consul/test/integration/connect/envoy/run-tests.sh

300 lines
6.3 KiB
Bash
Raw Normal View History

#!/usr/bin/env bash
set -eEuo pipefail
# DEBUG=1 enables set -x for this script so echos every command run
DEBUG=${DEBUG:-}
# ENVOY_VERSION to run each test against
ENVOY_VERSION=${ENVOY_VERSION:-"1.14.1"}
export ENVOY_VERSION
if [ ! -z "$DEBUG" ] ; then
set -x
fi
source helpers.bash
function command_error {
echo "ERR: command exited with status $1" 1>&2
echo " command: $2" 1>&2
echo " line: $3" 1>&2
echo " function: $4" 1>&2
echo " called at: $5" 1>&2
# printf '%s\n' "${FUNCNAME[@]}"
# printf '%s\n' "${BASH_SOURCE[@]}"
# printf '%s\n' "${BASH_LINENO[@]}"
}
trap 'command_error $? "${BASH_COMMAND}" "${LINENO}" "${FUNCNAME[0]:-main}" "${BASH_SOURCE[0]}:${BASH_LINENO[0]}"' ERR
function init_workdir {
local DC="$1"
if test -z "$DC"
then
DC=primary
fi
# Note, we use explicit set of dirs so we don't delete .gitignore. Also,
# don't wipe logs between runs as they are already split and we need them to
# upload as artifacts later.
rm -rf workdir/${DC}
mkdir -p workdir/${DC}/{consul,envoy,bats,statsd,data}
# Reload consul config from defaults
cp consul-base-cfg/* workdir/${DC}/consul/
# Add any overrides if there are any (no op if not)
connect: reconcile how upstream configuration works with discovery chains (#6225) * connect: reconcile how upstream configuration works with discovery chains The following upstream config fields for connect sidecars sanely integrate into discovery chain resolution: - Destination Namespace/Datacenter: Compilation occurs locally but using different default values for namespaces and datacenters. The xDS clusters that are created are named as they normally would be. - Mesh Gateway Mode (single upstream): If set this value overrides any value computed for any resolver for the entire discovery chain. The xDS clusters that are created may be named differently (see below). - Mesh Gateway Mode (whole sidecar): If set this value overrides any value computed for any resolver for the entire discovery chain. If this is specifically overridden for a single upstream this value is ignored in that case. The xDS clusters that are created may be named differently (see below). - Protocol (in opaque config): If set this value overrides the value computed when evaluating the entire discovery chain. If the normal chain would be TCP or if this override is set to TCP then the result is that we explicitly disable L7 Routing and Splitting. The xDS clusters that are created may be named differently (see below). - Connect Timeout (in opaque config): If set this value overrides the value for any resolver in the entire discovery chain. The xDS clusters that are created may be named differently (see below). If any of the above overrides affect the actual result of compiling the discovery chain (i.e. "tcp" becomes "grpc" instead of being a no-op override to "tcp") then the relevant parameters are hashed and provided to the xDS layer as a prefix for use in naming the Clusters. This is to ensure that if one Upstream discovery chain has no overrides and tangentially needs a cluster named "api.default.XXX", and another Upstream does have overrides for "api.default.XXX" that they won't cross-pollinate against the operator's wishes. Fixes #6159
2019-08-02 03:03:34 +00:00
find ${CASE_DIR} -maxdepth 1 -name '*.hcl' -type f -exec cp -f {} workdir/${DC}/consul \;
# Copy all the test files
connect: reconcile how upstream configuration works with discovery chains (#6225) * connect: reconcile how upstream configuration works with discovery chains The following upstream config fields for connect sidecars sanely integrate into discovery chain resolution: - Destination Namespace/Datacenter: Compilation occurs locally but using different default values for namespaces and datacenters. The xDS clusters that are created are named as they normally would be. - Mesh Gateway Mode (single upstream): If set this value overrides any value computed for any resolver for the entire discovery chain. The xDS clusters that are created may be named differently (see below). - Mesh Gateway Mode (whole sidecar): If set this value overrides any value computed for any resolver for the entire discovery chain. If this is specifically overridden for a single upstream this value is ignored in that case. The xDS clusters that are created may be named differently (see below). - Protocol (in opaque config): If set this value overrides the value computed when evaluating the entire discovery chain. If the normal chain would be TCP or if this override is set to TCP then the result is that we explicitly disable L7 Routing and Splitting. The xDS clusters that are created may be named differently (see below). - Connect Timeout (in opaque config): If set this value overrides the value for any resolver in the entire discovery chain. The xDS clusters that are created may be named differently (see below). If any of the above overrides affect the actual result of compiling the discovery chain (i.e. "tcp" becomes "grpc" instead of being a no-op override to "tcp") then the relevant parameters are hashed and provided to the xDS layer as a prefix for use in naming the Clusters. This is to ensure that if one Upstream discovery chain has no overrides and tangentially needs a cluster named "api.default.XXX", and another Upstream does have overrides for "api.default.XXX" that they won't cross-pollinate against the operator's wishes. Fixes #6159
2019-08-02 03:03:34 +00:00
find ${CASE_DIR} -maxdepth 1 -name '*.bats' -type f -exec cp -f {} workdir/${DC}/bats \;
# Copy DC specific bats
cp helpers.bash workdir/${DC}/bats
# Add any DC overrides
if test -d "${CASE_DIR}/${DC}"
then
find ${CASE_DIR}/${DC} -type f -name '*.hcl' -exec cp -f {} workdir/${DC}/consul \;
find ${CASE_DIR}/${DC} -type f -name '*.bats' -exec cp -f {} workdir/${DC}/bats \;
fi
if test -d "${CASE_DIR}/data"
then
cp -r ${CASE_DIR}/data/* workdir/${DC}/data
fi
return 0
}
function start_consul {
local DC=${1:-primary}
# Start consul now as setup script needs it up
docker-compose rm -s -v -f consul-${DC} || true
docker-compose up -d consul-${DC}
}
function pre_service_setup {
local DC=${1:-primary}
# Run test case setup (e.g. generating Envoy bootstrap, starting containers)
if [ -f "${CASE_DIR}/${DC}/setup.sh" ]
then
source ${CASE_DIR}/${DC}/setup.sh
else
source ${CASE_DIR}/setup.sh
fi
}
function start_services {
# Push the state to the shared docker volume (note this is because CircleCI
# can't use shared volumes)
docker cp workdir/. envoy_workdir_1:/workdir
# Start containers required
if [ ! -z "$REQUIRED_SERVICES" ] ; then
docker-compose rm -s -v -f $REQUIRED_SERVICES || true
docker-compose up --build -d $REQUIRED_SERVICES
fi
return 0
}
function verify {
local DC=$1
if test -z "$DC"; then
DC=primary
fi
# Execute tests
res=0
# Nuke any previous case's verify container.
docker-compose rm -s -v -f verify-${DC} || true
if docker-compose up --abort-on-container-exit --exit-code-from verify-${DC} verify-${DC} ; then
echogreen "✓ PASS"
else
echored " FAIL"
res=1
fi
return $res
}
function capture_logs {
echo "Capturing Logs"
mkdir -p "$LOG_DIR"
services="$REQUIRED_SERVICES consul-primary"
if is_set $REQUIRE_SECONDARY
then
services="$services consul-secondary"
fi
if [ -f "${CASE_DIR}/capture.sh" ]
then
echo "Executing ${CASE_DIR}/capture.sh"
source ${CASE_DIR}/capture.sh || true
fi
for cont in $services
do
echo "Capturing log for $cont"
docker-compose logs --no-color "$cont" 2>&1 > "${LOG_DIR}/${cont}.log"
done
}
function stop_services {
# Teardown
if [ -f "${CASE_DIR}/teardown.sh" ] ; then
source "${CASE_DIR}/teardown.sh"
fi
docker-compose rm -s -v -f $REQUIRED_SERVICES || true
}
function init_vars {
source "defaults.sh"
if [ -f "${CASE_DIR}/vars.sh" ] ; then
source "${CASE_DIR}/vars.sh"
fi
}
wan federation via mesh gateways (#6884) This is like a Möbius strip of code due to the fact that low-level components (serf/memberlist) are connected to high-level components (the catalog and mesh-gateways) in a twisty maze of references which make it hard to dive into. With that in mind here's a high level summary of what you'll find in the patch: There are several distinct chunks of code that are affected: * new flags and config options for the server * retry join WAN is slightly different * retry join code is shared to discover primary mesh gateways from secondary datacenters * because retry join logic runs in the *agent* and the results of that operation for primary mesh gateways are needed in the *server* there are some methods like `RefreshPrimaryGatewayFallbackAddresses` that must occur at multiple layers of abstraction just to pass the data down to the right layer. * new cache type `FederationStateListMeshGatewaysName` for use in `proxycfg/xds` layers * the function signature for RPC dialing picked up a new required field (the node name of the destination) * several new RPCs for manipulating a FederationState object: `FederationState:{Apply,Get,List,ListMeshGateways}` * 3 read-only internal APIs for debugging use to invoke those RPCs from curl * raft and fsm changes to persist these FederationStates * replication for FederationStates as they are canonically stored in the Primary and replicated to the Secondaries. * a special derivative of anti-entropy that runs in secondaries to snapshot their local mesh gateway `CheckServiceNodes` and sync them into their upstream FederationState in the primary (this works in conjunction with the replication to distribute addresses for all mesh gateways in all DCs to all other DCs) * a "gateway locator" convenience object to make use of this data to choose the addresses of gateways to use for any given RPC or gossip operation to a remote DC. This gets data from the "retry join" logic in the agent and also directly calls into the FSM. * RPC (`:8300`) on the server sniffs the first byte of a new connection to determine if it's actually doing native TLS. If so it checks the ALPN header for protocol determination (just like how the existing system uses the type-byte marker). * 2 new kinds of protocols are exclusively decoded via this native TLS mechanism: one for ferrying "packet" operations (udp-like) from the gossip layer and one for "stream" operations (tcp-like). The packet operations re-use sockets (using length-prefixing) to cut down on TLS re-negotiation overhead. * the server instances specially wrap the `memberlist.NetTransport` when running with gateway federation enabled (in a `wanfed.Transport`). The general gist is that if it tries to dial a node in the SAME datacenter (deduced by looking at the suffix of the node name) there is no change. If dialing a DIFFERENT datacenter it is wrapped up in a TLS+ALPN blob and sent through some mesh gateways to eventually end up in a server's :8300 port. * a new flag when launching a mesh gateway via `consul connect envoy` to indicate that the servers are to be exposed. This sets a special service meta when registering the gateway into the catalog. * `proxycfg/xds` notice this metadata blob to activate additional watches for the FederationState objects as well as the location of all of the consul servers in that datacenter. * `xds:` if the extra metadata is in place additional clusters are defined in a DC to bulk sink all traffic to another DC's gateways. For the current datacenter we listen on a wildcard name (`server.<dc>.consul`) that load balances all servers as well as one mini-cluster per node (`<node>.server.<dc>.consul`) * the `consul tls cert create` command got a new flag (`-node`) to help create an additional SAN in certs that can be used with this flavor of federation.
2020-03-09 20:59:02 +00:00
function global_setup {
if [ -f "${CASE_DIR}/global-setup.sh" ] ; then
source "${CASE_DIR}/global-setup.sh"
wan federation via mesh gateways (#6884) This is like a Möbius strip of code due to the fact that low-level components (serf/memberlist) are connected to high-level components (the catalog and mesh-gateways) in a twisty maze of references which make it hard to dive into. With that in mind here's a high level summary of what you'll find in the patch: There are several distinct chunks of code that are affected: * new flags and config options for the server * retry join WAN is slightly different * retry join code is shared to discover primary mesh gateways from secondary datacenters * because retry join logic runs in the *agent* and the results of that operation for primary mesh gateways are needed in the *server* there are some methods like `RefreshPrimaryGatewayFallbackAddresses` that must occur at multiple layers of abstraction just to pass the data down to the right layer. * new cache type `FederationStateListMeshGatewaysName` for use in `proxycfg/xds` layers * the function signature for RPC dialing picked up a new required field (the node name of the destination) * several new RPCs for manipulating a FederationState object: `FederationState:{Apply,Get,List,ListMeshGateways}` * 3 read-only internal APIs for debugging use to invoke those RPCs from curl * raft and fsm changes to persist these FederationStates * replication for FederationStates as they are canonically stored in the Primary and replicated to the Secondaries. * a special derivative of anti-entropy that runs in secondaries to snapshot their local mesh gateway `CheckServiceNodes` and sync them into their upstream FederationState in the primary (this works in conjunction with the replication to distribute addresses for all mesh gateways in all DCs to all other DCs) * a "gateway locator" convenience object to make use of this data to choose the addresses of gateways to use for any given RPC or gossip operation to a remote DC. This gets data from the "retry join" logic in the agent and also directly calls into the FSM. * RPC (`:8300`) on the server sniffs the first byte of a new connection to determine if it's actually doing native TLS. If so it checks the ALPN header for protocol determination (just like how the existing system uses the type-byte marker). * 2 new kinds of protocols are exclusively decoded via this native TLS mechanism: one for ferrying "packet" operations (udp-like) from the gossip layer and one for "stream" operations (tcp-like). The packet operations re-use sockets (using length-prefixing) to cut down on TLS re-negotiation overhead. * the server instances specially wrap the `memberlist.NetTransport` when running with gateway federation enabled (in a `wanfed.Transport`). The general gist is that if it tries to dial a node in the SAME datacenter (deduced by looking at the suffix of the node name) there is no change. If dialing a DIFFERENT datacenter it is wrapped up in a TLS+ALPN blob and sent through some mesh gateways to eventually end up in a server's :8300 port. * a new flag when launching a mesh gateway via `consul connect envoy` to indicate that the servers are to be exposed. This sets a special service meta when registering the gateway into the catalog. * `proxycfg/xds` notice this metadata blob to activate additional watches for the FederationState objects as well as the location of all of the consul servers in that datacenter. * `xds:` if the extra metadata is in place additional clusters are defined in a DC to bulk sink all traffic to another DC's gateways. For the current datacenter we listen on a wildcard name (`server.<dc>.consul`) that load balances all servers as well as one mini-cluster per node (`<node>.server.<dc>.consul`) * the `consul tls cert create` command got a new flag (`-node`) to help create an additional SAN in certs that can be used with this flavor of federation.
2020-03-09 20:59:02 +00:00
fi
}
function run_tests {
CASE_DIR="${CASE_DIR?CASE_DIR must be set to the path of the test case}"
CASE_NAME=$( basename $CASE_DIR | cut -c6- )
export CASE_NAME
export LOG_DIR="workdir/logs/${CASE_DIR}/${ENVOY_VERSION}"
init_vars
# Initialize the workdir
init_workdir primary
if is_set $REQUIRE_SECONDARY
then
init_workdir secondary
fi
wan federation via mesh gateways (#6884) This is like a Möbius strip of code due to the fact that low-level components (serf/memberlist) are connected to high-level components (the catalog and mesh-gateways) in a twisty maze of references which make it hard to dive into. With that in mind here's a high level summary of what you'll find in the patch: There are several distinct chunks of code that are affected: * new flags and config options for the server * retry join WAN is slightly different * retry join code is shared to discover primary mesh gateways from secondary datacenters * because retry join logic runs in the *agent* and the results of that operation for primary mesh gateways are needed in the *server* there are some methods like `RefreshPrimaryGatewayFallbackAddresses` that must occur at multiple layers of abstraction just to pass the data down to the right layer. * new cache type `FederationStateListMeshGatewaysName` for use in `proxycfg/xds` layers * the function signature for RPC dialing picked up a new required field (the node name of the destination) * several new RPCs for manipulating a FederationState object: `FederationState:{Apply,Get,List,ListMeshGateways}` * 3 read-only internal APIs for debugging use to invoke those RPCs from curl * raft and fsm changes to persist these FederationStates * replication for FederationStates as they are canonically stored in the Primary and replicated to the Secondaries. * a special derivative of anti-entropy that runs in secondaries to snapshot their local mesh gateway `CheckServiceNodes` and sync them into their upstream FederationState in the primary (this works in conjunction with the replication to distribute addresses for all mesh gateways in all DCs to all other DCs) * a "gateway locator" convenience object to make use of this data to choose the addresses of gateways to use for any given RPC or gossip operation to a remote DC. This gets data from the "retry join" logic in the agent and also directly calls into the FSM. * RPC (`:8300`) on the server sniffs the first byte of a new connection to determine if it's actually doing native TLS. If so it checks the ALPN header for protocol determination (just like how the existing system uses the type-byte marker). * 2 new kinds of protocols are exclusively decoded via this native TLS mechanism: one for ferrying "packet" operations (udp-like) from the gossip layer and one for "stream" operations (tcp-like). The packet operations re-use sockets (using length-prefixing) to cut down on TLS re-negotiation overhead. * the server instances specially wrap the `memberlist.NetTransport` when running with gateway federation enabled (in a `wanfed.Transport`). The general gist is that if it tries to dial a node in the SAME datacenter (deduced by looking at the suffix of the node name) there is no change. If dialing a DIFFERENT datacenter it is wrapped up in a TLS+ALPN blob and sent through some mesh gateways to eventually end up in a server's :8300 port. * a new flag when launching a mesh gateway via `consul connect envoy` to indicate that the servers are to be exposed. This sets a special service meta when registering the gateway into the catalog. * `proxycfg/xds` notice this metadata blob to activate additional watches for the FederationState objects as well as the location of all of the consul servers in that datacenter. * `xds:` if the extra metadata is in place additional clusters are defined in a DC to bulk sink all traffic to another DC's gateways. For the current datacenter we listen on a wildcard name (`server.<dc>.consul`) that load balances all servers as well as one mini-cluster per node (`<node>.server.<dc>.consul`) * the `consul tls cert create` command got a new flag (`-node`) to help create an additional SAN in certs that can be used with this flavor of federation.
2020-03-09 20:59:02 +00:00
global_setup
# Wipe state
docker-compose up wipe-volumes
# Push the state to the shared docker volume (note this is because CircleCI
# can't use shared volumes)
docker cp workdir/. envoy_workdir_1:/workdir
start_consul primary
if [ $? -ne 0 ]
then
capture_logs
return 1
fi
if is_set $REQUIRE_SECONDARY
then
start_consul secondary
if [ $? -ne 0 ]
then
capture_logs
return 1
fi
fi
echo "Setting up the primary datacenter"
pre_service_setup primary
if [ $? -ne 0 ]
then
echo "Setting up the primary datacenter failed"
capture_logs
return 1
fi
if is_set $REQUIRE_SECONDARY
then
echo "Setting up the secondary datacenter"
pre_service_setup secondary
if [ $? -ne 0 ]
then
echo "Setting up the secondary datacenter failed"
capture_logs
return 1
fi
fi
echo "Starting services"
start_services
if [ $? -ne 0 ]
then
capture_logs
return 1
fi
# Run the verify container and report on the output
verify primary
TESTRESULT=$?
if is_set $REQUIRE_SECONDARY && test "$TESTRESULT" -eq 0
then
verify secondary
SECONDARYRESULT=$?
if [ "$SECONDARYRESULT" -ne 0 ]
then
TESTRESULT=$SECONDARYRESULT
fi
fi
if [ "$TESTRESULT" -ne 0 ]
then
capture_logs
fi
stop_services primary
if is_set $REQUIRE_SECONDARY
then
stop_services secondary
fi
return $TESTRESULT
}
function suite_setup {
# Set a log dir to prevent docker-compose warning about unset var
export LOG_DIR="workdir/logs/"
# Cleanup from any previous unclean runs.
docker-compose down --volumes --timeout 0 --remove-orphans
# Start the volume container
docker-compose up -d workdir
}
function suite_teardown {
# Set a log dir to prevent docker-compose warning about unset var
export LOG_DIR="workdir/logs/"
docker-compose down --volumes --timeout 0 --remove-orphans
}
case "${1-}" in
"")
echo "command required"
exit 1 ;;
*)
"$@" ;;
esac