From d81182cf69a4f188f304fcce6d651ffd56b67aac Mon Sep 17 00:00:00 2001 From: "Kaleb S. KEITHLEY" Date: Tue, 17 Mar 2015 09:27:05 -0400 Subject: NFS-Ganesha: Install scripts, config files, and resource agent scripts Resubmitting after a gerrit bug bungled the merge of http://review.gluster.org/9621 (was it really a gerrit bug?) Scripts related to NFS-Ganesha are in extras/ganesha/scripts. Config files are in extras/ganesha/config. Resource Agent files are in extras/ganesha/ocf Files are copied to appropriate locations. Change-Id: I137169f4d653ee2b7d6df14d41e2babd0ae8d10c BUG: 1188184 Signed-off-by: Kaleb S. KEITHLEY Reviewed-on: http://review.gluster.org/9912 Tested-by: Gluster Build System --- extras/Makefile.am | 2 +- extras/ganesha/Makefile.am | 2 + extras/ganesha/config/Makefile.am | 4 + extras/ganesha/config/ganesha-ha.conf.sample | 15 + extras/ganesha/ocf/Makefile.am | 12 + extras/ganesha/ocf/ganesha_grace | 168 ++++++ extras/ganesha/ocf/ganesha_mon | 159 ++++++ extras/ganesha/ocf/ganesha_nfsd | 196 +++++++ extras/ganesha/scripts/Makefile.am | 4 + extras/ganesha/scripts/create-export-ganesha.sh | 55 ++ extras/ganesha/scripts/dbus-send.sh | 74 +++ extras/ganesha/scripts/ganesha-ha.sh | 665 ++++++++++++++++++++++++ 12 files changed, 1355 insertions(+), 1 deletion(-) create mode 100644 extras/ganesha/Makefile.am create mode 100644 extras/ganesha/config/Makefile.am create mode 100644 extras/ganesha/config/ganesha-ha.conf.sample create mode 100644 extras/ganesha/ocf/Makefile.am create mode 100644 extras/ganesha/ocf/ganesha_grace create mode 100644 extras/ganesha/ocf/ganesha_mon create mode 100644 extras/ganesha/ocf/ganesha_nfsd create mode 100644 extras/ganesha/scripts/Makefile.am create mode 100755 extras/ganesha/scripts/create-export-ganesha.sh create mode 100755 extras/ganesha/scripts/dbus-send.sh create mode 100755 extras/ganesha/scripts/ganesha-ha.sh (limited to 'extras') diff --git a/extras/Makefile.am b/extras/Makefile.am index 28a66003042..e2b29f2da45 100644 --- a/extras/Makefile.am +++ b/extras/Makefile.am @@ -2,7 +2,7 @@ EditorModedir = $(docdir) EditorMode_DATA = glusterfs-mode.el glusterfs.vim SUBDIRS = init.d systemd benchmarking hook-scripts $(OCF_SUBDIR) LinuxRPM \ - $(GEOREP_EXTRAS_SUBDIR) + $(GEOREP_EXTRAS_SUBDIR) ganesha confdir = $(sysconfdir)/glusterfs conf_DATA = glusterfs-logrotate gluster-rsyslog-7.2.conf gluster-rsyslog-5.8.conf \ diff --git a/extras/ganesha/Makefile.am b/extras/ganesha/Makefile.am new file mode 100644 index 00000000000..9eaa401b6c8 --- /dev/null +++ b/extras/ganesha/Makefile.am @@ -0,0 +1,2 @@ +SUBDIRS = scripts config ocf +CLEANFILES = diff --git a/extras/ganesha/config/Makefile.am b/extras/ganesha/config/Makefile.am new file mode 100644 index 00000000000..c729273096e --- /dev/null +++ b/extras/ganesha/config/Makefile.am @@ -0,0 +1,4 @@ +EXTRA_DIST= ganesha-ha.conf.sample + +confdir = $(sysconfdir)/ganesha +conf_DATA = ganesha-ha.conf.sample diff --git a/extras/ganesha/config/ganesha-ha.conf.sample b/extras/ganesha/config/ganesha-ha.conf.sample new file mode 100644 index 00000000000..68408e0e175 --- /dev/null +++ b/extras/ganesha/config/ganesha-ha.conf.sample @@ -0,0 +1,15 @@ +# Name of the HA cluster created. +HA_NAME="ganesha-ha-360" +# Shared volume to store NFS state. +HA_VOL_NAME="ha-state-volname" +# Mount point of the shared volume. +HA_VOL_MNT="/mnt-nfs" +# The server on which shared volume is created. +HA_VOL_SERVER="server1" +# The subset of nodes of the Gluster Trusted Pool +# that forms the ganesha HA cluster. IP/Hostname +# is specified. +HA_CLUSTER_NODES="server1,server2,..." +# Virtual IPs of each of the nodes specified above. +VIP_server1="10.x.x.x" +VIP_server2="10.x.x.x" diff --git a/extras/ganesha/ocf/Makefile.am b/extras/ganesha/ocf/Makefile.am new file mode 100644 index 00000000000..6aed9548a0f --- /dev/null +++ b/extras/ganesha/ocf/Makefile.am @@ -0,0 +1,12 @@ +EXTRA_DIST= ganesha_grace ganesha_mon ganesha_nfsd + +# The root of the OCF resource agent hierarchy +# Per the OCF standard, it's always "lib", +# not "lib64" (even on 64-bit platforms). +ocfdir = $(prefix)/lib/ocf + +# The provider directory +radir = $(ocfdir)/resource.d/heartbeat + +ra_SCRIPTS = ganesha_grace ganesha_mon ganesha_nfsd + diff --git a/extras/ganesha/ocf/ganesha_grace b/extras/ganesha/ocf/ganesha_grace new file mode 100644 index 00000000000..45819c9a619 --- /dev/null +++ b/extras/ganesha/ocf/ganesha_grace @@ -0,0 +1,168 @@ +#!/bin/bash +# +# Copyright (c) 2014 Anand Subramanian anands@redhat.com +# Copyright (c) 2015 Red Hat Inc. +# All Rights Reserved. +# +# This program is free software; you can redistribute it and/or modify +# it under the terms of version 2 of the GNU General Public License as +# published by the Free Software Foundation. +# +# This program is distributed in the hope that it would be useful, but +# WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. +# +# Further, this software is distributed without any warranty that it is +# free of the rightful claim of any third person regarding infringement +# or the like. Any license provided herein, whether implied or +# otherwise, applies only to this software file. Patent licenses, if +# any, provided herein do not apply to combinations of this program with +# other software, or any other product whatsoever. +# +# You should have received a copy of the GNU General Public License +# along with this program; if not, write the Free Software Foundation, +# Inc., 59 Temple Place - Suite 330, Boston MA 02111-1307, USA. +# +# + +# Initialization: +: ${OCF_FUNCTIONS_DIR=${OCF_ROOT}/lib/heartbeat} +. ${OCF_FUNCTIONS_DIR}/ocf-shellfuncs + +if [ -n "$OCF_DEBUG_LIBRARY" ]; then + . $OCF_DEBUG_LIBRARY +else + : ${OCF_FUNCTIONS_DIR=${OCF_ROOT}/lib/heartbeat} +. ${OCF_FUNCTIONS_DIR}/ocf-shellfuncs +fi + +ganesha_meta_data() { + cat < + + +1.0 + + +This Linux-specific resource agent acts as a dummy +resource agent for nfs-ganesha. + + +Manages the user-space nfs-ganesha NFS server + + + + + + + + + + + + +END + +return $OCF_SUCCESS +} + +ganesha_grace_usage() { + echo "ganesha.nfsd USAGE" +} + +# Make sure meta-data and usage always succeed +case $__OCF_ACTION in + meta-data) ganesha_meta_data + exit $OCF_SUCCESS + ;; + usage|help) ganesha_usage + exit $OCF_SUCCESS + ;; + *) + ;; +esac + +ganesha_grace_start() +{ + local result="" + local resourcename="" + local deadserver="" + local tmpIFS=${IFS} + + # logger "ganesha_grace_start()" + # we're here because somewhere in the cluster one or more + # of the ganesha.nfsds have died, triggering a floating IP + # address to move. Resource constraint location rules ensure + # that this is invoked before the floating IP is moved. + if [ -d /proc/$(cat /var/run/ganesha.nfsd.pid) ]; then + # my ganesha.nfsd is still running + # find out which one died? + + pcs status | grep dead_ip-1 | sort > /tmp/.pcs_status + + result=$(diff /var/run/ganesha/pcs_status /tmp/.pcs_status | grep '^>') + if [[ ${result} ]]; then + # logger "ganesha_grace_start(), ${result}" + IFS=$'\n' + for line in ${result}; do + resourcename=$(echo ${line} | cut -f 1 | cut -d ' ' -f 3) + deadserver=${resourcename%"-dead_ip-1"} + + if [[ ${deadserver} ]]; then + # logger "ganesha_grace_start(), ${line}" + # logger "ganesha_grace_start(), dbus-send --print-reply --system --dest=org.ganesha.nfsd /org/ganesha/nfsd/admin org.ganesha.nfsd.admin.grace string:${deadserver}" + dbus-send --print-reply --system --dest=org.ganesha.nfsd /org/ganesha/nfsd/admin org.ganesha.nfsd.admin.grace string:${deadserver} + if [ $? -ne 0 ]; then + logger "warning: dbus-send --print-reply --system --dest=org.ganesha.nfsd /org/ganesha/nfsd/admin org.ganesha.nfsd.admin.grace string:${deadserver} failed" + fi + fi + done + IFS=${tmpIFS} + fi + + fi + return $OCF_SUCCESS +} + +ganesha_grace_stop() +{ + + # logger "ganesha_grace_stop()" + return $OCF_SUCCESS +} + +ganesha_grace_monitor() +{ + # logger "ganesha_grace_monitor()" + pcs status | grep dead_ip-1 | sort > /var/run/ganesha/pcs_status + return $OCF_SUCCESS +} + +ganesha_grace_validate() +{ + return $OCF_SUCCESS +} + +ganesha_grace_validate + +# logger "ganesha_grace ${OCF_RESOURCE_INSTANCE} $__OCF_ACTION" + +# Translate each action into the appropriate function call +case $__OCF_ACTION in +start) ganesha_grace_start + ;; +stop) ganesha_grace_stop + ;; +status|monitor) ganesha_grace_monitor + ;; +*) ganesha_grace_usage + exit $OCF_ERR_UNIMPLEMENTED + ;; +esac + +rc=$? + +# The resource agent may optionally log a debug message +ocf_log debug "${OCF_RESOURCE_INSTANCE} $__OCF_ACTION returned $rc" +exit $rc + diff --git a/extras/ganesha/ocf/ganesha_mon b/extras/ganesha/ocf/ganesha_mon new file mode 100644 index 00000000000..c03d7ee5ec9 --- /dev/null +++ b/extras/ganesha/ocf/ganesha_mon @@ -0,0 +1,159 @@ +#!/bin/bash +# +# Copyright (c) 2014 Anand Subramanian anands@redhat.com +# Copyright (c) 2015 Red Hat Inc. +# All Rights Reserved. +# +# This program is free software; you can redistribute it and/or modify +# it under the terms of version 2 of the GNU General Public License as +# published by the Free Software Foundation. +# +# This program is distributed in the hope that it would be useful, but +# WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. +# +# Further, this software is distributed without any warranty that it is +# free of the rightful claim of any third person regarding infringement +# or the like. Any license provided herein, whether implied or +# otherwise, applies only to this software file. Patent licenses, if +# any, provided herein do not apply to combinations of this program with +# other software, or any other product whatsoever. +# +# You should have received a copy of the GNU General Public License +# along with this program; if not, write the Free Software Foundation, +# Inc., 59 Temple Place - Suite 330, Boston MA 02111-1307, USA. +# +# + +# Initialization: +: ${OCF_FUNCTIONS_DIR=${OCF_ROOT}/lib/heartbeat} +. ${OCF_FUNCTIONS_DIR}/ocf-shellfuncs + +if [ -n "$OCF_DEBUG_LIBRARY" ]; then + . $OCF_DEBUG_LIBRARY +else + : ${OCF_FUNCTIONS_DIR=${OCF_ROOT}/lib/heartbeat} +. ${OCF_FUNCTIONS_DIR}/ocf-shellfuncs +fi + +ganesha_meta_data() { + cat < + + +1.0 + + +This Linux-specific resource agent acts as a dummy +resource agent for nfs-ganesha. + + +Manages the user-space nfs-ganesha NFS server + + + + + + + + + + + + +END + +return $OCF_SUCCESS +} + +ganesha_mon_usage() { + echo "ganesha.nfsd USAGE" +} + +# Make sure meta-data and usage always succeed +case $__OCF_ACTION in + meta-data) ganesha_meta_data + exit $OCF_SUCCESS + ;; + usage|help) ganesha_usage + exit $OCF_SUCCESS + ;; + *) + ;; +esac + +ganesha_mon_start() +{ + return $OCF_SUCCESS +} + +ganesha_mon_stop() +{ + return $OCF_SUCCESS +} + +ganesha_mon_monitor() +{ + local short_host=$(hostname -s) + + if [ -d /proc/$(cat /var/run/ganesha.nfsd.pid) ]; then + # logger "ganesha_mon_monitor(), attrd_updater -n ganesha-active -v 1" + pcs resource delete ${short_host}-dead_ip-1 + # if [ $? -ne 0 ]; then + # logger "warning: pcs resource delete ${short_host}-dead_ip-1" + # fi + + sleep 1 + + attrd_updater -n ganesha-active -v 1 + if [ $? -ne 0 ]; then + logger "warning: attrd_updater -n ganesha-active -v 1 failed" + fi + + else + # logger "ganesha_mon_monitor(), attrd_updater --D -n ganesha-active" + + pcs resource create ${short_host}-dead_ip-1 ocf:heartbeat:Dummy + if [ $? -ne 0 ]; then + logger "warning: pcs resource create ${short_host}-dead_ip-1 ocf:heartbeat:Dummy failed" + fi + + sleep 1 + + attrd_updater -D -n ganesha-active + if [ $? -ne 0 ]; then + logger "warning: attrd_updater -D -n ganesha-active failed" + fi + fi + + return $OCF_SUCCESS +} + +ganesha_mon_validate() +{ + return $OCF_SUCCESS +} + +ganesha_mon_validate + +# logger "ganesha_mon ${OCF_RESOURCE_INSTANCE} $__OCF_ACTION" + +# Translate each action into the appropriate function call +case $__OCF_ACTION in +start) ganesha_mon_start + ;; +stop) ganesha_mon_stop + ;; +status|monitor) ganesha_mon_monitor + ;; +*) ganesha_mon_usage + exit $OCF_ERR_UNIMPLEMENTED + ;; +esac + +rc=$? + +# The resource agent may optionally log a debug message +ocf_log debug "${OCF_RESOURCE_INSTANCE} $__OCF_ACTION returned $rc" +exit $rc + diff --git a/extras/ganesha/ocf/ganesha_nfsd b/extras/ganesha/ocf/ganesha_nfsd new file mode 100644 index 00000000000..2c0b32c7817 --- /dev/null +++ b/extras/ganesha/ocf/ganesha_nfsd @@ -0,0 +1,196 @@ +#!/bin/bash +# +# Copyright (c) 2014 Anand Subramanian anands@redhat.com +# Copyright (c) 2015 Red Hat Inc. +# All Rights Reserved. +# +# This program is free software; you can redistribute it and/or modify +# it under the terms of version 2 of the GNU General Public License as +# published by the Free Software Foundation. +# +# This program is distributed in the hope that it would be useful, but +# WITHOUT ANY WARRANTY; without even the implied warranty of +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. +# +# Further, this software is distributed without any warranty that it is +# free of the rightful claim of any third person regarding infringement +# or the like. Any license provided herein, whether implied or +# otherwise, applies only to this software file. Patent licenses, if +# any, provided herein do not apply to combinations of this program with +# other software, or any other product whatsoever. +# +# You should have received a copy of the GNU General Public License +# along with this program; if not, write the Free Software Foundation, +# Inc., 59 Temple Place - Suite 330, Boston MA 02111-1307, USA. +# +# + +# Initialization: +: ${OCF_FUNCTIONS_DIR=${OCF_ROOT}/lib/heartbeat} +. ${OCF_FUNCTIONS_DIR}/ocf-shellfuncs + +if [ -n "$OCF_DEBUG_LIBRARY" ]; then + . $OCF_DEBUG_LIBRARY +else + : ${OCF_FUNCTIONS_DIR=${OCF_ROOT}/lib/heartbeat} +. ${OCF_FUNCTIONS_DIR}/ocf-shellfuncs +fi + +ganesha_meta_data() { + cat < + + +1.0 + + +This Linux-specific resource agent acts as a dummy +resource agent for nfs-ganesha. + + +Manages the user-space nfs-ganesha NFS server + + + +HA State Volume Name +HA_State Volume Name + + + + +HA State Volume Mount Point +HA_State Volume Mount Point + + + + +HA State Volume Server +HA_State Volume Server + + + + + + + + + + + + +END + +return $OCF_SUCCESS +} + +ganesha_nfsd_usage() { + echo "ganesha.nfsd USAGE" +} + +# Make sure meta-data and usage always succeed +case $__OCF_ACTION in + meta-data) ganesha_meta_data + exit $OCF_SUCCESS + ;; + usage|help) ganesha_usage + exit $OCF_SUCCESS + ;; + *) + ;; +esac + +ganesha_nfsd_start() +{ + return $OCF_SUCCESS +} + +ganesha_nfsd_stop() +{ + local mounted="" + local mntptinuse="" + local instance_host="" + local short_host="" + local resource_prefix=${OCF_RESOURCE_INSTANCE:0:9} + + if [ "X${resource_prefix}X" = "Xnfs_startX" ]; then + + mounted=$(mount | grep $OCF_RESKEY_ha_vol_name) + mntptinuse=$(mount | grep -o $OCF_RESKEY_ha_vol_mnt) + short_host=$(hostname -s) + long_host=$(hostname) + + if [[ ! ${mounted} ]]; then + + if [ -d $OCF_RESKEY_ha_vol_mnt ]; then + if [[ ${mntptinuse} ]]; then + return $OCF_ERR_GENERIC + fi + else + mkdir ${mntpt} + fi + + mount -t glusterfs $OCF_RESKEY_ha_vol_server:$OCF_RESKEY_ha_vol_name $OCF_RESKEY_ha_vol_mnt + if [ $? -ne 0 ]; then + logger "warning: mount -t glusterfs $OCF_RESKEY_ha_vol_server:$OCF_RESKEY_ha_vol_name $OCF_RESKEY_ha_vol_mnt failed" + fi + + mv /var/lib/nfs /var/lib/nfs.backup + ln -s $OCF_RESKEY_ha_vol_mnt/${long_host}/nfs /var/lib/nfs + if [ $? -ne 0 ]; then + logger "warning: ln -s $OCF_RESKEY_ha_vol_mnt/${long_host}/nfs /var/lib/nfs failed" + fi + + fi + + service nfs-ganesha start + if [ $? -ne 0 ]; then + logger "warning: service nfs-ganesha start failed" + fi + else + umount $OCF_RESKEY_ha_vol_mnt + if [ $? -ne 0 ]; then + logger "warning: umount $OCF_RESKEY_ha_vol_mnt failed" + fi + + service nfs-ganesha stop + if [ $? -ne 0 ]; then + logger "warning: service nfs-ganesha stop failed" + fi + fi + + return $OCF_SUCCESS +} + +ganesha_nfsd_monitor() +{ + return $OCF_SUCCESS +} + +ganesha_nfsd_validate() +{ + return $OCF_SUCCESS +} + +ganesha_nfsd_validate + +# logger "ganesha_nfsd ${OCF_RESOURCE_INSTANCE} $__OCF_ACTION" + +# Translate each action into the appropriate function call +case $__OCF_ACTION in +start) ganesha_nfsd_start + ;; +stop) ganesha_nfsd_stop + ;; +status|monitor) ganesha_nfsd_monitor + ;; +*) ganesha_nfsd_usage + exit $OCF_ERR_UNIMPLEMENTED + ;; +esac + +rc=$? + +# The resource agent may optionally log a debug message +ocf_log debug "${OCF_RESOURCE_INSTANCE} $__OCF_ACTION returned $rc" +exit $rc + diff --git a/extras/ganesha/scripts/Makefile.am b/extras/ganesha/scripts/Makefile.am new file mode 100644 index 00000000000..f18ad66494d --- /dev/null +++ b/extras/ganesha/scripts/Makefile.am @@ -0,0 +1,4 @@ +EXTRA_DIST= ganesha-ha.sh dbus-send.sh create-export-ganesha.sh + +scriptsdir = $(libexecdir)/ganesha +scripts_DATA = create-export-ganesha.sh dbus-send.sh ganesha-ha.sh diff --git a/extras/ganesha/scripts/create-export-ganesha.sh b/extras/ganesha/scripts/create-export-ganesha.sh new file mode 100755 index 00000000000..8140a11f548 --- /dev/null +++ b/extras/ganesha/scripts/create-export-ganesha.sh @@ -0,0 +1,55 @@ +#/bin/bash + +#This script is called by glusterd when the user +#tries to export a volume via NFS-Ganesha. +#An export file specific to a volume +#is created in GANESHA_DIR/exports. + +GANESHA_DIR=$1 +VOL=$2 + +function check_cmd_status() +{ + if [ "$1" != "0" ] + then + rm -rf $GANESHA_DIR/exports/export.$VOL.conf + exit 1 + fi +} + + +if [ ! -d "$GANESHA_DIR/exports" ]; + then + mkdir $GANESHA_DIR/exports + check_cmd_status `echo $?` +fi + +CONF=$(cat /etc/sysconfig/ganesha | grep "CONFFILE" | cut -f 2 -d "=") +check_cmd_status `echo $?` + + +function write_conf() +{ +echo -e "# WARNING : Using Gluster CLI will overwrite manual +# changes made to this file. To avoid it, edit the +# file, copy it over to all the NFS-Ganesha nodes +# and run ganesha-ha.sh --refresh-config." + +echo "EXPORT{" +echo " Export_Id = 1;" +echo " FSAL {" +echo " name = "GLUSTER";" +echo " hostname=\"localhost\";" +echo " volume=\"$VOL\";" +echo " }" +echo " Access_type = RW;" +echo ' Squash="No_root_squash";' +echo " Pseudo=\"/$VOL\";" +echo ' Protocols = "3,4" ;' +echo ' Transports = "UDP,TCP";' +echo ' SecType = "sys";' +echo " }" +} + +write_conf $@ > $GANESHA_DIR/exports/export.$VOL.conf +echo "%include \"$GANESHA_DIR/exports/export.$VOL.conf\"" >> $CONF diff --git a/extras/ganesha/scripts/dbus-send.sh b/extras/ganesha/scripts/dbus-send.sh new file mode 100755 index 00000000000..406162f0747 --- /dev/null +++ b/extras/ganesha/scripts/dbus-send.sh @@ -0,0 +1,74 @@ +#/bin/bash + +declare -i EXPORT_ID +GANESHA_DIR=$1 +OPTION=$1 +VOL=$2 + +function check_cmd_status() +{ + if [ "$1" != "0" ] + then + rm -rf $GANESHA_DIR/exports/export.$VOL.conf + exit 1 + fi +} + +#This function keeps track of export IDs and increments it with every new entry +function dynamic_export_add() +{ + count=`ls -l $GANESHA_DIR/exports/*.conf | wc -l` + if [ "$count" = "1" ] ; + then + EXPORT_ID=2 + else + #if [ -s /var/lib/ganesha/export_removed ]; + # then + # EXPORT_ID=`head -1 /var/lib/ganesha/export_removed` + # sed -i -e "1d" /var/lib/ganesha/export_removed + # else + + EXPORT_ID=`cat $GANESHA_DIR/.export_added` + check_cmd_status `echo $?` + EXPORT_ID=EXPORT_ID+1 + #fi + fi + echo $EXPORT_ID > $GANESHA_DIR/.export_added + check_cmd_status `echo $?` + sed -i s/Export_Id.*/"Export_Id= $EXPORT_ID ;"/ \ +$GANESHA_DIR/exports/export.$VOL.conf + check_cmd_status `echo $?` + dbus-send --print-reply --system \ +--dest=org.ganesha.nfsd /org/ganesha/nfsd/ExportMgr \ +org.ganesha.nfsd.exportmgr.AddExport string:$GANESHA_DIR/exports/export.$VOL.conf \ +string:"EXPORT(Path=/$VOL)" + +} + +#This function removes an export dynamically(uses the export_id of the export) +function dynamic_export_remove() +{ + removed_id=`cat $GANESHA_DIR/exports/export.$VOL.conf |\ +grep Export_Id | cut -d " " -f3` + echo $removed_id + check_cmd_status `echo $?` + dbus-send --print-reply --system \ +--dest=org.ganesha.nfsd /org/ganesha/nfsd/ExportMgr \ +org.ganesha.nfsd.exportmgr.RemoveExport uint16:$removed_id + check_cmd_status `echo $?` + rm -rf $GANESHA_DIR/exports/export.$VOL.conf + +} + +if [ "$OPTION" = "on" ]; +then + dynamic_export_add $@ + check_cmd_status `echo $?` +fi + +if [ "$OPTION" = "off" ]; +then + dynamic_export_remove $@ + check_cmd_status `echo $?` +fi + diff --git a/extras/ganesha/scripts/ganesha-ha.sh b/extras/ganesha/scripts/ganesha-ha.sh new file mode 100755 index 00000000000..8fea7ff6dfd --- /dev/null +++ b/extras/ganesha/scripts/ganesha-ha.sh @@ -0,0 +1,665 @@ +#!/bin/bash + +# Pacemaker+Corosync High Availability for NFS-Ganesha +# +# setup, teardown, add-node, delete-node, refresh-config, and status +# +# Each participating node in the cluster is assigned a virtual IP (VIP) +# which fails over to another node when its associated ganesha.nfsd dies +# for any reason. After the VIP is moved to another node all the +# ganesha.nfsds are send a signal using DBUS to put them into NFS GRACE. +# +# There are six resource agent types used: ganesha_mon, ganesha_grace, +# ganesha_nfsd, IPaddr, and Dummy. ganesha_mon is used to monitor the +# ganesha.nfsd. ganesha_grace is used to send the DBUS signal to put +# the remaining ganesha.nfsds into grace. ganesha_nfsd is used to start +# and stop the ganesha.nfsd during setup and teardown. IPaddr manages +# the VIP. A Dummy resource named $hostname-trigger_ip-1 is used to +# ensure that the NFS GRACE DBUS signal is sent after the VIP moves to +# the new host. + +HA_NUM_SERVERS=0 +HA_SERVERS="" +HA_CONFDIR="" + +RHEL6_PCS_CNAME_OPTION="--name" + +check_cluster_exists() +{ + local name=${1} + local cluster_name="" + + if [ -e /var/run/corosync.pid ]; then + cluster_name=$(pcs status | grep "Cluster name:" | cut -d ' ' -f 3) + if [ ${cluster_name} -a ${cluster_name} = ${name} ]; then + logger "$name already exists, exiting" + exit 0 + fi + fi +} + +determine_servers() +{ + local cmd=${1} + local num_servers=0 + local tmp_ifs=${IFS} + local ha_servers="" + + if [[ "X${cmd}X" != "XteardownX" ]]; then + IFS=$',' + for server in ${HA_CLUSTER_NODES} ; do + num_servers=$(expr ${num_servers} + 1) + done + IFS=${tmp_ifs} + HA_NUM_SERVERS=${num_servers} + HA_SERVERS="${HA_CLUSTER_NODES//,/ }" + else + ha_servers=$(pcs status | grep "Online:" | grep -o '\[.*\]' | sed -e 's/\[//' | sed -e 's/\]//') + IFS=$' ' + for server in ${ha_servers} ; do + num_servers=$(expr ${num_servers} + 1) + done + IFS=${tmp_ifs} + HA_NUM_SERVERS=${num_servers} + HA_SERVERS="${ha_servers}" + fi +} + +setup_cluster() +{ + local name=${1} + local num_servers=${2} + local servers=${3} + local unclean="" + + logger "setting up cluster ${name} with the following ${servers}" + + pcs cluster auth ${servers} +# fedora pcs cluster setup ${name} ${servers} +# rhel6 pcs cluster setup --name ${name} ${servers} + pcs cluster setup ${RHEL6_PCS_CNAME_OPTION} ${name} ${servers} + if [ $? -ne 0 ]; then + logger "pcs cluster setup ${RHEL6_PCS_CNAME_OPTION} ${name} ${servers} failed" + exit 1; + fi + pcs cluster start --all + if [ $? -ne 0 ]; then + logger "pcs cluster start failed" + exit 1; + fi + + sleep 3 + unclean=$(pcs status | grep -u "UNCLEAN") + while [[ "${unclean}X" = "UNCLEANX" ]]; do + sleep 1 + unclean=$(pcs status | grep -u "UNCLEAN") + done + sleep 1 + + if [ ${num_servers} -lt 3 ]; then + pcs property set no-quorum-policy=ignore + if [ $? -ne 0 ]; then + logger "warning: pcs property set no-quorum-policy=ignore failed" + fi + fi + pcs property set stonith-enabled=false + if [ $? -ne 0 ]; then + logger "warning: pcs property set stonith-enabled=false failed" + fi +} + +setup_finalize() +{ + local cibfile=${1} + local stopped="" + + stopped=$(pcs status | grep -u "Stopped") + while [[ "${stopped}X" = "StoppedX" ]]; do + sleep 1 + stopped=$(pcs status | grep -u "Stopped") + done + + pcs status | grep dead_ip-1 | sort > /var/run/ganesha/pcs_status + +} + +teardown_cluster() +{ + local name=${1} + + logger "tearing down cluster $name" + + for server in ${HA_SERVERS} ; do + if [[ ${HA_CLUSTER_NODES} != *${server}* ]]; then + logger "info: ${server} is not in config, removing" + + pcs cluster stop ${server} + if [ $? -ne 0 ]; then + logger "pcs cluster stop ${server}" + fi + + pcs cluster node remove ${server} + if [ $? -ne 0 ]; then + logger "warning: pcs cluster node remove ${server} failed" + fi + fi + done + +# BZ 1193433 - pcs doesn't reload cluster.conf after modification +# after teardown completes, a subsequent setup will appear to have +# 'remembered' the deleted node. You can work around this by +# issuing another `pcs cluster node remove $node`, +# `crm_node -f -R $server`, or +# `cibadmin --delete --xml-text '' + + pcs cluster stop --all + if [ $? -ne 0 ]; then + logger "warning pcs cluster stop --all failed" + fi + + pcs cluster destroy + if [ $? -ne 0 ]; then + logger "error pcs cluster destroy failed" + exit 1 + fi +} + +do_create_virt_ip_constraints() +{ + local cibfile=${1}; shift + local primary=${1}; shift + local weight="1000" + + # first a constraint location rule that says the VIP must be where + # there's a ganesha.nfsd running + pcs -f ${cibfile} constraint location ${primary}-cluster_ip-1 rule score=-INFINITY ganesha-active ne 1 + if [ $? -ne 0 ]; then + logger "warning: pcs constraint location ${primary}-cluster_ip-1 rule score=-INFINITY ganesha-active ne 1 failed" + fi + + # then a set of constraint location prefers to set the prefered order + # for where a VIP should move + while [[ ${1} ]]; do + pcs -f ${cibfile} constraint location ${primary}-cluster_ip-1 prefers ${1}=${weight} + if [ $? -ne 0 ]; then + logger "warning: pcs constraint location ${primary}-cluster_ip-1 prefers ${1}=${weight} failed" + fi + weight=$(expr ${weight} + 1000) + shift + done + # and finally set the highest preference for the VIP to its home node + # default weight when created is/was 100. + # on Fedora setting appears to be additive, so to get the desired + # value we adjust the weight + # weight=$(expr ${weight} - 100) + pcs -f ${cibfile} constraint location ${primary}-cluster_ip-1 prefers ${primary}=${weight} + if [ $? -ne 0 ]; then + logger "warning: pcs constraint location ${primary}-cluster_ip-1 prefers ${primary}=${weight} failed" + fi +} + +wrap_create_virt_ip_constraints() +{ + local cibfile=${1}; shift + local primary=${1}; shift + local head="" + local tail="" + + # build a list of peers, e.g. for a four node cluster, for node1, + # the result is "node2 node3 node4"; for node2, "node3 node4 node1" + # and so on. + while [[ ${1} ]]; do + if [ "${1}" = "${primary}" ]; then + shift + while [[ ${1} ]]; do + tail=${tail}" "${1} + shift + done + else + head=${head}" "${1} + fi + shift + done + do_create_virt_ip_constraints ${cibfile} ${primary} ${tail} ${head} +} + +create_virt_ip_constraints() +{ + local cibfile=${1}; shift + while [[ ${1} ]]; do + wrap_create_virt_ip_constraints ${cibfile} ${1} ${HA_SERVERS} + shift + done +} + + +setup_create_resources() +{ + local cibfile=$(mktemp -u) + + # mount the HA-state volume and start ganesha.nfsd on all nodes + pcs resource create nfs_start ganesha_nfsd ha_vol_name=${HA_VOL_NAME} ha_vol_mnt=${HA_VOL_MNT} ha_vol_server=${HA_VOL_SERVER} --clone + if [ $? -ne 0 ]; then + logger "warning: pcs resource create nfs_start ganesha_nfsd --clone failed" + fi + sleep 1 + # cloned resources seem to never have their start() invoked when they + # are created, but stop() is invoked when they are destroyed. Why???. + # No matter, we don't want this resource agent hanging around anyway + pcs resource delete nfs_start-clone + if [ $? -ne 0 ]; then + logger "warning: pcs resource delete nfs_start-clone failed" + fi + + pcs resource create nfs-mon ganesha_mon --clone + if [ $? -ne 0 ]; then + logger "warning: pcs resource create nfs-mon ganesha_mon --clone failed" + fi + + pcs resource create nfs-grace ganesha_grace --clone + if [ $? -ne 0 ]; then + logger "warning: pcs resource create nfs-grace ganesha_grace --clone failed" + fi + + pcs cluster cib ${cibfile} + + while [[ ${1} ]]; do + + # ipaddr=$(grep ^${1} ${HA_CONFIG_FILE} | cut -d = -f 2) + ipaddrx="VIP_${1//-/_}" + + ipaddr=${!ipaddrx} + + pcs -f ${cibfile} resource create ${1}-cluster_ip-1 ocf:heartbeat:IPaddr ip=${ipaddr} cidr_netmask=32 op monitor interval=15s + if [ $? -ne 0 ]; then + logger "warning pcs resource create ${1}-cluster_ip-1 ocf:heartbeat:IPaddr ip=${ipaddr} cidr_netmask=32 op monitor interval=10s failed" + fi + + pcs -f ${cibfile} resource create ${1}-trigger_ip-1 ocf:heartbeat:Dummy + if [ $? -ne 0 ]; then + logger "warning: pcs resource create ${1}-trigger_ip-1 ocf:heartbeat:Dummy failed" + fi + + pcs -f ${cibfile} constraint colocation add ${1}-cluster_ip-1 with ${1}-trigger_ip-1 + if [ $? -ne 0 ]; then + logger "warning: pcs constraint colocation add ${1}-cluster_ip-1 with ${1}-trigger_ip-1 failed" + fi + + pcs -f ${cibfile} constraint order ${1}-trigger_ip-1 then nfs-grace-clone + if [ $? -ne 0 ]; then + logger "warning: pcs constraint order ${1}-trigger_ip-1 then nfs-grace-clone failed" + fi + + pcs -f ${cibfile} constraint order nfs-grace-clone then ${1}-cluster_ip-1 + if [ $? -ne 0 ]; then + logger "warning: pcs constraint order nfs-grace-clone then ${1}-cluster_ip-1 failed" + fi + + shift + done + + create_virt_ip_constraints ${cibfile} ${HA_SERVERS} + + pcs cluster cib-push ${cibfile} + if [ $? -ne 0 ]; then + logger "warning pcs cluster cib-push ${cibfile} failed" + fi + rm -f ${cibfile} +} + +teardown_resources() +{ + # local mntpt=$(grep ha-vol-mnt ${HA_CONFIG_FILE} | cut -d = -f 2) + + # unmount the HA-state volume and terminate ganesha.nfsd on all nodes + pcs resource create nfs_stop ganesha_nfsd ha_vol_name=dummy ha_vol_mnt=${HA_VOL_MNT} ha_vol_server=dummy --clone + if [ $? -ne 0 ]; then + logger "warning: pcs resource create nfs_stop ganesha_nfsd --clone failed" + fi + sleep 1 + # cloned resources seem to never have their start() invoked when they + # are created, but stop() is invoked when they are destroyed. Why???. + pcs resource delete nfs_stop-clone + if [ $? -ne 0 ]; then + logger "warning: pcs resource delete nfs_stop-clone failed" + fi + + while [[ ${1} ]]; do + pcs resource delete ${1}-cluster_ip-1 + if [ $? -ne 0 ]; then + logger "warning: pcs resource delete ${1}-cluster_ip-1 failed" + fi + pcs resource delete ${1}-trigger_ip-1 + if [ $? -ne 0 ]; then + logger "warning: pcs resource delete ${1}-trigger_ip-1 failed" + fi + pcs resource delete ${1}-dead_ip-1 + if [ $? -ne 0 ]; then + logger "info: pcs resource delete ${1}-dead_ip-1 failed" + fi + shift + done + + # delete -clone resource agents + pcs resource delete nfs-mon-clone + if [ $? -ne 0 ]; then + logger "warning: pcs resource delete nfs-mon-clone failed" + fi + + pcs resource delete nfs-grace-clone + if [ $? -ne 0 ]; then + logger "warning: pcs resource delete nfs-grace-clone failed" + fi + +} + + +recreate_resources() +{ + local cibfile=${1}; shift + local add_node=${1}; shift + local add_vip=${1}; shift + + while [[ ${1} ]]; do + + # ipaddr=$(grep ^${1} ${HA_CONFIG_FILE} | cut -d = -f 2) + ipaddrx="VIP_${1//-/_}" + + ipaddr=${!ipaddrx} + + pcs -f ${cibfile} resource create ${1}-cluster_ip-1 ocf:heartbeat:IPaddr ip=${ipaddr} cidr_netmask=32 op monitor interval=15s + if [ $? -ne 0 ]; then + logger "warning pcs resource create ${1}-cluster_ip-1 ocf:heartbeat:IPaddr ip=${ipaddr} cidr_netmask=32 op monitor interval=10s failed" + fi + + pcs -f ${cibfile} resource create ${1}-trigger_ip-1 ocf:heartbeat:Dummy + if [ $? -ne 0 ]; then + logger "warning: pcs resource create ${1}-trigger_ip-1 ocf:heartbeat:Dummy failed" + fi + + pcs -f ${cibfile} constraint colocation add ${1}-cluster_ip-1 with ${1}-trigger_ip-1 + if [ $? -ne 0 ]; then + logger "warning: pcs constraint colocation add ${1}-cluster_ip-1 with ${1}-trigger_ip-1 failed" + fi + + pcs -f ${cibfile} constraint order ${1}-trigger_ip-1 then nfs-grace-clone + if [ $? -ne 0 ]; then + logger "warning: pcs constraint order ${1}-trigger_ip-1 then nfs-grace-clone failed" + fi + + pcs -f ${cibfile} constraint order nfs-grace-clone then ${1}-cluster_ip-1 + if [ $? -ne 0 ]; then + logger "warning: pcs constraint order nfs-grace-clone then ${1}-cluster_ip-1 failed" + fi + + shift + done + + pcs -f ${cibfile} resource create ${add_node}-cluster_ip-1 ocf:heartbeat:IPaddr ip=${add_vip} cidr_netmask=32 op monitor interval=15s + if [ $? -ne 0 ]; then + logger "warning pcs resource create ${add_node}-cluster_ip-1 ocf:heartbeat:IPaddr ip=${add_vip} cidr_netmask=32 op monitor interval=10s failed" + fi + + pcs -f ${cibfile} resource create ${add_node}-trigger_ip-1 ocf:heartbeat:Dummy + if [ $? -ne 0 ]; then + logger "warning: pcs resource create ${add_node}-trigger_ip-1 ocf:heartbeat:Dummy failed" + fi + + pcs -f ${cibfile} constraint colocation add ${add_node}-cluster_ip-1 with ${add_node}-trigger_ip-1 + if [ $? -ne 0 ]; then + logger "warning: pcs constraint colocation add ${add_node}-cluster_ip-1 with ${add_node}-trigger_ip-1 failed" + fi + + pcs -f ${cibfile} constraint order ${add_node}-trigger_ip-1 then nfs-grace-clone + if [ $? -ne 0 ]; then + logger "warning: pcs constraint order ${add_node}-trigger_ip-1 then nfs-grace-clone failed" + fi + + pcs -f ${cibfile} constraint order nfs-grace-clone then ${add_node}-cluster_ip-1 + if [ $? -ne 0 ]; then + logger "warning: pcs constraint order nfs-grace-clone then ${add_node}-cluster_ip-1 failed" + fi + +} + + +clear_and_recreate_resources() +{ + local cibfile=${1}; shift + local add_node=${1}; shift + local add_vip=${1}; shift + + while [[ ${1} ]]; do + + pcs -f ${cibfile} resource delete ${1}-cluster_ip-1 + if [ $? -ne 0 ]; then + logger "warning: pcs -f ${cibfile} resource delete ${1}-cluster_ip-1" + fi + + pcs -f ${cibfile} resource delete ${1}-trigger_ip-1 + if [ $? -ne 0 ]; then + logger "warning: pcs -f ${cibfile} resource delete ${1}-trigger_ip-1" + fi + + shift + done + + recreate_resources ${cibfile} ${add_node} ${add_vip} ${HA_SERVERS} + +} + + +addnode_create_resources() +{ + local add_node=${1}; shift + local add_vip=${1}; shift + local cibfile=$(mktemp -u) + + # mount the HA-state volume and start ganesha.nfsd on the new node + pcs cluster cib ${cibfile} + if [ $? -ne 0 ]; then + logger "warning: pcs cluster cib ${cibfile} failed" + fi + + pcs -f ${cibfile} resource create nfs_start-${add_node} ganesha_nfsd ha_vol_name=${HA_VOL_NAME} ha_vol_mnt=${HA_VOL_MNT} ha_vol_server=${HA_VOL_SERVER} + if [ $? -ne 0 ]; then + logger "warning: pcs -f ${cibfile} resource create nfs_start-${add_node} ganesha_nfsd ha_vol_name=${HA_VOL_NAME} ha_vol_mnt=${HA_VOL_MNT} ha_vol_server=${HA_VOL_SERVER} failed" + fi + + pcs -f ${cibfile} constraint location nfs_start-${add_node} prefers ${newnode}=INFINITY + if [ $? -ne 0 ]; then + logger "warning: pcs -f ${cibfile} constraint location nfs_start-${add_node} prefers ${newnode}=INFINITY failed" + fi + + pcs -f ${cibfile} constraint order nfs_start-${add_node} then nfs-mon-clone + if [ $? -ne 0 ]; then + logger "warning: pcs -f ${cibfile} constraint order nfs_start-${add_node} then nfs-mon-clone failed" + fi + + pcs cluster cib-push ${cibfile} + if [ $? -ne 0 ]; then + logger "warning: pcs cluster cib-push ${cibfile} failed" + fi + + rm -f ${cibfile} + + # start HA on the new node + pcs cluster start ${add_node} + if [ $? -ne 0 ]; then + logger "warning: pcs cluster start ${add_node} failed" + fi + + pcs resource delete nfs_start-${add_node} + if [ $? -ne 0 ]; then + logger "warning: pcs resource delete nfs_start-${add_node} failed" + fi + + + pcs cluster cib ${cibfile} + if [ $? -ne 0 ]; then + logger "warning: pcs cluster cib ${cibfile} failed" + fi + + # delete all the -cluster_ip-1 and -trigger_ip-1 resources, + # clearing their constraints, then create them again so we can + # rejigger their constraints + clear_and_recreate_resources ${cibfile} ${add_node} ${add_vip} ${HA_SERVERS} + + HA_SERVERS="${HA_SERVERS} ${add_node}" + + create_virt_ip_constraints ${cibfile} ${HA_SERVERS} + + pcs cluster cib-push ${cibfile} + if [ $? -ne 0 ]; then + logger "warning: pcs cluster cib-push ${cibfile} failed" + fi +} + + +deletenode_delete_resources() +{ + local node=${1}; shift + + pcs cluster cib ${cibfile} + if [ $? -ne 0 ]; then + logger "warning: pcs cluster cib ${cibfile} failed" + fi + + pcs cluster cib-push ${cibfile} + if [ $? -ne 0 ]; then + logger "warning: pcs cluster cib-push ${cibfile} failed" + fi +} + +setup_state_volume() +{ + local mnt=$(mktemp -d) + local longname="" + local shortname="" + local dname="" + + mount -t glusterfs ${HA_VOL_SERVER}:/${HA_VOL_NAME} ${mnt} + + longname=$(hostname) + dname=${longname#$(hostname -s)} + + while [[ ${1} ]]; do + mkdir ${mnt}/${1}${dname} + mkdir ${mnt}/${1}${dname}/nfs + mkdir ${mnt}/${1}${dname}/nfs/ganesha + mkdir ${mnt}/${1}${dname}/nfs/statd + touch ${mnt}/${1}${dname}/nfs/state + mkdir ${mnt}/${1}${dname}/nfs/ganesha/v4recov + mkdir ${mnt}/${1}${dname}/nfs/ganesha/v4old + mkdir ${mnt}/${1}${dname}/nfs/statd/sm + mkdir ${mnt}/${1}${dname}/nfs/statd/sm.bak + mkdir ${mnt}/${1}${dname}/nfs/statd/state + for server in ${HA_SERVERS} ; do + if [ ${server} != ${1}${dname} ]; then + ln -s ${mnt}/${server}/nfs/ganesha ${mnt}/${1}${dname}/nfs/ganesha/${server} + ln -s ${mnt}/${server}/nfs/statd ${mnt}/${1}${dname}/nfs/statd/${server} + fi + done + shift + done + + umount ${mnt} + rmdir ${mnt} +} + +main() +{ + local cmd=${1}; shift + HA_CONFDIR=${1}; shift + local node="" + local vip="" + + . ${HA_CONFDIR}/ganesha-ha.conf + + if [ -e /etc/os-release ]; then + RHEL6_PCS_CNAME_OPTION="" + fi + + case "${cmd}" in + + setup | --setup) + logger "setting up ${HA_NAME}" + + check_cluster_exists ${HA_NAME} + + determine_servers "setup" + + if [ "X${HA_NUM_SERVERS}X" != "X1X" ]; then + + # setup_state_volume ${HA_SERVERS} + + setup_cluster ${HA_NAME} ${HA_NUM_SERVERS} "${HA_SERVERS}" + + setup_create_resources ${HA_SERVERS} + + setup_finalize + else + + logger "insufficient servers for HA, aborting" + fi + ;; + + teardown | --teardown) + logger "tearing down ${HA_NAME}" + + determine_servers "teardown" + + teardown_resources ${HA_SERVERS} + + teardown_cluster ${HA_NAME} + ;; + + add | --add) + node=${1}; shift + vip=${1}; shift + + logger "adding ${node} with ${vip} to ${HA_NAME}" + + determine_servers "add" + + pcs cluster node add ${node} + if [ $? -ne 0 ]; then + logger "warning: pcs cluster node add ${node} failed" + fi + + addnode_create_resources ${node} ${vip} + + ;; + + delete | --delete) + node=${1}; shift + + logger "deleting ${node} from ${HA_NAME}" + + determine_servers "delete" + + deletenode_delete_resources ${node} + + pcs cluster node remove ${node} + if [ $? -ne 0 ]; then + logger "warning: pcs cluster node remove ${node} failed" + fi + + ;; + + status | --status) + ;; + + refresh-config | --refresh-config) + ;; + + *) + logger "Usage: ganesha-ha.sh setup|teardown|add|delete|status" + ;; + + esac +} + +main $* + -- cgit