sonic-buildimage/files/scripts/syncd_common.sh
mprabhu-nokia 2adf4e9026 [systemd] ASIC status based service bringup on VOQ chassis (#7477)
Changes to allow starting per asic services like swss and syncd only if the platform vendor codedetects the asic is detected and notified. The systemd services ordering we want is database->database@->pmon->swss@->syncd@->teamd@->lldp@
There is also a requirement that management, telemetry, snmp dockers can start even if all asic services are not up.

Why I did it
For VOQ chassis, the fabric cards will have 1-N asics. Also, there could be multiple removable fabric cards. On the supervisor, swss and syncd containers need to be started only if the fabric-card is in Online state and respective asics are detected by the kernel. Using systemd, the dependent services can be in inactive state.

How I did it
Introduce a mechanism where all ASIC dependent service wait on its state to be published via PMON to REDIS. Once the subscription is received, the service proceeds to create respective dockers.
For fixed platforms, systemd is unchanged i.e. the service bring up and docker creation happens in the start()/ExecStartPre routine of the .sh scripts.
For VOQ chassis platform on supervisor, the service bringup skips docker creation in the start() routine, but does it in the wait()/ExecStart routine of the .sh scrips.
Management dockers are decoupled from ASIC docker creation.
2021-08-03 23:52:51 -07:00

160 lines
3.8 KiB
Bash
Executable File

#!/bin/bash
#
# common functions used by "syncd" scipts (syncd.sh, gbsyncd.sh, etc..)
# scripts using this must provide implementations of the following functions:
#
# startplatform
# waitplatform
# stopplatform1 and stopplatform2
#
# For examples of these, see gbsyncd.sh and syncd.sh.
#
. /usr/local/bin/asic_status.sh
function debug()
{
/usr/bin/logger $1
/bin/echo `date` "- $1" >> ${DEBUGLOG}
}
function lock_service_state_change()
{
debug "Locking ${LOCKFILE} from ${SERVICE}$DEV service"
exec {LOCKFD}>${LOCKFILE}
/usr/bin/flock -x ${LOCKFD}
trap "/usr/bin/flock -u ${LOCKFD}" 0 2 3 15
debug "Locked ${LOCKFILE} (${LOCKFD}) from ${SERVICE}$DEV service"
}
function unlock_service_state_change()
{
debug "Unlocking ${LOCKFILE} (${LOCKFD}) from ${SERVICE}$DEV service"
/usr/bin/flock -u ${LOCKFD}
}
function check_warm_boot()
{
SYSTEM_WARM_START=`$SONIC_DB_CLI STATE_DB hget "WARM_RESTART_ENABLE_TABLE|system" enable`
SERVICE_WARM_START=`$SONIC_DB_CLI STATE_DB hget "WARM_RESTART_ENABLE_TABLE|${SERVICE}" enable`
# SYSTEM_WARM_START could be empty, always make WARM_BOOT meaningful.
if [[ x"$SYSTEM_WARM_START" == x"true" ]] || [[ x"$SERVICE_WARM_START" == x"true" ]]; then
WARM_BOOT="true"
else
WARM_BOOT="false"
fi
}
function wait_for_database_service()
{
# Wait for redis server start before database clean
until [[ $($SONIC_DB_CLI PING | grep -c PONG) -gt 0 ]]; do
sleep 1;
done
# Wait for configDB initialization
until [[ $($SONIC_DB_CLI CONFIG_DB GET "CONFIG_DB_INITIALIZED") ]];
do sleep 1;
done
}
function getBootType()
{
# same code snippet in files/build_templates/docker_image_ctl.j2
case "$(cat /proc/cmdline)" in
*SONIC_BOOT_TYPE=warm*)
TYPE='warm'
;;
*SONIC_BOOT_TYPE=fastfast*)
TYPE='fastfast'
;;
*SONIC_BOOT_TYPE=fast*|*fast-reboot*)
# check that the key exists
if [[ $($SONIC_DB_CLI STATE_DB GET "FAST_REBOOT|system") == "1" ]]; then
TYPE='fast'
else
TYPE='cold'
fi
;;
*)
TYPE='cold'
esac
echo "${TYPE}"
}
start() {
debug "Starting ${SERVICE}$DEV service..."
lock_service_state_change
mkdir -p /host/warmboot
wait_for_database_service
check_warm_boot
debug "Warm boot flag: ${SERVICE}$DEV ${WARM_BOOT}."
if [[ x"$WARM_BOOT" == x"true" ]]; then
# Leave a mark for syncd scripts running inside docker.
touch /host/warmboot/warm-starting
else
rm -f /host/warmboot/warm-starting
fi
startplatform
# On supervisor card, skip starting asic related services here. In wait(),
# wait until the asic is detected by pmon and published via database.
if ! is_chassis_supervisor; then
# start service docker
/usr/bin/${SERVICE}.sh start $DEV
debug "Started ${SERVICE}$DEV service..."
fi
unlock_service_state_change
}
wait() {
# On supervisor card, wait for asic to be online before starting the docker.
if is_chassis_supervisor; then
check_asic_status
ASIC_STATUS=$?
# start service docker
if [[ $ASIC_STATUS == 0 ]]; then
/usr/bin/${SERVICE}.sh start $DEV
debug "Started ${SERVICE}$DEV service..."
fi
fi
waitplatform
/usr/bin/${SERVICE}.sh wait $DEV
}
stop() {
debug "Stopping ${SERVICE}$DEV service..."
lock_service_state_change
check_warm_boot
debug "Warm boot flag: ${SERVICE}$DEV ${WARM_BOOT}."
if [[ x"$WARM_BOOT" == x"true" ]]; then
TYPE=warm
else
TYPE=cold
fi
stopplatform1
/usr/bin/${SERVICE}.sh stop $DEV
debug "Stopped ${SERVICE}$DEV service..."
stopplatform2
unlock_service_state_change
}