blob: da1ef42c786db03f221f66724f6107be38feafe3 [file] [log] [blame]
/*
* Licensed to the Apache Software Foundation (ASF) under one
* or more contributor license agreements. See the NOTICE file
* distributed with this work for additional information
* regarding copyright ownership. The ASF licenses this file
* to you under the Apache License, Version 2.0 (the
* "License"); you may not use this file except in compliance
* with the License. You may obtain a copy of the License at
*
* http://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing,
* software distributed under the License is distributed on an
* "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
* KIND, either express or implied. See the License for the
* specific language governing permissions and limitations
* under the License.
*/
package datareq
import (
"fmt"
"net/http"
"net/url"
"github.com/apache/trafficcontrol/lib/go-log"
"github.com/apache/trafficcontrol/lib/go-tc"
"github.com/apache/trafficcontrol/traffic_monitor/peer"
)
func srvTRState(
params url.Values,
localStates peer.CRStatesThreadsafe,
combinedStates peer.CRStatesThreadsafe,
peerStates peer.CRStatesPeersThreadsafe,
distributedPollingEnabled bool,
) ([]byte, int, error) {
_, raw := params["raw"] // peer polling case
_, local := params["local"] // distributed peer polling case
if raw {
data, err := srvTRStateSelf(localStates, distributedPollingEnabled)
return data, http.StatusOK, err
}
// This covers the case where we have lost connectivity to all peers, but multiple peers exist. In this case, it is
// more likely that the local machine has lost all connectivity than both peers losing connectivity or crashing. If
// the peers really did crash, the health protocol is essentially broken, and serving a 503 will cause Traffic Router
// to use the last good state fetched from a Traffic Monitor within the CDN. If the peers are simply unreachable from
// this Traffic Monitor, serving 503s until connectivity is restored will cause Traffic Router to ignore this instance
// until the health protocol can be relied upon once again.
if peerStates.OptimisticQuorumEnabled() {
optimisticQuorum, peersAvailable, peerCount, minimum := peerStates.HasOptimisticQuorum()
log.Debugf("optimisticQuorum=%v, peerCount=%v, peersAvailable=%v, minimum=%v", optimisticQuorum, peerCount, peersAvailable, minimum)
if !optimisticQuorum {
return nil, http.StatusServiceUnavailable, fmt.Errorf("number of peers available (%d/%d) is less than the minimum number of %d required for optimistic peer quorum", peersAvailable, peerCount, minimum)
}
}
data, err := srvTRStateDerived(combinedStates, local && distributedPollingEnabled)
return data, http.StatusOK, err
}
func srvTRStateDerived(combinedStates peer.CRStatesThreadsafe, directlyPolledOnly bool) ([]byte, error) {
if !directlyPolledOnly {
return tc.CRStatesMarshall(combinedStates.Get())
}
unfiltered := combinedStates.Get()
return tc.CRStatesMarshall(filterDirectlyPolledCaches(unfiltered))
}
func filterDirectlyPolledCaches(crstates tc.CRStates) tc.CRStates {
filtered := tc.CRStates{
Caches: make(map[tc.CacheName]tc.IsAvailable),
DeliveryService: crstates.DeliveryService,
}
for cacheName, availability := range crstates.Caches {
if availability.DirectlyPolled {
filtered.Caches[cacheName] = availability
}
}
return filtered
}
func srvTRStateSelf(localStates peer.CRStatesThreadsafe, directlyPolledOnly bool) ([]byte, error) {
if !directlyPolledOnly {
return tc.CRStatesMarshall(localStates.Get())
}
unfiltered := localStates.Get()
return tc.CRStatesMarshall(filterDirectlyPolledCaches(unfiltered))
}