#!/usr/bin/env bash # Copyright (C) 2011-2017 Red Hat, Inc. All rights reserved. # # This copyrighted material is made available to anyone wishing to use, # modify, copy, or redistribute it subject to the terms and conditions # of the GNU General Public License v.2. # # You should have received a copy of the GNU General Public License # along with this program; if not, write to the Free Software Foundation, # Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA . lib/utils test -n "$BASH" && set -euE -o pipefail run_valgrind() { # Execute script which may use $TESTNAME for creating individual # log files for each execute command exec "${VALGRIND:-valgrind}" "$@" } expect_failure() { echo "TEST EXPECT FAILURE" } check_daemon_in_builddir() { # skip if we don't have our own deamon... if test -z "${installed_testsuite+varset}"; then (which "$1" 2>/dev/null | grep "$abs_builddir" >/dev/null ) || skip "$1 is not in executed path." fi rm -f debug.log strace.log } create_corosync_conf() { COROSYNC_CONF="/etc/corosync/corosync.conf" COROSYNC_NODE=$(hostname) if test -a "$COROSYNC_CONF"; then if ! grep "created by lvm test suite" "$COROSYNC_CONF"; then rm "$COROSYNC_CONF" else mv "$COROSYNC_CONF" "$COROSYNC_CONF.prelvmtest" fi fi sed -e "s/@LOCAL_NODE@/$COROSYNC_NODE/" lib/test-corosync-conf > "$COROSYNC_CONF" echo "created new $COROSYNC_CONF" } DLM_CONF="/etc/dlm/dlm.conf" create_dlm_conf() { if test -a "$DLM_CONF"; then if ! grep "created by lvm test suite" "$DLM_CONF"; then rm "$DLM_CONF" else mv "$DLM_CONF" "$DLM_CONF.prelvmtest" fi fi cp lib/test-dlm-conf "$DLM_CONF" echo "created new $DLM_CONF" } prepare_dlm() { if pgrep dlm_controld ; then echo "Cannot run while existing dlm_controld process exists." exit 1 fi if pgrep corosync; then echo "Cannot run while existing corosync process exists." exit 1 fi create_corosync_conf create_dlm_conf systemctl start corosync sleep 1 if ! pgrep corosync; then echo "Failed to start corosync." exit 1 fi systemctl start dlm sleep 1 if ! pgrep dlm_controld; then echo "Failed to start dlm." exit 1 fi } SANLOCK_CONF="/etc/sanlock/sanlock.conf" create_sanlock_conf() { if test -a "$SANLOCK_CONF"; then if ! grep "created by lvm test suite" "$SANLOCK_CONF"; then rm "$SANLOCK_CONF" else mv "$SANLOCK_CONF" "$SANLOCK_CONF.prelvmtest" fi fi cp lib/test-sanlock-conf "$SANLOCK_CONF" echo "created new $SANLOCK_CONF" } prepare_sanlock() { if pgrep sanlock ; then echo "Cannot run while existing sanlock process exists" exit 1 fi create_sanlock_conf systemctl start sanlock if ! pgrep sanlock; then echo "Failed to start sanlock" exit 1 fi } prepare_idm() { if pgrep seagate_ilm; then echo "Cannot run while existing seagate_ilm process exists" exit 1 fi seagate_ilm -D 0 -l 0 -L 7 -E 7 -S 7 if ! pgrep seagate_ilm; then echo "Failed to start seagate_ilm" exit 1 fi } prepare_lvmlockd() { if pgrep lvmlockd ; then echo "Cannot run while existing lvmlockd process exists" exit 1 fi if test -n "$LVM_TEST_LOCK_TYPE_SANLOCK"; then # make check_lvmlockd_sanlock echo "starting lvmlockd for sanlock" lvmlockd -o 2 elif test -n "$LVM_TEST_LOCK_TYPE_DLM"; then # make check_lvmlockd_dlm echo "starting lvmlockd for dlm" lvmlockd elif test -n "$LVM_TEST_LOCK_TYPE_IDM"; then # make check_lvmlockd_idm echo "starting lvmlockd for idm" lvmlockd -g idm elif test -n "$LVM_TEST_LVMLOCKD_TEST_DLM"; then # make check_lvmlockd_test echo "starting lvmlockd --test (dlm)" lvmlockd --test -g dlm elif test -n "$LVM_TEST_LVMLOCKD_TEST_SANLOCK"; then # FIXME: add option for this combination of --test and sanlock echo "starting lvmlockd --test (sanlock)" lvmlockd --test -g sanlock -o 2 elif test -n "$LVM_TEST_LVMLOCKD_TEST_IDM"; then # make check_lvmlockd_test echo "starting lvmlockd --test (idm)" lvmlockd --test -g idm else echo "not starting lvmlockd" exit 0 fi sleep 1 if ! pgrep lvmlockd; then echo "Failed to start lvmlockd" exit 1 fi } prepare_clvmd() { test "${LVM_TEST_LOCKING:-0}" -ne 3 && return # not needed if pgrep clvmd ; then skip "Cannot use fake cluster locking with real clvmd ($(pgrep clvmd)) running." fi check_daemon_in_builddir clvmd test -e "$DM_DEV_DIR/control" || dmsetup table >/dev/null # create control node # skip if singlenode is not compiled in (clvmd --help 2>&1 | grep "Available cluster managers" | grep "singlenode" >/dev/null) || \ skip "Compiled clvmd does not support singlenode for testing." # lvmconf "activation/monitoring = 1" local run_valgrind="" test "${LVM_VALGRIND_CLVMD:-0}" -eq 0 || run_valgrind="run_valgrind" rm -f "$CLVMD_PIDFILE" echo "<======== Starting CLVMD ========>" echo -n "## preparing clvmd..." # lvs is executed from clvmd - use our version LVM_LOG_FILE_EPOCH=CLVMD LVM_LOG_FILE_MAX_LINES=1000000 $run_valgrind clvmd -Isinglenode -d 1 -f & echo $! > LOCAL_CLVMD for i in {200..0} ; do test "$i" -eq 0 && die "Startup of clvmd is too slow." test -e "$CLVMD_PIDFILE" && test -e "${CLVMD_PIDFILE%/*}/lvm/clvmd.sock" && break echo -n . sleep .1 done echo ok } prepare_dmeventd() { if pgrep dmeventd ; then skip "Cannot test dmeventd with real dmeventd ($(pgrep dmeventd)) running." fi check_daemon_in_builddir dmeventd lvmconf "activation/monitoring = 1" local run_valgrind="" test "${LVM_VALGRIND_DMEVENTD:-0}" -eq 0 || run_valgrind="run_valgrind" echo -n "## preparing dmeventd..." # LVM_LOG_FILE_EPOCH=DMEVENTD $run_valgrind dmeventd -fddddl "$@" 2>&1 & LVM_LOG_FILE_EPOCH=DMEVENTD $run_valgrind dmeventd -fddddl "$@" >debug.log_DMEVENTD_out 2>&1 & echo $! > LOCAL_DMEVENTD # FIXME wait for pipe in /var/run instead for i in {200..0} ; do test "$i" -eq 0 && die "Startup of dmeventd is too slow." test -e "${DMEVENTD_PIDFILE}" && break echo -n . sleep .1 done echo ok } prepare_lvmpolld() { test -e LOCAL_LVMPOLLD || lvmconf "global/use_lvmpolld = 1" local run_valgrind="" test "${LVM_VALGRIND_LVMPOLLD:-0}" -eq 0 || run_valgrind="run_valgrind" kill_sleep_kill_ LOCAL_LVMPOLLD "${LVM_VALGRIND_LVMPOLLD:-0}" echo -n "## preparing lvmpolld..." $run_valgrind lvmpolld -f "$@" -s "$TESTDIR/lvmpolld.socket" -B "$TESTDIR/lib/lvm" -l all & echo $! > LOCAL_LVMPOLLD for i in {200..0} ; do test -e "$TESTDIR/lvmpolld.socket" && break echo -n .; sleep .1; done # wait for the socket test "$i" -gt 0 || die "Startup of lvmpolld is too slow." echo ok } lvmpolld_talk() { local use=nc if type -p socat >& /dev/null; then use=socat elif echo | not nc -U "$TESTDIR/lvmpolld.socket" ; then echo "WARNING: Neither socat nor nc -U seems to be available." 1>&2 echo "## failed to contact lvmpolld." return 1 fi if test "$use" = nc ; then nc -U "$TESTDIR/lvmpolld.socket" else socat "unix-connect:$TESTDIR/lvmpolld.socket" - fi | tee -a lvmpolld-talk.txt } lvmpolld_dump() { (echo 'request="dump"'; echo '##') | lvmpolld_talk "$@" } prepare_lvmdbusd() { local daemon rm -f debug.log_LVMDBUSD_out kill_sleep_kill_ LOCAL_LVMDBUSD 0 # FIXME: This is not correct! Daemon is auto started. echo -n "## checking lvmdbusd is NOT running..." if pgrep -f -l lvmdbusd | grep python3 || pgrep -x -l lvmdbusd ; then skip "Cannot run lvmdbusd while existing lvmdbusd process exists" fi echo ok # skip if we don't have our own lvmdbusd... echo -n "## find lvmdbusd to use..." if test -z "${installed_testsuite+varset}"; then # NOTE: this is always present - additional checks are needed: daemon="$abs_top_builddir/daemons/lvmdbusd/lvmdbusd" if test -x "$daemon" || chmod ugo+x "$daemon"; then echo "$daemon" else echo "Failed to make '$daemon' executable">&2 return 1 fi # Setup the python path so we can run export PYTHONPATH="$abs_top_builddir/daemons" else daemon=$(which lvmdbusd || :) echo "$daemon" fi test -x "$daemon" || skip "The lvmdbusd daemon is missing" which python3 >/dev/null || skip "Missing python3" python3 -c "import pyudev, dbus, gi.repository" || skip "Missing python modules" # Copy the needed file to run on the system bus if it doesn't # already exist if [ ! -f /etc/dbus-1/system.d/com.redhat.lvmdbus1.conf ]; then install -m 644 "$abs_top_builddir/scripts/com.redhat.lvmdbus1.conf" /etc/dbus-1/system.d/ fi echo "## preparing lvmdbusd..." lvmconf "global/notify_dbus = 1" "$daemon" --debug > debug.log_LVMDBUSD_out 2>&1 & local pid=$! sleep 1 echo -n "## checking lvmdbusd IS running..." comm= # TODO: Is there a better check than wait 1 second and check pid? if ! comm=$(ps -p $pid -o comm=) >/dev/null || [[ $comm != lvmdbusd ]]; then echo "Failed to start lvmdbusd daemon" return 1 fi echo "$pid" > LOCAL_LVMDBUSD echo ok } # # Temporary solution to create some occupied thin metadata # This heavily depends on thin metadata output format to stay as is. # Currently it expects 2MB thin metadata and 200MB data volume size # Argument specifies how many devices should be created. # prepare_thin_metadata() { local devices=$1 local transaction_id=${2:-0} local data_block_size=${3:-128} local nr_data_blocks=${4:-3200} local i echo '' for i in $(seq 1 "$devices") do echo ' ' echo ' ' echo ' ' done echo "" } teardown_devs_prefixed() { local prefix=$1 local stray=${2:-0} local IFS=$IFS_NL local once=1 local dm rm -rf "${TESTDIR:?}/dev/$prefix*" # Send idle message to frozen raids (with hope to unfreeze them) for dm in $(dm_status | grep -E "$prefix.*raid.*frozen"); do echo "## unfreezing: dmsetup message \"${dm%:*}\"" dmsetup message "${dm%:*}" 0 "idle" & done # Resume suspended devices first for dm in $(dm_info name -S "name=~$PREFIX&&suspended=Suspended"); do test "$dm" != "No devices found" || break echo "## resuming: dmsetup resume \"$dm\"" dmsetup clear "$dm" dmsetup resume "$dm" & done wait local mounts=( $(grep "$prefix" /proc/mounts | cut -d' ' -f1) ) if test ${#mounts[@]} -gt 0; then test "$stray" -eq 0 || echo "## removing stray mounted devices containing $prefix:" "${mounts[@]}" if umount -fl "${mounts[@]}"; then udev_wait fi fi # Remove devices, start with closed (sorted by open count) # Run 'dmsetup remove' in parallel rm -f REMOVE_FAILED #local listdevs=( $(dm_info name,open --sort open,name | grep "$prefix.*:0") ) #dmsetup remove --deferred ${listdevs[@]%%:0} || touch REMOVE_FAILED # 2nd. loop is trying --force removal which can possibly 'unstuck' some bloked operations for i in 0 1; do test "$i" = 1 && test "$stray" = 0 && break # no stray device removal while :; do local sortby="name" local progress=0 # HACK: sort also by minors - so we try to close 'possibly later' created device first test "$i" = 0 || sortby="-minor" for dm in $(dm_info name,open --separator ';' --nameprefixes --unquoted --sort open,"$sortby" -S "name=~$prefix" --mangle none || true) ; do test "$dm" != "No devices found" || break 2 DM_NAME=${dm##DM_NAME=} DM_NAME=${DM_NAME%%;DM_OPEN*} DM_OPEN=${dm##*;DM_OPEN=} local force="-f" if test "$i" = 0; then if test "$once" = 1 ; then once=0 echo "## removing stray mapped devices with names beginning with $prefix: " fi test "$DM_OPEN" = 0 || break # stop loop with 1st. opened device force="" fi # Succesfull 'remove' signals progress dmsetup remove $force "$DM_NAME" --mangle none && progress=1 done test "$i" = 0 || break test "$progress" = 1 || break udev_wait wait done # looping till there are some removed devices done } teardown_devs() { # Delete any remaining dm/udev semaphores teardown_udev_cookies test ! -f MD_DEV || cleanup_md_dev test ! -f DEVICES || teardown_devs_prefixed "$PREFIX" test ! -f RAMDISK || { modprobe -r brd || true ; } # NOTE: SCSI_DEBUG_DEV test must come before the LOOP test because # prepare_scsi_debug_dev() also sets LOOP to short-circuit prepare_loop() if test -f SCSI_DEBUG_DEV; then udev_wait test "${LVM_TEST_PARALLEL:-0}" -eq 1 || modprobe -r scsi_debug else test ! -f LOOP || losetup -d "$(< LOOP)" || true test ! -f LOOPFILE || rm -f "$(< LOOPFILE)" fi not diff LOOP BACKING_DEV >/dev/null 2>&1 || rm -f BACKING_DEV rm -f DEVICES LOOP RAMDISK # Attempt to remove any loop devices that failed to get torn down if earlier tests aborted test "${LVM_TEST_PARALLEL:-0}" -eq 1 || test -z "$COMMON_PREFIX" || { local stray_loops=( $(losetup -a | grep "$COMMON_PREFIX" | cut -d: -f1) ) test ${#stray_loops[@]} -eq 0 || { teardown_devs_prefixed "$COMMON_PREFIX" 1 echo "## removing stray loop devices containing $COMMON_PREFIX:" "${stray_loops[@]}" for i in "${stray_loops[@]}" ; do test ! -b "$i" || losetup -d "$i" || true ; done # Leave test when udev processed all removed devices udev_wait } } restore_dm_mirror } kill_sleep_kill_() { local pidfile=$1 local slow=$2 if test -s "$pidfile" ; then pid=$(< "$pidfile") rm -f "$pidfile" kill -TERM "$pid" 2>/dev/null || return 0 for i in {0..10} ; do ps "$pid" >/dev/null || return 0 if test "$slow" -eq 0 ; then sleep .2 ; else sleep 1 ; fi kill -KILL "$pid" 2>/dev/null || true done fi } print_procs_by_tag_() { (ps -o pid,args ehax | grep -we"LVM_TEST_TAG=${1:-kill_me_$PREFIX}") || true } count_processes_with_tag() { print_procs_by_tag_ | wc -l } kill_tagged_processes() { local pid local wait # read uses all vars within pipe subshell local pids=() while read -r pid wait; do if test -n "$pid" ; then echo "## killing tagged process: $pid ${wait:0:120}..." kill -TERM "$pid" 2>/dev/null || true fi pids+=( "$pid" ) done < <(print_procs_by_tag_ "$@") test ${#pids[@]} -eq 0 && return # wait if process exited and eventually -KILL wait=0 for pid in "${pids[@]}" ; do while ps "$pid" > /dev/null && test "$wait" -le 10; do sleep .2 wait=$(( wait + 1 )) done test "$wait" -le 10 || kill -KILL "$pid" 2>/dev/null || true done } teardown() { local TEST_LEAKED_DEVICES="" echo -n "## teardown..." unset LVM_LOG_FILE_EPOCH if test -f TESTNAME ; then if test ! -f SKIP_THIS_TEST ; then # Evaluate left devices only for non-skipped tests TEST_LEAKED_DEVICES=$(dmsetup table | grep "$PREFIX" | \ grep -Ev "${PREFIX}(pv|[0-9])" | \ grep -v "$(cat ERR_DEV_NAME 2>/dev/null)" | \ grep -v "$(cat ZERO_DEV_NAME 2>/dev/null)") || true fi kill_tagged_processes if test -n "$LVM_TEST_LVMLOCKD_TEST" ; then echo "" echo "## stopping lvmlockd in teardown" killall lvmlockd sleep 1 killall lvmlockd || true sleep 1 killall -9 lvmlockd || true fi dm_table | not grep -E -q "$vg|$vg1|$vg2|$vg3|$vg4" || { # Avoid activation of dmeventd if there is no pid cfg=$(test -s LOCAL_DMEVENTD || echo "--config activation{monitoring=0}") if dm_info suspended,name | grep "^Suspended:.*$PREFIX" >/dev/null ; then echo "## skipping vgremove, suspended devices detected." else vgremove -ff "$cfg" \ "$vg" "$vg1" "$vg2" "$vg3" "$vg4" &>/dev/null || rm -f debug.log strace.log fi } kill_sleep_kill_ LOCAL_LVMDBUSD 0 echo -n . kill_sleep_kill_ LOCAL_LVMPOLLD "${LVM_VALGRIND_LVMPOLLD:-0}" echo -n . kill_sleep_kill_ LOCAL_CLVMD "${LVM_VALGRIND_CLVMD:-0}" echo -n . kill_sleep_kill_ LOCAL_DMEVENTD "${LVM_VALGRIND_DMEVENTD:-0}" echo -n . test -d "$DM_DEV_DIR/mapper" && teardown_devs echo -n . fi test -z "$TEST_LEAKED_DEVICES" || { echo "## unexpected devices left dm table:" echo "$TEST_LEAKED_DEVICES" return 1 } if test "${LVM_TEST_PARALLEL:-0}" = 0 && test -z "$RUNNING_DMEVENTD"; then not pgrep dmeventd &>/dev/null # printed in STACKTRACE fi echo -n . test -n "$TESTDIR" && { cd "$TESTOLDPWD" || die "Failed to enter $TESTOLDPWD" # after this delete no further write is possible rm -rf "${TESTDIR:?}" || echo BLA } # Remove any dangling symlink in /dev/disk (our tests can confuse udev) test -d /dev/disk && { find /dev/disk -type l ! -exec /usr/bin/test -e {} \; -print0 | xargs -0 rm -f || true } # Remove any metadata archives and backups from this test on system rm -f /etc/lvm/archive/"${PREFIX}"* /etc/lvm/backup/"${PREFIX}"* echo "ok" } prepare_loop() { local size=$1 shift # all other params are directly passed to all 'losetup' calls local i local slash test -f LOOP && LOOP=$(< LOOP) echo -n "## preparing loop device..." # skip if prepare_scsi_debug_dev() was used if test -f SCSI_DEBUG_DEV -a -f LOOP ; then echo "(skipped)" return 0 fi test ! -e LOOP test -n "$DM_DEV_DIR" for i in 0 1 2 3 4 5 6 7; do test -e "$DM_DEV_DIR/loop$i" || mknod "$DM_DEV_DIR/loop$i" b 7 $i done echo -n . local LOOPFILE="$PWD/test.img" rm -f "$LOOPFILE" dd if=/dev/zero of="$LOOPFILE" bs=$((1024*1024)) count=0 seek=$(( size + 1 )) 2> /dev/null if LOOP=$(losetup "$@" -s -f "$LOOPFILE" 2>/dev/null); then : elif LOOP=$(losetup -f) && losetup "$@" "$LOOP" "$LOOPFILE"; then # no -s support : else # no -f support # Iterate through $DM_DEV_DIR/loop{,/}{0,1,2,3,4,5,6,7} for slash in '' /; do for i in 0 1 2 3 4 5 6 7; do local dev="$DM_DEV_DIR/loop$slash$i" ! losetup "$dev" >/dev/null 2>&1 || continue # got a free losetup "$@" "$dev" "$LOOPFILE" LOOP=$dev break done test -z "$LOOP" || break done fi test -n "$LOOP" # confirm or fail BACKING_DEV=$LOOP echo "$LOOP" > LOOP echo "$LOOP" > BACKING_DEV echo "ok ($LOOP)" } prepare_ramdisk() { local size=$1 echo -n "## preparing ramdisk device..." modprobe brd rd_size=$((size * 1024)) || return BACKING_DEV=/dev/ram0 echo "ok ($BACKING_DEV)" touch RAMDISK } prepare_real_devs() { aux lvmconf 'devices/scan = "/dev"' touch REAL_DEVICES if test -n "$LVM_TEST_DEVICE_LIST"; then local count=0 while read path; do REAL_DEVICES[$count]=$path count=$(( count + 1 )) aux extend_filter "a|$path|" dd if=/dev/zero of="$path" bs=32k count=1 wipefs -a "$path" 2>/dev/null || true done < "$LVM_TEST_DEVICE_LIST" fi printf "%s\\n" "${REAL_DEVICES[@]}" > REAL_DEVICES } # A drop-in replacement for prepare_loop() that uses scsi_debug to create # a ramdisk-based SCSI device upon which all LVM devices will be created # - scripts must take care not to use a DEV_SIZE that will enduce OOM-killer prepare_scsi_debug_dev() { local DEV_SIZE=$1 shift # rest of params directly passed to modprobe local DEBUG_DEV rm -f debug.log strace.log test ! -f "SCSI_DEBUG_DEV" || return 0 test ! -f LOOP test -n "$DM_DEV_DIR" # Skip test if scsi_debug module is unavailable or is already in use modprobe --dry-run scsi_debug || skip lsmod | not grep scsi_debug >/dev/null || skip # Create the scsi_debug device and determine the new scsi device's name # NOTE: it will _never_ make sense to pass num_tgts param; # last param wins.. so num_tgts=1 is imposed touch SCSI_DEBUG_DEV modprobe scsi_debug dev_size_mb="$(( DEV_SIZE + 2 ))" "$@" num_tgts=1 || skip for i in {1..20} ; do sleep .1 # allow for async Linux SCSI device registration DEBUG_DEV="/dev/$(grep -H scsi_debug /sys/block/sd*/device/model | cut -f4 -d /)" test -b "$DEBUG_DEV" && break done test -b "$DEBUG_DEV" || return 1 # should not happen # Create symlink to scsi_debug device in $DM_DEV_DIR SCSI_DEBUG_DEV="$DM_DEV_DIR/$(basename "$DEBUG_DEV")" echo "$SCSI_DEBUG_DEV" > SCSI_DEBUG_DEV echo "$SCSI_DEBUG_DEV" > BACKING_DEV # Setting $LOOP provides means for prepare_devs() override test "$DEBUG_DEV" = "$SCSI_DEBUG_DEV" || ln -snf "$DEBUG_DEV" "$SCSI_DEBUG_DEV" } cleanup_scsi_debug_dev() { teardown_devs rm -f SCSI_DEBUG_DEV LOOP } mdadm_create() { local devid local mddev which mdadm >/dev/null || skip "mdadm tool is missing!" cleanup_md_dev rm -f debug.log strace.log # try to find free MD node # using the old naming /dev/mdXXX # if we need more MD arrays test suite more likely leaked them for devid in {127..150} ; do grep -q "md${devid}" /proc/mdstat || break done test "$devid" -lt "150" || skip "Cannot find free /dev/mdXXX node!" mddev=/dev/md${devid} mdadm --create "$mddev" "$@" || { # Some older 'mdadm' version managed to open and close devices internaly # and reporting non-exclusive access on such device # let's just skip the test if this happens. # Note: It's pretty complex to get rid of consequences # the following sequence avoid leaks on f19 # TODO: maybe try here to recreate few times.... mdadm --stop "$mddev" || true udev_wait while [ "$#" -ne 0 ] ; do case "$1" in *"$PREFIX"*) mdadm --zero-superblock "$1" || true ;; esac shift done udev_wait skip "Test skipped, unreliable mdadm detected!" } for i in {10..0} ; do test -e "$mddev" && break echo "Waiting for $mddev." sleep .5 done test -b "$mddev" || skip "mdadm has not created device!" echo "$mddev" > MD_DEV # LVM/DM will see this device case "$DM_DEV_DIR" in "/dev") echo "$mddev" > MD_DEV_PV ;; *) rm -f "$DM_DEV_DIR/md${devid}" cp -LR "$mddev" "$DM_DEV_DIR" echo "${DM_DEV_DIR}/md${devid}" > MD_DEV_PV ;; esac rm -f MD_DEVICES while [ "$#" -ne 0 ] ; do case "$1" in *"$PREFIX"*) echo "$1" >> MD_DEVICES ;; esac shift done } mdadm_assemble() { STRACE= [ "$DM_DEV_DIR" = "/dev" ] && mdadm -V 2>&1 | grep " v3.2" && { # use this 'trick' to slow down mdadm which otherwise # is racing with udev rule since mdadm internally # opens and closes raid leg devices in RW mode and then # tries to get exlusive access to the leg device during # insertion to kernel and fails during assembly # There can be some other affected version of mdadm. STRACE="strace -f -o /dev/null" } $STRACE mdadm --assemble "$@" udev_wait } cleanup_md_dev() { local IFS=$IFS_NL local i local dev local base local mddev test -f MD_DEV || return 0 mddev=$(< MD_DEV) base=$(basename "$mddev") # try to find and remove any DM device on top of cleaned MD # assume /dev/mdXXX is 9:MINOR local minor=${mddev##/dev/md} for i in $(dmsetup table | grep 9:"$minor" | cut -d: -f1) ; do dmsetup remove "$i" || { dmsetup --force remove "$i" || true } done for i in {0..10} ; do grep -q "$base" /proc/mdstat || break test "$i" = 0 || { sleep .1 echo "$mddev is still present, stopping again" cat /proc/mdstat } mdadm --stop "$mddev" || true udev_wait # wait till events are process, not zeroing to early done test "$DM_DEV_DIR" = "/dev" || rm -f "$(< MD_DEV_PV)" for dev in $(< MD_DEVICES); do mdadm --zero-superblock "$dev" 2>/dev/null || true done udev_wait rm -f MD_DEV MD_DEVICES MD_DEV_PV } wipefs_a() { local dev=$1 local have_wipefs=1 shift if test -n "$LVM_TEST_DEVICES_FILE"; then lvmdevices --deldev "$dev" || true fi if test -f HAVE_WIPEFS ; then have_wipefs=$(< HAVE_WIPEFS) else wipefs -V >/dev/null 2>&1 || have_wipefs=0 echo "$have_wipefs" > HAVE_WIPEFS fi udev_wait if [ "$have_wipefs" = "1" ] ; then wipefs -a "$dev" || { echo "$dev: device in-use, retrying wipe again." sleep 1 udev_wait wipefs -a "$dev" } else dd if=/dev/zero of="$dev" bs=4096 count=8 oflag=direct >/dev/null || true mdadm --zero-superblock "$dev" 2>/dev/null || true fi if test -n "$LVM_TEST_DEVICES_FILE"; then lvmdevices --adddev "$dev" || true fi udev_wait } cleanup_idm_context() { local dev=$1 if [ -n "$LVM_TEST_LOCK_TYPE_IDM" ]; then sg_dev=`sg_map26 ${dev}` echo "Cleanup IDM context for drive ${dev} ($sg_dev)" sg_raw -v -r 512 -o /tmp/idm_tmp_data.bin $sg_dev \ 88 00 01 00 00 00 00 20 FF 01 00 00 00 01 00 00 sg_raw -v -s 512 -i /tmp/idm_tmp_data.bin $sg_dev \ 8E 00 FF 00 00 00 00 00 00 00 00 00 00 01 00 00 rm /tmp/idm_tmp_data.bin fi } prepare_backing_dev() { local size=${1=32} shift if test -n "$LVM_TEST_BACKING_DEVICE"; then IFS=',' read -r -a BACKING_DEVICE_ARRAY <<< "$LVM_TEST_BACKING_DEVICE" for d in "${BACKING_DEVICE_ARRAY[@]}"; do if test ! -b "$d"; then echo "Device $d doesn't exist!" return 1 fi done fi if test -f BACKING_DEV; then BACKING_DEV=$(< BACKING_DEV) return 0 elif test -n "$LVM_TEST_BACKING_DEVICE"; then BACKING_DEV=${BACKING_DEVICE_ARRAY[0]} echo "$BACKING_DEV" > BACKING_DEV return 0 elif test "${LVM_TEST_PREFER_BRD-1}" = "1" && \ test ! -d /sys/block/ram0 && \ kernel_at_least 4 16 0 && \ test "$size" -lt 16384; then # try to use ramdisk if possible, but for # big allocs (>16G) do not try to use ramdisk # Also we can't use BRD device prior kernel 4.16 # since they were DAX based and lvm2 often relies # in save table loading between exiting backend device # and bio-based 'error' device. # However with request based DAX brd device we get this: # device-mapper: ioctl: can't change device type after initial table load. prepare_ramdisk "$size" "$@" && return echo "(failed)" fi prepare_loop "$size" "$@" } prepare_devs() { local n=${1:-3} local devsize=${2:-34} local pvname=${3:-pv} local header_shift=1 # shift header from begin & end of device by 1MiB # sanlock requires more space for the internal sanlock lv # This could probably be lower, but what are the units? if test -n "$LVM_TEST_LOCK_TYPE_SANLOCK" ; then devsize=1024 fi touch DEVICES prepare_backing_dev $(( n * devsize + 2 * header_shift )) blkdiscard "$BACKING_DEV" 2>/dev/null || true echo -n "## preparing $n devices..." local size=$(( devsize * 2048 )) # sectors local count=0 rm -f CREATE_FAILED init_udev_transaction for i in $(seq 1 "$n"); do local name="${PREFIX}$pvname$i" local dev="$DM_DEV_DIR/mapper/$name" DEVICES[$count]=$dev count=$(( count + 1 )) # If the backing device number can meet the requirement for PV devices, # then allocate a dedicated backing device for PV; otherwise, rollback # to use single backing device for device-mapper. if [ -n "$LVM_TEST_BACKING_DEVICE" ] && [ $n -le ${#BACKING_DEVICE_ARRAY[@]} ]; then echo 0 $size linear "${BACKING_DEVICE_ARRAY[$(( count - 1 ))]}" $(( header_shift * 2048 )) > "$name.table" else echo 0 $size linear "$BACKING_DEV" $(( ( i - 1 ) * size + ( header_shift * 2048 ) )) > "$name.table" fi dmsetup create -u "TEST-$name" "$name" "$name.table" || touch CREATE_FAILED & test -f CREATE_FAILED && break; done wait finish_udev_transaction if test -f CREATE_FAILED ; then if test -z "$LVM_TEST_BACKING_DEVICE"; then echo "failed" return 1 fi LVM_TEST_BACKING_DEVICE= rm -f BACKING_DEV CREATE_FAILED prepare_devs "$@" return $? fi if [ -n "$LVM_TEST_BACKING_DEVICE" ]; then for d in "${BACKING_DEVICE_ARRAY[@]}"; do cnt=$((`blockdev --getsize64 $d` / 1024 / 1024)) cnt=$(( cnt < 1000 ? cnt : 1000 )) dd if=/dev/zero of="$d" bs=1MB count=$cnt wipefs -a "$d" 2>/dev/null || true cleanup_idm_context "$d" done fi # non-ephemeral devices need to be cleared between tests test -f LOOP -o -f RAMDISK || for d in "${DEVICES[@]}"; do # ensure disk header is always zeroed dd if=/dev/zero of="$d" bs=32k count=1 wipefs -a "$d" 2>/dev/null || true done if test -n "$LVM_TEST_DEVICES_FILE"; then mkdir -p "$TESTDIR/etc/lvm/devices" || true rm "$TESTDIR/etc/lvm/devices/system.devices" || true touch "$TESTDIR/etc/lvm/devices/system.devices" for d in "${DEVICES[@]}"; do lvmdevices --adddev "$d" || true done fi #for i in `seq 1 $n`; do # local name="${PREFIX}$pvname$i" # dmsetup info -c $name #done #for i in `seq 1 $n`; do # local name="${PREFIX}$pvname$i" # dmsetup table $name #done printf "%s\\n" "${DEVICES[@]}" > DEVICES # ( IFS=$'\n'; echo "${DEVICES[*]}" ) >DEVICES echo "ok" } common_dev_() { local tgtype=$1 local dev=$2 local name=${dev##*/} shift 2 local read_ms=${1-0} local write_ms=${2-0} case "$tgtype" in delay) test "$read_ms" -eq 0 && test "$write_ms" -eq 0 && { # zero delay is just equivalent to 'enable_dev' enable_dev "$dev" return } shift 2 ;; delayzero) shift 2 # zero delay is just equivalent to 'zero_dev' test "$read_ms" -eq 0 && test "$write_ms" -eq 0 && tgtype="zero" ;; # error|zero target does not take read_ms & write_ms only offset list esac local pos local size local type local pvdev local offset read -r pos size type pvdev offset < "$name.table" for fromlen in "${@-0:}"; do from=${fromlen%%:*} len=${fromlen##*:} if test "$len" = "$fromlen"; then # Missing the colon at the end: empty len len= fi test -n "$len" || len=$(( size - from )) diff=$(( from - pos )) if test $diff -gt 0 ; then echo "$pos $diff $type $pvdev $(( pos + offset ))" pos=$(( pos + diff )) elif test $diff -lt 0 ; then die "Position error" fi case "$tgtype" in delay) echo "$from $len delay $pvdev $(( pos + offset )) $read_ms $pvdev $(( pos + offset )) $write_ms" ;; writeerror) echo "$from $len delay $pvdev $(( pos + offset )) 0 $(cat ERR_DEV) 0 0" ;; delayzero) echo "$from $len delay $(cat ZERO_DEV) 0 $read_ms $(cat ZERO_DEV) 0 $write_ms" ;; error|zero) echo "$from $len $tgtype" ;; esac pos=$(( pos + len )) done > "$name.devtable" diff=$(( size - pos )) test "$diff" -gt 0 && echo "$pos $diff $type $pvdev $(( pos + offset ))" >>"$name.devtable" restore_from_devtable "$dev" } # Replace linear PV device with its 'delayed' version # Could be used to more deterministicaly hit some problems. # Parameters: {device path} [read delay ms] [write delay ms] [offset[:[size]]]... # Original device is restored when both delay params are 0 (or missing). # If the size is missing, the remaing portion of device is taken # i.e. delay_dev "$dev1" 0 200 256: delay_dev() { if test ! -f HAVE_DM_DELAY ; then target_at_least dm-delay 1 1 0 || return 0 touch HAVE_DM_DELAY fi common_dev_ delay "$@" } disable_dev() { local dev local silent="" local error="" local notify="" while test -n "$1"; do if test "$1" = "--silent"; then silent=1 shift elif test "$1" = "--error"; then error=1 shift else break fi done udev_wait for dev in "$@"; do maj=$(($(stat -L --printf=0x%t "$dev"))) min=$(($(stat -L --printf=0x%T "$dev"))) echo "Disabling device $dev ($maj:$min)" notify="$notify $maj:$min" if test -n "$error"; then echo 0 10000000 error | dmsetup load "$dev" dmsetup resume "$dev" else dmsetup remove -f "$dev" 2>/dev/null || true fi done } enable_dev() { local dev local silent="" if test "$1" = "--silent"; then silent=1 shift fi rm -f debug.log strace.log init_udev_transaction for dev in "$@"; do local name=${dev##*/} dmsetup create -u "TEST-$name" "$name" "$name.table" 2>/dev/null || \ dmsetup load "$name" "$name.table" # using device name (since device path does not exists yes with udev) dmsetup resume "$name" done finish_udev_transaction } # Try to remove list of DM device from table remove_dm_devs() { local remove=( "$@" ) local held local i for i in {1..50}; do held=() for d in "${remove[@]}" ; do dmsetup remove "$d" 2>/dev/null || { dmsetup info -c "$d" 2>/dev/null && { held+=( "$d" ) dmsetup status "$d" } } done test ${#held[@]} -eq 0 && { rm -f debug.log* return } remove=( "${held[@]}" ) done die "Can't remove device(s) ${held[*]}" } # Throttle down performance of kcopyd when mirroring i.e. disk image throttle_sys="/sys/module/dm_mirror/parameters/raid1_resync_throttle" throttle_dm_mirror() { # if the kernel config file is present, validate whether the kernel uses HZ_1000 # and return failure for this 'throttling' when it does NOT as without this setting # whole throttling is pointless on modern hardware local kconfig="/boot/config-$(uname -r)" if test -e "$kconfig" ; then grep -q "CONFIG_HZ_1000=y" "$kconfig" 2>/dev/null || { echo "WARNING: CONFIG_HZ_1000=y is NOT set in $kconfig -> throttling is unusable" return 1 } fi test -e "$throttle_sys" || return test -f THROTTLE || cat "$throttle_sys" > THROTTLE echo ${1-1} > "$throttle_sys" } # Restore original kcopyd throttle value and have mirroring fast again restore_dm_mirror() { test ! -f THROTTLE || { cat THROTTLE > "$throttle_sys" rm -f THROTTLE } } # Once there is $name.devtable # this is a quick way to restore to this table entry restore_from_devtable() { local dev local silent="" if test "$1" = "--silent"; then silent=1 shift fi rm -f debug.log strace.log init_udev_transaction for dev in "$@"; do local name=${dev##*/} dmsetup load "$name" "$name.devtable" if not dmsetup resume "$name" ; then dmsetup clear $name dmsetup resume $name finish_udev_transaction echo "Device $name has unusable table \"$(cat $name.devtable)\"" return 1 fi done finish_udev_transaction } # # Convert device to device with errors # Takes the list of pairs of error segment from:len # Combination with zero or delay is unsupported # Original device table is replaced with multiple lines # i.e. error_dev "$dev1" 8:32 96:8 error_dev() { common_dev_ error "$@" } # # Convert device to device with write errors but normal reads. # For this 'delay' dev is used and reroutes 'reads' back to original device # and for writes it will use extra new TEST-errordev (huge error target) # i.e. writeerror_dev "$dev1" 8:32 writeerror_dev() { local name=${PREFIX}-errordev if test ! -e ERR_DEV; then # delay target is used for error mapping if test ! -f HAVE_DM_DELAY ; then target_at_least dm-delay 1 1 0 || return 0 touch HAVE_DM_DELAY fi dmsetup create -u "TEST-$name" "$name" --table "0 4611686018427387904 error" # Take major:minor of our error device echo "$name" > ERR_DEV_NAME dmsetup info -c --noheadings -o major,minor "$name" > ERR_DEV fi common_dev_ writeerror "$@" } # # Convert device to device with sections of delayed zero read and writes. # For this 'delay' dev will use extra new TEST-zerodev (huge zero target) # and reroutes reads and writes # i.e. delayzero_dev "$dev1" 8:32 delayzero_dev() { local name=${PREFIX}-zerodev if test ! -e ZERO_DEV; then # delay target is used for error mapping if test ! -f HAVE_DM_DELAY ; then target_at_least dm-delay 1 1 0 || return 0 touch HAVE_DM_DELAY fi dmsetup create -u "TEST-$name" "$name" --table "0 4611686018427387904 zero" # Take major:minor of our error device echo "$name" > ZERO_DEV_NAME dmsetup info -c --noheadings -o major,minor "$name" > ZERO_DEV fi common_dev_ delayzero "$@" } # # Convert existing device to a device with zero segments # Takes the list of pairs of zero segment from:len # Combination with error or delay is unsupported # Original device table is replaced with multiple lines # i.e. zero_dev "$dev1" 8:32 96:8 zero_dev() { common_dev_ zero "$@" } backup_dev() { local dev for dev in "$@"; do dd if="$dev" of="${dev}.backup" bs=1024 done } restore_dev() { local dev for dev in "$@"; do test -e "${dev}.backup" || \ die "Internal error: $dev not backed up, can't restore!" dd of="$dev" if="${dev}.backup" bs=1024 done } prepare_pvs() { prepare_devs "$@" pvcreate -ff "${DEVICES[@]}" } prepare_vg() { teardown_devs prepare_devs "$@" vgcreate $SHARED -s 512K "$vg" "${DEVICES[@]}" } extend_devices() { test -z "$LVM_TEST_DEVICES_FILE" && return for dev in "$@"; do lvmdevices --adddev $dev done } extend_filter() { local filter test -n "$LVM_TEST_DEVICES_FILE" && return filter=$(grep ^devices/global_filter CONFIG_VALUES | tail -n 1) for rx in "$@"; do filter=$(echo "$filter" | sed -e "s:\\[:[ \"$rx\", :") done lvmconf "$filter" "devices/scan_lvs = 1" } extend_filter_md() { local filter test -n "$LVM_TEST_DEVICES_FILE" && return filter=$(grep ^devices/global_filter CONFIG_VALUES | tail -n 1) for rx in "$@"; do filter=$(echo "$filter" | sed -e "s:\\[:[ \"$rx\", :") done lvmconf "$filter" lvmconf "devices/scan = [ \"$DM_DEV_DIR\", \"/dev\" ]" } extend_filter_LVMTEST() { extend_filter "a|$DM_DEV_DIR/$PREFIX|" "$@" } hide_dev() { local filter if test -n "$LVM_TEST_DEVICES_FILE"; then for dev in "$@"; do lvmdevices --deldev $dev done else filter=$(grep ^devices/global_filter CONFIG_VALUES | tail -n 1) for dev in "$@"; do filter=$(echo "$filter" | sed -e "s:\\[:[ \"r|$dev|\", :") done lvmconf "$filter" fi } unhide_dev() { local filter if test -n "$LVM_TEST_DEVICES_FILE"; then for dev in "$@"; do lvmdevices -y --adddev $dev done else filter=$(grep ^devices/global_filter CONFIG_VALUES | tail -n 1) for dev in "$@"; do filter=$(echo "$filter" | sed -e "s:\"r|$dev|\", ::") done lvmconf "$filter" fi } mkdev_md5sum() { rm -f debug.log strace.log mkfs.ext2 "$DM_DEV_DIR/$1/$2" || return 1 md5sum "$DM_DEV_DIR/$1/$2" > "md5.$1-$2" } generate_config() { if test -n "$profile_name"; then config_values="PROFILE_VALUES_$profile_name" config="PROFILE_$profile_name" touch "$config_values" else config_values=CONFIG_VALUES config=CONFIG fi LVM_TEST_LOCKING=${LVM_TEST_LOCKING:-1} LVM_TEST_LVMPOLLD=${LVM_TEST_LVMPOLLD:-0} LVM_TEST_LVMLOCKD=${LVM_TEST_LVMLOCKD:-0} LVM_TEST_DEVICES_FILE=${LVM_TEST_DEVICES_FILE:-0} # FIXME:dct: This is harmful! Variables are unused here and are tested not being empty elsewhere: #LVM_TEST_LOCK_TYPE_SANLOCK=${LVM_TEST_LOCK_TYPE_SANLOCK:-0} #LVM_TEST_LOCK_TYPE_DLM=${LVM_TEST_LOCK_TYPE_DLM:-0} if test "$DM_DEV_DIR" = "/dev"; then LVM_VERIFY_UDEV=${LVM_VERIFY_UDEV:-0} else LVM_VERIFY_UDEV=${LVM_VERIFY_UDEV:-1} fi test -f "$config_values" || { cat > "$config_values" <<-EOF activation/checks = 1 activation/monitoring = 0 activation/polling_interval = 1 activation/retry_deactivation = 1 activation/snapshot_autoextend_percent = 50 activation/snapshot_autoextend_threshold = 50 activation/verify_udev_operations = $LVM_VERIFY_UDEV activation/raid_region_size = 512 allocation/wipe_signatures_when_zeroing_new_lvs = 0 allocation/vdo_slab_size_mb = 128 allocation/zero_metadata = 0 backup/archive = 0 backup/backup = 0 devices/cache_dir = "$TESTDIR/etc" devices/default_data_alignment = 1 devices/dir = "$DM_DEV_DIR" devices/md_component_detection = 0 devices/scan = "$DM_DEV_DIR" devices/sysfs_scan = 1 devices/write_cache_state = 0 devices/use_devicesfile = $LVM_TEST_DEVICES_FILE devices/filter = "a|.*|" devices/global_filter = [ "a|$DM_DEV_DIR/mapper/${PREFIX}.*pv[0-9_]*$|", "r|.*|" ] global/abort_on_internal_errors = 1 global/cache_check_executable = "$LVM_TEST_CACHE_CHECK_CMD" global/cache_dump_executable = "$LVM_TEST_CACHE_DUMP_CMD" global/cache_repair_executable = "$LVM_TEST_CACHE_REPAIR_CMD" global/detect_internal_vg_cache_corruption = 1 global/fallback_to_local_locking = 0 global/locking_type=$LVM_TEST_LOCKING global/notify_dbus = 0 global/si_unit_consistency = 1 global/thin_check_executable = "$LVM_TEST_THIN_CHECK_CMD" global/thin_dump_executable = "$LVM_TEST_THIN_DUMP_CMD" global/thin_repair_executable = "$LVM_TEST_THIN_REPAIR_CMD" global/use_lvmpolld = $LVM_TEST_LVMPOLLD global/use_lvmlockd = $LVM_TEST_LVMLOCKD log/activation = 1 log/file = "$TESTDIR/debug.log" log/indent = 1 log/level = 9 log/overwrite = 1 log/syslog = 0 log/verbose = 0 EOF # For 'rpm' builds use system installed binaries # and libraries and locking dir and some more built-in # defaults # For test suite run use binaries from builddir. test -z "${abs_top_builddir+varset}" || { cat >> "$config_values" <<-EOF dmeventd/executable = "$abs_top_builddir/test/lib/dmeventd" activation/udev_rules = 1 activation/udev_sync = 1 global/fsadm_executable = "$abs_top_builddir/test/lib/fsadm" global/library_dir = "$TESTDIR/lib" global/locking_dir = "$TESTDIR/var/lock/lvm" EOF } } # append all parameters (avoid adding empty \n) local v test $# -gt 0 && printf "%s\\n" "$@" >> "$config_values" declare -A CONF 2>/dev/null || { # Associative arrays is not available local s for s in $(cut -f1 -d/ "$config_values" | sort | uniq); do echo "$s {" local k for k in $(grep ^"$s"/ "$config_values" | cut -f1 -d= | sed -e 's, *$,,' | sort | uniq); do grep "^${k}[ \t=]" "$config_values" | tail -n 1 | sed -e "s,^$s/, ," || true done echo "}" echo done | tee "$config" | sed -e "s,^,## LVMCONF: ," return 0 } local sec local last_sec="" # read sequential list and put into associative array while IFS= read -r v; do CONF["${v%%[={ ]*}"]=${v#*/} done < "$config_values" # sort by section and iterate through them printf "%s\\n" "${!CONF[@]}" | sort | while read -r v ; do sec=${v%%/*} # split on section'/'param_name test "$sec" = "$last_sec" || { test -z "$last_sec" || echo "}" echo "$sec {" last_sec=$sec } echo " ${CONF[$v]}" done > "$config" echo "}" >> "$config" sed -e "s,^,## LVMCONF: ," "$config" } lvmconf() { local profile_name="" test $# -eq 0 || { # Compare if passed args aren't already all in generated lvm.conf local needed=0 for i in "$@"; do val=$(grep "${i%%[={ ]*}" CONFIG_VALUES 2>/dev/null | tail -1) || { needed=1; break; } test "$val" = "$i" || { needed=1; break; } done test "$needed" -eq 0 && { echo "## Skipping reconfiguring for: (" "$@" ")" return 0 # not needed } } generate_config "$@" mv -f CONFIG "$LVM_SYSTEM_DIR/lvm.conf" } profileconf() { local pdir="$LVM_SYSTEM_DIR/profile" local profile_name=$1 shift generate_config "$@" mkdir -p "$pdir" mv -f "PROFILE_$profile_name" "$pdir/$profile_name.profile" } prepare_profiles() { local pdir="$LVM_SYSTEM_DIR/profile" local profile_name mkdir -p "$pdir" for profile_name in "$@"; do test -L "lib/$profile_name.profile" || skip cp "lib/$profile_name.profile" "$pdir/$profile_name.profile" done } unittest() { test -x "$TESTOLDPWD/unit/unit-test" || skip "$TESTOLDPWD/unit/unit-test" "${@}" } mirror_recovery_works() { case "$(uname -r)" in 3.3.4-5.fc17.i686|3.3.4-5.fc17.x86_64) return 1 ;; esac } raid456_replace_works() { # The way kmem_cache aliasing is done in the kernel is broken. # It causes RAID 4/5/6 tests to fail. # # The problem with kmem_cache* is this: # *) Assume CONFIG_SLUB is set # 1) kmem_cache_create(name="foo-a") # - creates new kmem_cache structure # 2) kmem_cache_create(name="foo-b") # - If identical cache characteristics, it will be merged with the previously # created cache associated with "foo-a". The cache's refcount will be # incremented and an alias will be created via sysfs_slab_alias(). # 3) kmem_cache_destroy() # - Attempting to destroy cache associated with "foo-a", but instead the # refcount is simply decremented. I don't even think the sysfs aliases are # ever removed... # 4) kmem_cache_create(name="foo-a") # - This FAILS because kmem_cache_sanity_check colides with the existing # name ("foo-a") associated with the non-removed cache. # # This is a problem for RAID (specifically dm-raid) because the name used # for the kmem_cache_create is ("raid%d-%p", level, mddev). If the cache # persists for long enough, the memory address of an old mddev will be # reused for a new mddev - causing an identical formulation of the cache # name. Even though kmem_cache_destory had long ago been used to delete # the old cache, the merging of caches has cause the name and cache of that # old instance to be preserved and causes a colision (and thus failure) in # kmem_cache_create(). I see this regularly in testing the following # kernels: # # This seems to be finaly resolved with this patch: # http://www.redhat.com/archives/dm-devel/2014-March/msg00008.html # so we need to put here exlusion for kernes which do trace SLUB # case "$(uname -r)" in 3.6.*.fc18.i686*|3.6.*.fc18.x86_64) return 1 ;; 3.9.*.fc19.i686*|3.9.*.fc19.x86_64) return 1 ;; 3.1[0123].*.fc18.i686*|3.1[0123].*.fc18.x86_64) return 1 ;; 3.1[01234].*.fc19.i686*|3.1[01234].*.fc19.x86_64) return 1 ;; 3.1[123].*.fc20.i686*|3.1[123].*.fc20.x86_64) return 1 ;; 3.14.*.fc21.i686*|3.14.*.fc21.x86_64) return 1 ;; 3.15.*rc6*.fc21.i686*|3.15.*rc6*.fc21.x86_64) return 1 ;; 3.16.*rc4*.fc21.i686*|3.16.*rc4*.fc21.x86_64) return 1 ;; esac } # # Some 32bit kernel cannot pass some erroring magic which forces # thin-pool to be falling into Error state. # # Skip test on such kernels (see: https://bugzilla.redhat.com/1310661) # thin_pool_error_works_32() { case "$(uname -r)" in 2.6.32-618.*.i686) return 1 ;; 2.6.32-623.*.i686) return 1 ;; 2.6.32-573.1[28].1.el6.i686) return 1 ;; esac } thin_restore_needs_more_volumes() { case $("$LVM_TEST_THIN_RESTORE_CMD" -V) in # With older version of thin-tool we got slightly more compact metadata 0.[0-6]*|0.7.0*) return 0 ;; 0.8.5-2.el7) return 0 ;; esac return 1 } udev_wait() { pgrep udev >/dev/null || return 0 which udevadm &>/dev/null || return 0 if test -n "${1-}" ; then udevadm settle --exit-if-exists="$1" || true else udevadm settle --timeout=15 || true fi } # wait_for_sync wait_for_sync() { local i for i in {1..100} ; do check in_sync "$@" && return sleep .2 done echo "Sync is taking too long - assume stuck" echo t >/proc/sysrq-trigger 2>/dev/null return 1 } # Check if tests are running on 64bit architecture can_use_16T() { test "$(getconf LONG_BIT)" -eq 64 } # Check if major.minor.revision' string is 'at_least' version_at_least() { local major local minor local revision IFS=".-" read -r major minor revision <<< "$1" shift test -n "${1:-}" || return 0 test -n "$major" || return 1 test "$major" -gt "$1" && return 0 test "$major" -eq "$1" || return 1 test -n "${2:-}" || return 0 test -n "$minor" || return 1 test "$minor" -gt "$2" && return 0 test "$minor" -eq "$2" || return 1 test -n "${3:-}" || return 0 test "$revision" -ge "$3" 2>/dev/null || return 1 } # # Check wheter kernel [dm module] target exist # at least in expected version # # [dm-]target-name major minor revision # # i.e. dm_target_at_least dm-thin-pool 1 0 target_at_least() { rm -f debug.log strace.log case "$1" in dm-vdo) modprobe "kvdo" || true ;; dm-*) modprobe "$1" || true ;; esac if test "$1" = dm-raid; then case "$(uname -r)" in 3.12.0*) return 1 ;; esac fi local version version=$(dmsetup targets 2>/dev/null | grep "^${1##dm-} " 2>/dev/null) version=${version##* v} version_at_least "$version" "${@:2}" || { echo "Found $1 version $version, but requested ${*:2}." >&2 return 1 } } # Check whether the kernel driver version is greater or equal # to the specified version. This can be used to skip tests on # kernels where they are known to not be supported. # # e.g. driver_at_least 4 33 # driver_at_least() { local version version=$(dmsetup version | tail -1 2>/dev/null) version=${version##*:} version_at_least "$version" "$@" || { echo "Found driver version $version, but requested" "$@" "." >&2 return 1 } } have_thin() { lvm segtypes 2>/dev/null | grep thin$ >/dev/null || { echo "Thin is not built-in." >&2 return 1 } target_at_least dm-thin-pool "$@" declare -a CONF=() # disable thin_check if not present in system if test -n "$LVM_TEST_THIN_CHECK_CMD" && test ! -x "$LVM_TEST_THIN_CHECK_CMD"; then CONF[0]="global/thin_check_executable = \"\"" fi if test -n "$LVM_TEST_THIN_DUMP_CMD" && test ! -x "$LVM_TEST_THIN_DUMP_CMD"; then CONF[1]="global/thin_dump_executable = \"\"" fi if test -n "$LVM_TEST_THIN_REPAIR_CMD" && test ! -x "$LVM_TEST_THIN_REPAIR_CMD"; then CONF[2]="global/thin_repair_executable = \"\"" fi if test ${#CONF[@]} -ne 0 ; then echo "TEST WARNING: Reconfiguring" "${CONF[@]}" lvmconf "${CONF[@]}" fi } have_vdo() { lvm segtypes 2>/dev/null | grep 'vdo$' >/dev/null || { echo "VDO is not built-in." >&2 return 1 } target_at_least dm-vdo "$@" } have_writecache() { lvm segtypes 2>/dev/null | grep 'writecache$' >/dev/null || { echo "writecache is not built-in." >&2 return 1 } target_at_least dm-writecache "$@" } have_integrity() { lvm segtypes 2>/dev/null | grep 'integrity$' >/dev/null || { echo "integrity is not built-in." >&2 return 1 } target_at_least dm-integrity "$@" } have_raid() { target_at_least dm-raid "$@" # some kernels have broken mdraid bitmaps, don't use them! # may oops kernel, we know for sure all FC24 are currently broken # in general any 4.1, 4.2 is likely useless unless patched case "$(uname -r)" in 4.[12].*fc24*) return 1 ;; esac } have_raid4 () { local r=0 have_raid 1 8 0 && r=1 have_raid 1 9 1 && r=0 return $r } have_cache() { lvm segtypes 2>/dev/null | grep ' cache-pool$' >/dev/null || { echo "Cache is not built-in." >&2 return 1 } target_at_least dm-cache "$@" declare -a CONF=() # disable cache_check if not present in system if test -n "$LVM_TEST_CACHE_CHECK_CMD" -a ! -x "$LVM_TEST_CACHE_CHECK_CMD" ; then CONF[0]="global/cache_check_executable = \"\"" fi if test -n "$LVM_TEST_CACHE_DUMP_CMD" -a ! -x "$LVM_TEST_CACHE_DUMP_CMD" ; then CONF[1]="global/cache_dump_executable = \"\"" fi if test -n "$LVM_TEST_CACHE_REPAIR_CMD" -a ! -x "$LVM_TEST_CACHE_REPAIR_CMD" ; then CONF[2]="global/cache_repair_executable = \"\"" fi if test ${#CONF[@]} -ne 0 ; then echo "TEST WARNING: Reconfiguring" "${CONF[@]}" lvmconf "${CONF[@]}" fi } have_tool_at_least() { local version version=$("$1" -V 2>/dev/null) version=${version%%-*} shift version_at_least "$version" "$@" } # check if lvm shell is build-in (needs readline) have_readline() { echo version | lvm &>/dev/null } have_multi_core() { which nproc &>/dev/null || return 0 [ "$(nproc)" -ne 1 ] } dmsetup_wrapped() { udev_wait dmsetup "$@" } awk_parse_init_count_in_lvmpolld_dump() { printf '%s' \ \ $'BEGINFILE { x=0; answ=0 }' \ $'{' \ $'if (/.*{$/) { x++ }' \ $'else if (/.*}$/) { x-- }' \ $'else if ( x == 2 && $1 ~ "[[:space:]]*"vkey) { value=substr($2, 2); value=substr(value, 1, length(value) - 1); }' \ $'if ( x == 2 && value == vvalue && $1 ~ /[[:space:]]*init_requests_count/) { answ=$2 }' \ $'if (answ > 0) { exit 0 }' \ $'}' \ $'END { printf "%d", answ }' } check_lvmpolld_init_rq_count() { local ret ret=$(awk -v vvalue="$2" -v vkey="${3:-lvname}" -F= "$(awk_parse_init_count_in_lvmpolld_dump)" lvmpolld_dump.txt) test "$ret" -eq "$1" || { die "check_lvmpolld_init_rq_count failed. Expected $1, got $ret" } } wait_pvmove_lv_ready() { # given sleep .1 this is about 20 secs of waiting local retries=${2-100} if [ -e LOCAL_LVMPOLLD ]; then local lvid="" while : ; do test "$retries" -le 0 && die "Waiting for lvmpolld timed out" test -n "$lvid" || { # wait till wanted LV really appears lvid=$(dmsetup info --noheadings -c -o uuid "$1" 2>/dev/null || true) lvid=${lvid##LVM-} } test -z "$lvid" || { lvmpolld_dump > lvmpolld_dump.txt check_lvmpolld_init_rq_count 1 "$lvid" lvid && break; } sleep .1 retries=$((retries-1)) done else while : ; do test "$retries" -le 0 && die "Waiting for pvmove LV to get activated has timed out" dmsetup info -c -o tables_loaded "$1" >out 2>/dev/null|| true; not grep Live out >/dev/null || break sleep .1 retries=$((retries-1)) done fi # Adding settle here, to avoid remove, before processing of 'add' is finished # (masking systemd-udevd issue) udevadm settle --timeout=2 || true } # Holds device open with sleep which automatically expires after given timeout # Prints PID of running holding sleep process in background hold_device_open() { local vgname=$1 local lvname=$2 local sec=${3-20} # default 20sec sleep "$sec" < "$DM_DEV_DIR/$vgname/$lvname" >/dev/null 2>&1 & SLEEP_PID=$! # wait till device is openned for i in $(seq 1 50) ; do if test "$(dmsetup info --noheadings -c -o open "$vgname"-"$lvname")" -ne 0 ; then echo "$SLEEP_PID" return fi sleep .1 done die "$vgname-$lvname expected to be openned, but it's not!" } # return total memory size in kB units total_mem() { local a local b while IFS=":" read -r a b ; do case "$a" in MemTotal*) echo "${b%% kB}" ; break ;; esac done < /proc/meminfo } kernel_at_least() { version_at_least "$(uname -r)" "$@" } test "${LVM_TEST_AUX_TRACE-0}" = "0" || set -x test -f DEVICES && devs=$(< DEVICES) if test "$1" = "dmsetup" ; then shift dmsetup_wrapped "$@" else "$@" fi