sonic-buildimage/files/scripts/syncd.sh
shlomibitton 2a9aa0836c
[202012] [Mellanox] [pmon] Fix for PMON service not starting when restarting SWSS service after fast/warm reboot (#10902)
- Why I did it
Recent change to delay PMON service in case of fast/warm reboot introduce an issue when restarting only SWSS service after fast/warm reboot for Nvidia platform.
Since the timer is triggered only when the system boot, in a scenario when the system is after a fast/warm reboot and the user restart SWSS service, as part of syncd.sh script, PMON service will stop but the timer will not start again.

- How I did it
On syncd.sh script, in case of fast/warm indication, check if pmon.timer is running.
If it is running it means we are at the first boot and continue normally.
If it is not running, meaning the service was restarted, start the timer to keep the system behavior consistent.

- How to verify it
Run fast/warm reboot.
service swss restart.
Observe PMON service starting.
2022-06-08 09:46:54 +03:00

135 lines
3.9 KiB
Bash
Executable File

#!/bin/bash
. /usr/local/bin/syncd_common.sh
function startplatform() {
# platform specific tasks
# start mellanox drivers regardless of
# boot type
if [[ x"$sonic_asic_platform" == x"mellanox" ]]; then
BOOT_TYPE=`getBootType`
if [[ x"$WARM_BOOT" == x"true" || x"$BOOT_TYPE" == x"fast" ]]; then
export FAST_BOOT=1
fi
if [[ x"$WARM_BOOT" != x"true" ]]; then
if [[ x"$(/bin/systemctl is-active pmon)" == x"active" ]]; then
/bin/systemctl stop pmon
debug "pmon is active while syncd starting, stop it first"
fi
fi
debug "Starting Firmware update procedure"
/usr/bin/mst start --with_i2cdev
/usr/bin/mlnx-fw-upgrade.sh
/etc/init.d/sxdkernel start
debug "Firmware update procedure ended"
fi
if [[ x"$WARM_BOOT" != x"true" ]]; then
if [ x$sonic_asic_platform == x'cavium' ]; then
/etc/init.d/xpnet.sh start
fi
fi
}
function waitplatform() {
BOOT_TYPE=`getBootType`
if [[ x"$sonic_asic_platform" == x"mellanox" ]]; then
if [[ x"$BOOT_TYPE" = @(x"fast"|x"warm"|x"fastfast") ]]; then
PMON_TIMER_STATUS=$(systemctl is-active pmon.timer)
if [[ x"$PMON_TIMER_STATUS" = x"inactive" ]]; then
systemctl start pmon.timer
else
debug "PMON service is delayed by a timer for better fast/warm boot performance"
fi
else
debug "Starting pmon service..."
/bin/systemctl start pmon
debug "Started pmon service"
fi
fi
if [[ x"$BOOT_TYPE" = @(x"fast"|x"warm"|x"fastfast") ]]; then
debug "LLDP service is delayed by a timer for better fast/warm boot performance"
else
debug "Starting lldp service..."
/bin/systemctl start lldp
debug "Started lldp service"
fi
}
function stopplatform1() {
if [[ x$sonic_asic_platform == x"mellanox" ]] && [[ x$TYPE == x"cold" ]]; then
debug "Stopping pmon service ahead of syncd..."
/bin/systemctl stop pmon
debug "Stopped pmon service"
fi
if [[ x$sonic_asic_platform != x"mellanox" ]] || [[ x$TYPE != x"cold" ]]; then
debug "${TYPE} shutdown syncd process ..."
/usr/bin/docker exec -i syncd$DEV /usr/bin/syncd_request_shutdown --${TYPE}
# wait until syncd quits gracefully or force syncd to exit after
# waiting for 20 seconds
start_in_secs=${SECONDS}
end_in_secs=${SECONDS}
timer_threshold=20
while docker top syncd$DEV | grep -q /usr/bin/syncd \
&& [[ $((end_in_secs - start_in_secs)) -le $timer_threshold ]]; do
sleep 0.1
end_in_secs=${SECONDS}
done
if [[ $((end_in_secs - start_in_secs)) -gt $timer_threshold ]]; then
debug "syncd process in container syncd$DEV did not exit gracefully"
fi
/usr/bin/docker exec -i syncd$DEV /bin/sync
debug "Finished ${TYPE} shutdown syncd process ..."
fi
}
function stopplatform2() {
# platform specific tasks
if [[ x"$WARM_BOOT" != x"true" ]]; then
if [ x$sonic_asic_platform == x'mellanox' ]; then
/etc/init.d/sxdkernel stop
/usr/bin/mst stop
elif [ x$sonic_asic_platform == x'cavium' ]; then
/etc/init.d/xpnet.sh stop
/etc/init.d/xpnet.sh start
fi
fi
}
OP=$1
DEV=$2
SERVICE="syncd"
PEER="swss"
DEBUGLOG="/tmp/swss-syncd-debug$DEV.log"
LOCKFILE="/tmp/swss-syncd-lock$DEV"
NAMESPACE_PREFIX="asic"
if [ "$DEV" ]; then
NET_NS="$NAMESPACE_PREFIX$DEV" #name of the network namespace
SONIC_DB_CLI="sonic-db-cli -n $NET_NS"
else
NET_NS=""
SONIC_DB_CLI="sonic-db-cli"
fi
case "$1" in
start|wait|stop)
$1
;;
*)
echo "Usage: $0 {start|wait|stop}"
exit 1
;;
esac