Examples with ZooKeeperException org.apache.solr.common.cloud.ZooKeeperException used on opensource projects

Example 1 with ZooKeeperException

use of org.apache.solr.common.cloud.ZooKeeperException in project lucene-solr by apache.

the class ZkClientClusterStateProvider method connect.

@Override
public void connect() {
    if (zkStateReader == null) {
        synchronized (this) {
            if (zkStateReader == null) {
                ZkStateReader zk = null;
                try {
                    zk = new ZkStateReader(zkHost, zkClientTimeout, zkConnectTimeout);
                    zk.createClusterStateWatchersAndUpdate();
                    zkStateReader = zk;
                    log.info("Cluster at {} ready", zkHost);
                } catch (InterruptedException e) {
                    zk.close();
                    Thread.currentThread().interrupt();
                    throw new ZooKeeperException(SolrException.ErrorCode.SERVER_ERROR, "", e);
                } catch (KeeperException e) {
                    zk.close();
                    throw new ZooKeeperException(SolrException.ErrorCode.SERVER_ERROR, "", e);
                } catch (Exception e) {
                    if (zk != null)
                        zk.close();
                    // do not wrap because clients may be relying on the underlying exception being thrown
                    throw e;
                }
            }
        }
    }
}

Example 2 with ZooKeeperException

use of org.apache.solr.common.cloud.ZooKeeperException in project lucene-solr by apache.

the class LeaderElector method joinElection.

/**
     * Begin participating in the election process. Gets a new sequential number
     * and begins watching the node with the sequence number before it, unless it
     * is the lowest number, in which case, initiates the leader process. If the
     * node that is watched goes down, check if we are the new lowest node, else
     * watch the next lowest numbered node.
     *
     * @return sequential node number
     */
public int joinElection(ElectionContext context, boolean replacement, boolean joinAtHead) throws KeeperException, InterruptedException, IOException {
    context.joinedElectionFired();
    final String shardsElectZkPath = context.electionPath + LeaderElector.ELECTION_NODE;
    long sessionId = zkClient.getSolrZooKeeper().getSessionId();
    String id = sessionId + "-" + context.id;
    String leaderSeqPath = null;
    boolean cont = true;
    int tries = 0;
    while (cont) {
        try {
            if (joinAtHead) {
                log.debug("Node {} trying to join election at the head", id);
                List<String> nodes = OverseerTaskProcessor.getSortedElectionNodes(zkClient, shardsElectZkPath);
                if (nodes.size() < 2) {
                    leaderSeqPath = zkClient.create(shardsElectZkPath + "/" + id + "-n_", null, CreateMode.EPHEMERAL_SEQUENTIAL, false);
                } else {
                    String firstInLine = nodes.get(1);
                    log.debug("The current head: {}", firstInLine);
                    Matcher m = LEADER_SEQ.matcher(firstInLine);
                    if (!m.matches()) {
                        throw new IllegalStateException("Could not find regex match in:" + firstInLine);
                    }
                    leaderSeqPath = shardsElectZkPath + "/" + id + "-n_" + m.group(1);
                    zkClient.create(leaderSeqPath, null, CreateMode.EPHEMERAL, false);
                }
            } else {
                leaderSeqPath = zkClient.create(shardsElectZkPath + "/" + id + "-n_", null, CreateMode.EPHEMERAL_SEQUENTIAL, false);
            }
            log.debug("Joined leadership election with path: {}", leaderSeqPath);
            context.leaderSeqPath = leaderSeqPath;
            cont = false;
        } catch (ConnectionLossException e) {
            // we don't know if we made our node or not...
            List<String> entries = zkClient.getChildren(shardsElectZkPath, null, true);
            boolean foundId = false;
            for (String entry : entries) {
                String nodeId = getNodeId(entry);
                if (id.equals(nodeId)) {
                    // we did create our node...
                    foundId = true;
                    break;
                }
            }
            if (!foundId) {
                cont = true;
                if (tries++ > 20) {
                    throw new ZooKeeperException(SolrException.ErrorCode.SERVER_ERROR, "", e);
                }
                try {
                    Thread.sleep(50);
                } catch (InterruptedException e2) {
                    Thread.currentThread().interrupt();
                }
            }
        } catch (KeeperException.NoNodeException e) {
            // be working on it, lets try again
            if (tries++ > 20) {
                context = null;
                throw new ZooKeeperException(SolrException.ErrorCode.SERVER_ERROR, "", e);
            }
            cont = true;
            try {
                Thread.sleep(50);
            } catch (InterruptedException e2) {
                Thread.currentThread().interrupt();
            }
        }
    }
    checkIfIamLeader(context, replacement);
    return getSeq(context.leaderSeqPath);
}

Example 3 with ZooKeeperException

use of org.apache.solr.common.cloud.ZooKeeperException in project lucene-solr by apache.

the class CreateCollectionCmd method getConfName.

private static void getConfName(SolrZkClient zkClient, String collection, String collectionPath, Map<String, Object> collectionProps) throws KeeperException, InterruptedException {
    // check for configName
    log.debug("Looking for collection configName");
    if (collectionProps.containsKey("configName")) {
        log.info("configName was passed as a param {}", collectionProps.get("configName"));
        return;
    }
    List<String> configNames = null;
    int retry = 1;
    int retryLimt = 6;
    for (; retry < retryLimt; retry++) {
        if (zkClient.exists(collectionPath, true)) {
            ZkNodeProps cProps = ZkNodeProps.load(zkClient.getData(collectionPath, null, null, true));
            if (cProps.containsKey(ZkController.CONFIGNAME_PROP)) {
                break;
            }
        }
        // if there is only one conf, use that
        try {
            configNames = zkClient.getChildren(ZkConfigManager.CONFIGS_ZKNODE, null, true);
        } catch (NoNodeException e) {
        // just keep trying
        }
        if (configNames != null && configNames.size() == 1) {
            // no config set named, but there is only 1 - use it
            log.info("Only one config set found in zk - using it:" + configNames.get(0));
            collectionProps.put(ZkController.CONFIGNAME_PROP, configNames.get(0));
            break;
        }
        if (configNames != null && configNames.contains(collection)) {
            log.info("Could not find explicit collection configName, but found config name matching collection name - using that set.");
            collectionProps.put(ZkController.CONFIGNAME_PROP, collection);
            break;
        }
        log.info("Could not find collection configName - pausing for 3 seconds and trying again - try: " + retry);
        Thread.sleep(3000);
    }
    if (retry == retryLimt) {
        log.error("Could not find configName for collection " + collection);
        throw new ZooKeeperException(SolrException.ErrorCode.SERVER_ERROR, "Could not find configName for collection " + collection + " found:" + configNames);
    }
}

Example 4 with ZooKeeperException

use of org.apache.solr.common.cloud.ZooKeeperException in project lucene-solr by apache.

the class ZkController method preRegister.

public void preRegister(CoreDescriptor cd) {
    String coreNodeName = getCoreNodeName(cd);
    // this also gets us our assigned shard id if it was not specified
    try {
        checkStateInZk(cd);
        CloudDescriptor cloudDesc = cd.getCloudDescriptor();
        // make sure the node name is set on the descriptor
        if (cloudDesc.getCoreNodeName() == null) {
            cloudDesc.setCoreNodeName(coreNodeName);
        }
        publish(cd, Replica.State.DOWN, false, true);
        String collectionName = cd.getCloudDescriptor().getCollectionName();
        DocCollection collection = zkStateReader.getClusterState().getCollectionOrNull(collectionName);
        log.debug(collection == null ? "Collection {} not visible yet, but flagging it so a watch is registered when it becomes visible" : "Registering watch for collection {}", collectionName);
        zkStateReader.registerCore(collectionName);
    } catch (KeeperException e) {
        log.error("", e);
        throw new ZooKeeperException(SolrException.ErrorCode.SERVER_ERROR, "", e);
    } catch (InterruptedException e) {
        Thread.currentThread().interrupt();
        log.error("", e);
        throw new ZooKeeperException(SolrException.ErrorCode.SERVER_ERROR, "", e);
    }
    if (cd.getCloudDescriptor().getShardId() == null && needsToBeAssignedShardId(cd, zkStateReader.getClusterState(), coreNodeName)) {
        doGetShardIdAndNodeNameProcess(cd);
    } else {
        // still wait till we see us in local state
        doGetShardIdAndNodeNameProcess(cd);
    }
}

Example 5 with ZooKeeperException

use of org.apache.solr.common.cloud.ZooKeeperException in project lucene-solr by apache.

the class ZkController method init.

private void init(CurrentCoreDescriptorProvider registerOnReconnect) {
    try {
        createClusterZkNodes(zkClient);
        zkStateReader.createClusterStateWatchersAndUpdate();
        this.baseURL = zkStateReader.getBaseUrlForNodeName(this.nodeName);
        checkForExistingEphemeralNode();
        // start the overseer first as following code may need it's processing
        if (!zkRunOnly) {
            overseerElector = new LeaderElector(zkClient);
            this.overseer = new Overseer(cc.getShardHandlerFactory().getShardHandler(), cc.getUpdateShardHandler(), CommonParams.CORES_HANDLER_PATH, zkStateReader, this, cloudConfig);
            ElectionContext context = new OverseerElectionContext(zkClient, overseer, getNodeName());
            overseerElector.setup(context);
            overseerElector.joinElection(context, false);
        }
        Stat stat = zkClient.exists(ZkStateReader.LIVE_NODES_ZKNODE, null, true);
        if (stat != null && stat.getNumChildren() > 0) {
            publishAndWaitForDownStates();
        }
        // Do this last to signal we're up.
        createEphemeralLiveNode();
    } catch (IOException e) {
        log.error("", e);
        throw new SolrException(SolrException.ErrorCode.SERVER_ERROR, "Can't create ZooKeeperController", e);
    } catch (InterruptedException e) {
        // Restore the interrupted status
        Thread.currentThread().interrupt();
        log.error("", e);
        throw new ZooKeeperException(SolrException.ErrorCode.SERVER_ERROR, "", e);
    } catch (KeeperException e) {
        log.error("", e);
        throw new ZooKeeperException(SolrException.ErrorCode.SERVER_ERROR, "", e);
    }
}