmod 777 /var/lib/dhcpd/dhcpd.leases
+dhcpd -cf /etc/dhcpd.conf -lf /var/lib/dhcpd/dhcpd.leases
+#sh -i
+# Wait forever for the VM to die
+echo "Serving NFS mounts"
+while sleep 60; do sleep 60; done
+mount -n -o remount,ro /
+poweroff -f
diff --git a/test/TEST-50-MULTINIC/test.log b/test/TEST-50-MULTINIC/test.log
new file mode 100644
index 0000000..f3c0c59
--- /dev/null
+++ b/test/TEST-50-MULTINIC/test.log
@@ -0,0 +1,48 @@
+60+0 records in
+60+0 records out
+62914560 bytes (63 MB) copied, 0.158729 s, 396 MB/s
+mke2fs 1.41.4 (27-Jan-2009)
+Filesystem label=
+OS type: Linux
+Block size=1024 (log=0)
+Fragment size=1024 (log=0)
+15360 inodes, 61440 blocks
+3072 blocks (5.00%) reserved for the super user
+First data block=1
+Maximum filesystem blocks=62914560
+8 block groups
+8192 blocks per group, 8192 fragments per group
+1920 inodes per group
+Superblock backups stored on blocks: 
+	8193, 24577, 40961, 57345
+
+Writing inode tables: 0/8\b\b\b1/8\b\b\b2/8\b\b\b3/8\b\b\b4/8\b\b\b5/8\b\b\b6/8\b\b\b7/8\b\b\bdone                            
+Writing superblocks and filesystem accounting information: done
+
+This filesystem will be automatically checked every 27 mounts or
+180 days, whichever comes first.  Use tune2fs -c or -i to override.
+ln: creating symbolic link `mnt/bin/sh': File exists
+ln: creating symbolic link `mnt/usr/lib64/libnfsidmap_nsswitch.so.0': No such file or directory
+ln: creating symbolic link `mnt/nfs/client/bin/sh': File exists
+
+11804 blocks
+W: Possible missing firmware /lib/firmware/kaweth/trigger_code_fix.bin for module kaweth.ko
+W: Possible missing firmware /lib/firmware/kaweth/trigger_code.bin for module kaweth.ko
+W: Possible missing firmware /lib/firmware/kaweth/new_code_fix.bin for module kaweth.ko
+W: Possible missing firmware /lib/firmware/kaweth/new_code.bin for module kaweth.ko
+W: Possible missing firmware /lib/firmware/sun/cassini.bin for module cassini.ko
+ln: creating symbolic link `/tmp/initramfs.KxYIth/usr/lib64/libnfsidmap_nsswitch.so.0': No such file or directory
+
+42645 blocks
+MULTINIC TEST SETUP: Starting DHCP/NFS server
+setsockopt(IP_ADD_MEMBERSHIP): No such device
+Could not initialize device 'socket'
+Failed to start server
+cat: server.pid: Permission denied
+Usage:
+  kill pid ...              Send SIGTERM to every process listed.
+  kill signal pid ...       Send a signal to every process listed.
+  kill -s signal pid ...    Send a signal to every process listed.
+  kill -l                   List all signal names.
+  kill -L                   List all signal names in a nice table.
+  kill -l signal            Convert between signal numbers and names.
diff --git a/test/TEST-50-MULTINIC/test.sh b/test/TEST-50-MULTINIC/test.sh
new file mode 100755
index 0000000..3d43b13
--- /dev/null
+++ b/test/TEST-50-MULTINIC/test.sh
@@ -0,0 +1,225 @@
+#!/bin/bash
+TEST_DESCRIPTION="root filesystem on NFS with multiple nics"
+
+KVERSION=${KVERSION-$(uname -r)}
+
+# Uncomment this to debug failures
+#DEBUGFAIL="rdinitdebug rdnetdebug"
+
+run_server() {
+    # Start server first
+    echo "MULTINIC TEST SETUP: Starting DHCP/NFS server"
+
+    $testdir/run-qemu -hda server.ext2 -m 512M -nographic \
+	-net nic,macaddr=52:54:00:12:34:56,model=e1000 \
+	-net socket,mcast=230.0.0.1:1234 \
+	-serial udp:127.0.0.1:9999 \
+	-kernel /boot/vmlinuz-$KVERSION \
+	-append "root=/dev/sda rw quiet console=ttyS0,115200n81" \
+	-initrd initramfs.server -pidfile server.pid -daemonize || return 1
+    sudo chmod 644 server.pid || return 1
+
+    # Cleanup the terminal if we have one
+    tty -s && stty sane
+
+    echo Sleeping 10 seconds to give the server a head start
+    sleep 10
+}
+
+client_test() {
+    local test_name="$1"
+    local mac1="$2"
+    local mac2="$3"
+    local mac3="$4"
+    local cmdline="$5"
+    local check="$6"
+
+    echo "CLIENT TEST START: $test_name"
+
+    # Need this so kvm-qemu will boot (needs non-/dev/zero local disk)
+    if ! dd if=/dev/zero of=client.img bs=1M count=1; then
+	echo "Unable to make client sda image" 1>&2
+	return 1
+    fi
+
+    $testdir/run-qemu -hda client.img -m 512M -nographic \
+  	-net nic,macaddr=52:54:00:12:34:$mac1,model=e1000 \
+  	-net nic,macaddr=52:54:00:12:34:$mac2,model=e1000 \
+  	-net nic,macaddr=52:54:00:12:34:$mac3,model=e1000 \
+  	-net socket,mcast=230.0.0.1:1234 \
+  	-kernel /boot/vmlinuz-$KVERSION \
+  	-append "$cmdline $DEBUGFAIL ro quiet console=ttyS0,115200n81" \
+  	-initrd initramfs.testing
+
+    if [[ $? -ne 0 ]] || ! grep -m 1 -q OK client.img; then
+	echo "CLIENT TEST END: $test_name [FAILED - BAD EXIT]"
+	return 1
+    fi
+
+
+    for i in $check ; do
+	echo $i
+	if ! grep -m 1 -q net.$i.ifcfg client.img; then
+	    echo "CLIENT TEST END: $test_name [FAILED - BAD IF]"
+	    return 1
+	fi
+    done
+
+    echo "CLIENT TEST END: $test_name [OK]"
+    return 0
+}
+
+
+test_run() {
+    if ! run_server; then
+	echo "Failed to start server" 1>&2
+	return 1
+    fi
+
+    # Mac Numbering Scheme
+    # ...:00-02 receive IP adresses all others don't
+    # ...:02 receives a dhcp root-path
+
+    # PXE Style BOOTIF=
+    client_test "MULTINIC root=nfs,BOOTIF=" \
+	00 01 02 \
+	"root=nfs:192.168.50.1:/nfs/client BOOTIF=52-54-00-12-34-00" \
+	"eth0" || return 1
+
+    # PXE Style BOOTIF= with dhcp root-path
+    client_test "MULTINIC root=dhcp BOOTIF=" \
+	00 01 02 \
+	"root=dhcp BOOTIF=52-54-00-12-34-02" \
+	"eth2" || return 1
+
+    # Multinic case, where only one nic works
+    client_test "MULTINIC root=nfs ip=dhcp" \
+	FF 00 FE \
+	"root=nfs:192.168.50.1:/nfs/client ip=dhcp" \
+	"eth1" || return 1
+
+    # Require two interfaces
+    client_test "MULTINIC root=nfs ip=eth1:dhcp ip=eth2:dhcp bootdev=eth1" \
+	00 01 02 \
+	"root=nfs:192.168.50.1:/nfs/client ip=eth1:dhcp ip=eth2:dhcp bootdev=eth1" \
+	"eth1 eth2" || return 1
+
+    # Require three interfaces with dhcp root-path
+    client_test "MULTINIC root=dhcp ip=eth0:dhcp ip=eth1:dhcp ip=eth2:dhcp bootdev=eth2" \
+	00 01 02 \
+	"root=dhcp ip=eth0:dhcp ip=eth1:dhcp ip=eth2:dhcp bootdev=eth2" \
+	"eth0 eth1 eth2" || return 1
+}
+
+test_setup() {
+     # Make server root
+     dd if=/dev/zero of=server.ext2 bs=1M count=60
+     mke2fs -F server.ext2
+     mkdir mnt
+     sudo mount -o loop server.ext2 mnt
+
+     kernel=$KVERSION
+     (
+     	initdir=mnt
+ 	. $basedir/dracut-functions
+ 	dracut_install sh ls shutdown poweroff stty cat ps ln ip \
+ 	    /lib/terminfo/l/linux dmesg mkdir cp ping exportfs \
+ 	    modprobe rpc.nfsd rpc.mountd showmount tcpdump \
+ 	    /etc/services sleep mount chmod
+ 	which portmap >/dev/null 2>&1 && dracut_install portmap
+ 	which rpcbind >/dev/null 2>&1 && dracut_install rpcbind
+ 	[ -f /etc/netconfig ] && dracut_install /etc/netconfig 
+ 	which dhcpd >/dev/null 2>&1 && dracut_install dhcpd
+ 	[ -x /usr/sbin/dhcpd3 ] && inst /usr/sbin/dhcpd3 /usr/sbin/dhcpd
+ 	instmods nfsd sunrpc ipv6
+ 	inst ./server-init /sbin/init
+ 	inst ./hosts /etc/hosts
+ 	inst ./exports /etc/exports
+ 	inst ./dhcpd.conf /etc/dhcpd.conf
+ 	dracut_install /etc/nsswitch.conf /etc/rpc /etc/protocols
+ 	dracut_install rpc.idmapd /etc/idmapd.conf
+ 	if ldd $(which rpc.idmapd) |grep -q lib64; then
+ 	    LIBDIR="/lib64"
+ 	else
+ 	    LIBDIR="/lib"
+ 	fi
+
+ 	dracut_install $(ls {/usr,}$LIBDIR/libnfsidmap*.so* 2>/dev/null )
+ 	dracut_install $(ls {/usr,}$LIBDIR/libnss*.so 2>/dev/null)
+ 	(
+ 	    cd "$initdir";
+ 	    mkdir -p dev sys proc etc var/run tmp var/lib/{dhcpd,rpcbind}
+ 	    mkdir -p var/lib/nfs/{v4recovery,rpc_pipefs}
+ 	    chmod 777 var/lib/rpcbind var/lib/nfs
+ 	)
+ 	inst /etc/nsswitch.conf /etc/nsswitch.conf
+ 	inst /etc/passwd /etc/passwd
+ 	inst /etc/group /etc/group
+ 	for i in /lib*/libnss_files**;do
+ 	    inst_library $i
+ 	done
+
+ 	/sbin/depmod -a -b "$initdir" $kernel
+ 	ldconfig -n -r "$initdir" /lib* /usr/lib*
+     )
+
+    # Make client root inside server root
+    initdir=mnt/nfs/client
+    mkdir -p $initdir
+
+    (
+ 	. $basedir/dracut-functions
+ 	dracut_install sh shutdown poweroff stty cat ps ln ip \
+         	/lib/terminfo/l/linux mount dmesg mkdir \
+ 		cp ping grep ls
+ 	inst ./client-init /sbin/init
+ 	(
+ 	    cd "$initdir"
+ 	    mkdir -p dev sys proc etc
+ 	    mkdir -p var/lib/nfs/rpc_pipefs
+	)
+ 	inst /etc/nsswitch.conf /etc/nsswitch.conf
+ 	inst /etc/passwd /etc/passwd
+ 	inst /etc/group /etc/group
+ 	for i in /lib*/libnss_files*;do
+ 	    inst_library $i
+ 	done
+
+ 	ldconfig -n -r "$initdir" /lib* /usr/lib*
+     )
+
+     sudo umount mnt
+     rm -fr mnt
+
+    # Make an overlay with needed tools for the test harness
+    (
+ 	initdir=overlay
+ 	mkdir overlay
+ 	. $basedir/dracut-functions
+ 	dracut_install poweroff shutdown
+ 	inst_simple ./hard-off.sh /emergency/01hard-off.sh
+    )
+
+    # Make server's dracut image
+    $basedir/dracut -l -i overlay / \
+ 	-m "dash udev-rules base rootfs-block debug" \
+ 	-d "ata_piix ext2 sd_mod e1000" \
+ 	-f initramfs.server $KVERSION || return 1
+
+    # Make client's dracut image
+    $basedir/dracut -l -i overlay / \
+	-m "dash udev-rules base network nfs debug" \
+	-d "e1000 nfs sunrpc" \
+	-f initramfs.testing $KVERSION || return 1
+}
+
+test_cleanup() {
+    if [[ -s server.pid ]]; then
+	sudo kill -TERM $(cat server.pid)
+	rm -f server.pid
+    fi
+    rm -rf mnt overlay
+    rm -f server.ext2 client.img initramfs.server initramfs.testing
+}
+
+. $testdir/test-functions
-- 
1.6.0.4

--
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 4/5] Add DAD do dhclient-script

[David Dillow]

On Wed, 2009-06-24 at 20:08 +0200, Philippe Seewer wrote:
> The manpage for dhclient-script says:
> 
> Before actually configuring the address, dhclient-script should
> somehow ARP for it and exit with a nonzero status if it receives a
> reply.
> 
> By using arping in dracut this is very easy, since arping has a
> specific option to do just that.

> --- a/modules.d/40network/dhclient-script
> +++ b/modules.d/40network/dhclient-script
> @@ -64,6 +64,7 @@ case $reason in
>  	sleep 1
>  	;;
>      BOUND)
> +	arping -q -D -c 2 -I $netif $new_ip_address || exit 1

Where's the addition of arping to 40network/install?


--
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 4/5] Add DAD do dhclient-script

[Seewer Philippe]

David Dillow wrote:
> On Wed, 2009-06-24 at 20:08 +0200, Philippe Seewer wrote:
>> The manpage for dhclient-script says:
>>
>> Before actually configuring the address, dhclient-script should
>> somehow ARP for it and exit with a nonzero status if it receives a
>> reply.
>>
>> By using arping in dracut this is very easy, since arping has a
>> specific option to do just that.
> 
>> --- a/modules.d/40network/dhclient-script
>> +++ b/modules.d/40network/dhclient-script
>> @@ -64,6 +64,7 @@ case $reason in
>>  	sleep 1
>>  	;;
>>      BOUND)
>> +	arping -q -D -c 2 -I $netif $new_ip_address || exit 1
> 
> Where's the addition of arping to 40network/install?

*sigh* Thanks for noticing, forgot to add it to the commit. Complete
Patch below

---
 modules.d/40network/dhclient-script |    1 +
 modules.d/40network/install         |    2 +-
 2 files changed, 2 insertions(+), 1 deletions(-)

diff --git a/modules.d/40network/dhclient-script b/modules.d/40network/dhclient-script
index 6df630d..3fb8da6 100755
--- a/modules.d/40network/dhclient-script
+++ b/modules.d/40network/dhclient-script
@@ -64,6 +64,7 @@ case $reason in
 	sleep 1
 	;;
     BOUND)
+	arping -q -D -c 2 -I $netif $new_ip_address || exit 1
 	setup_interface 
 	set | while read line; do
 	    [ "${line#new_}" = "$line" ] && continue
diff --git a/modules.d/40network/install b/modules.d/40network/install
index 0b76cbd..513503c 100755
--- a/modules.d/40network/install
+++ b/modules.d/40network/install
@@ -1,5 +1,5 @@
 #!/bin/bash
-dracut_install ip dhclient hostname
+dracut_install ip dhclient hostname arping
 # Include wired net drivers, excluding wireless
 for modname in $(find "/lib/modules/$kernel/kernel/drivers" -name '*.ko'); do
   if nm -uPA $modname | grep -q eth_type_trans; then
--
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 4/5] Add DAD do dhclient-script

[David Dillow]

On Wed, 2009-06-24 at 21:56 +0200, Seewer Philippe wrote:
> David Dillow wrote:
> > On Wed, 2009-06-24 at 20:08 +0200, Philippe Seewer wrote:
> >> The manpage for dhclient-script says:
> >>
> >> Before actually configuring the address, dhclient-script should
> >> somehow ARP for it and exit with a nonzero status if it receives a
> >> reply.
> >>
> >> By using arping in dracut this is very easy, since arping has a
> >> specific option to do just that.
> > 
> >> --- a/modules.d/40network/dhclient-script
> >> +++ b/modules.d/40network/dhclient-script
> >> @@ -64,6 +64,7 @@ case $reason in
> >>  	sleep 1
> >>  	;;
> >>      BOUND)
> >> +	arping -q -D -c 2 -I $netif $new_ip_address || exit 1

Also, this seems like it is screaming for a message -- it will confuse
the user if it just doesn't work without saying why....

--
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 5/5] Add multinic test suite

[David Dillow]

On Wed, 2009-06-24 at 20:08 +0200, Philippe Seewer wrote:
> This is a start at a test suite for the multinic part of dracut

Seems sane, some minor issues.

> diff --git a/test/TEST-50-MULTINIC/test.log b/test/TEST-50-MULTINIC/test.log
> new file mode 100644
> index 0000000..f3c0c59
> --- /dev/null
> +++ b/test/TEST-50-MULTINIC/test.log

This shouldn't be part of the commit... 

> +MULTINIC TEST SETUP: Starting DHCP/NFS server
> +setsockopt(IP_ADD_MEMBERSHIP): No such device
> +Could not initialize device 'socket'
> +Failed to start server
> +cat: server.pid: Permission denied
> +Usage:
> +  kill pid ...              Send SIGTERM to every process listed.

What's this all about? Does the test suite work?



--
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 2/5] Give the kernel some time to up interfaces

[David Dillow]

On Wed, 2009-06-24 at 20:08 +0200, Philippe Seewer wrote:
> It does not really matter if the kernel needs some time or if
> the ethernet handshake takes a while. It's a good idea to wait
> a second after upping an interface to let things settle.

Hmm, shouldn't DHCP retries take care of this issue, and the server ARP
ping we've been talking about take care of it for static configs?

I would be a bit concerned that the backoff of dhclient may slow us down
more than a second, but I've not investigated if that is true.

I'm not really keen on adding a mandatory delay unless there is no other
option.

--
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 0/5] Netroot updates and multinic test-suite

[David Dillow]

On Wed, 2009-06-24 at 20:08 +0200, Philippe Seewer wrote:
> This is a first start at a multinic testsuite, some
> improvements to netroot and a bugfix.
> 
> Patches can be pulled from 
> git://github.com/seewer/dracut-seewer.git
> as well

The BOOTIF bug fix could go as-is, but I think the rest of the series
needs to be respun once the issues are addressed.

--
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 5/5] Add multinic test suite

[Seewer Philippe]

David Dillow wrote:
> On Wed, 2009-06-24 at 20:08 +0200, Philippe Seewer wrote:
>> This is a start at a test suite for the multinic part of dracut
> 
> Seems sane, some minor issues.
> 
>> diff --git a/test/TEST-50-MULTINIC/test.log b/test/TEST-50-MULTINIC/test.log
>> new file mode 100644
>> index 0000000..f3c0c59
>> --- /dev/null
>> +++ b/test/TEST-50-MULTINIC/test.log
> 
> This shouldn't be part of the commit... 

*sigh* Yes. Apologies.

>> +MULTINIC TEST SETUP: Starting DHCP/NFS server
>> +setsockopt(IP_ADD_MEMBERSHIP): No such device
>> +Could not initialize device 'socket'
>> +Failed to start server
>> +cat: server.pid: Permission denied
>> +Usage:
>> +  kill pid ...              Send SIGTERM to every process listed.
> 
> What's this all about? Does the test suite work?

That is weird. The server startup code is the same as with the
NFS and NDB (except for 512MB allocated instead of 128MB).

The only thing I noticed, but that goes for all test-suites, is that if 
you don't have any local network connections the servers won't start 
producing exactly your error.

Do the other tests run with you?

Thanks,
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 2/5] Give the kernel some time to up interfaces

[Seewer Philippe]

David Dillow wrote:
> On Wed, 2009-06-24 at 20:08 +0200, Philippe Seewer wrote:
>> It does not really matter if the kernel needs some time or if
>> the ethernet handshake takes a while. It's a good idea to wait
>> a second after upping an interface to let things settle.
> 
> Hmm, shouldn't DHCP retries take care of this issue, and the server ARP
> ping we've been talking about take care of it for static configs?
> 
> I would be a bit concerned that the backoff of dhclient may slow us down
> more than a second, but I've not investigated if that is true.
> 
> I'm not really keen on adding a mandatory delay unless there is no other
> option.
> 

In the DHCP case, yes we could skip the sleep as dhclient will just
retry if the interface isn't ready. But for all other cases, we really
should give the hardware some time to initialize. 

During the years, I've had to use this for different cases, be it
slow firmware loading, weird drivers or just hardware that takes
longer. I've even had interfaces failing configuration if they didn't 
get some time to initialize. Another example: My e1000e card actually
takes just a bit more than 1 second to got to LOWER_UP when I do
modprobe e1000e && ip link set eth0 up

So this patch is just plain paranoia to make sure there's enough time
between upping the interface and trying to configure it. Yes, we could 
try to wait for LOWER_UP or something similar, but sleep 1 is simpler
and I haven't yet seen a case where that isn't enough and the slow down
is negligible.

As for other references: Stock isc 4.1 dhclient-scripts do it as well
for example. I admit, others doing it doesn't make it right. But it's 
just a pointer.

--
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