Examples with TabletStats org.apache.accumulo.core.tabletserver.thrift.TabletStats used on opensource projects

Example 6 with TabletStats

use of org.apache.accumulo.core.tabletserver.thrift.TabletStats in project accumulo by apache.

the class DefaultLoadBalancer method move.

/**
 * Select a tablet based on differences between table loads; if the loads are even, use the busiest table
 */
List<TabletMigration> move(ServerCounts tooMuch, ServerCounts tooLittle, int count, Map<Table.ID, Map<KeyExtent, TabletStats>> donerTabletStats) {
    List<TabletMigration> result = new ArrayList<>();
    if (count == 0)
        return result;
    // Copy counts so we can update them as we propose migrations
    Map<Table.ID, Integer> tooMuchMap = tabletCountsPerTable(tooMuch.status);
    Map<Table.ID, Integer> tooLittleMap = tabletCountsPerTable(tooLittle.status);
    for (int i = 0; i < count; i++) {
        Table.ID table;
        Integer tooLittleCount;
        if (tableToBalance == null) {
            // find a table to migrate
            // look for an uneven table count
            int biggestDifference = 0;
            Table.ID biggestDifferenceTable = null;
            for (Entry<Table.ID, Integer> tableEntry : tooMuchMap.entrySet()) {
                Table.ID tableID = tableEntry.getKey();
                if (tooLittleMap.get(tableID) == null)
                    tooLittleMap.put(tableID, 0);
                int diff = tableEntry.getValue() - tooLittleMap.get(tableID);
                if (diff > biggestDifference) {
                    biggestDifference = diff;
                    biggestDifferenceTable = tableID;
                }
            }
            if (biggestDifference < 2) {
                table = busiest(tooMuch.status.tableMap);
            } else {
                table = biggestDifferenceTable;
            }
        } else {
            // just balance the given table
            table = tableToBalance;
        }
        Map<KeyExtent, TabletStats> onlineTabletsForTable = donerTabletStats.get(table);
        try {
            if (onlineTabletsForTable == null) {
                onlineTabletsForTable = new HashMap<>();
                List<TabletStats> stats = getOnlineTabletsForTable(tooMuch.server, table);
                if (null == stats) {
                    log.warn("Unable to find tablets to move");
                    return result;
                }
                for (TabletStats stat : stats) onlineTabletsForTable.put(new KeyExtent(stat.extent), stat);
                donerTabletStats.put(table, onlineTabletsForTable);
            }
        } catch (Exception ex) {
            log.error("Unable to select a tablet to move", ex);
            return result;
        }
        KeyExtent extent = selectTablet(tooMuch.server, onlineTabletsForTable);
        onlineTabletsForTable.remove(extent);
        if (extent == null)
            return result;
        tooMuchMap.put(table, tooMuchMap.get(table) - 1);
        /**
         * If a table grows from 1 tablet then tooLittleMap.get(table) can return a null, since there is only one tabletserver that holds all of the tablets. Here
         * we check to see if in fact that is the case and if so set the value to 0.
         */
        tooLittleCount = tooLittleMap.get(table);
        if (tooLittleCount == null) {
            tooLittleCount = 0;
        }
        tooLittleMap.put(table, tooLittleCount + 1);
        tooMuch.count--;
        tooLittle.count++;
        result.add(new TabletMigration(extent, tooMuch.server, tooLittle.server));
    }
    return result;
}

Example 7 with TabletStats

use of org.apache.accumulo.core.tabletserver.thrift.TabletStats in project accumulo by apache.

the class TabletServerResource method doCurrentOperations.

private List<CurrentOperations> doCurrentOperations(List<TabletStats> tsStats) throws Exception {
    List<CurrentOperations> currentOperations = new ArrayList<>();
    for (TabletStats info : tsStats) {
        if (info.extent == null) {
            historical = info;
            continue;
        }
        total.numEntries += info.numEntries;
        ActionStatsUpdator.update(total.minors, info.minors);
        ActionStatsUpdator.update(total.majors, info.majors);
        KeyExtent extent = new KeyExtent(info.extent);
        Table.ID tableId = extent.getTableId();
        MessageDigest digester = MessageDigest.getInstance("MD5");
        if (extent.getEndRow() != null && extent.getEndRow().getLength() > 0) {
            digester.update(extent.getEndRow().getBytes(), 0, extent.getEndRow().getLength());
        }
        String obscuredExtent = Base64.getEncoder().encodeToString(digester.digest());
        String displayExtent = String.format("[%s]", obscuredExtent);
        String tableName = Tables.getPrintableTableInfoFromId(HdfsZooInstance.getInstance(), tableId);
        currentOperations.add(new CurrentOperations(tableName, tableId, displayExtent, info.numEntries, info.ingestRate, info.queryRate, info.minors.num != 0 ? info.minors.elapsed / info.minors.num : null, stddev(info.minors.elapsed, info.minors.num, info.minors.sumDev), info.minors.elapsed != 0 ? info.minors.count / info.minors.elapsed : null, info.majors.num != 0 ? info.majors.elapsed / info.majors.num : null, stddev(info.majors.elapsed, info.majors.num, info.majors.sumDev), info.majors.elapsed != 0 ? info.majors.count / info.majors.elapsed : null));
    }
    return currentOperations;
}

Example 8 with TabletStats

use of org.apache.accumulo.core.tabletserver.thrift.TabletStats in project accumulo by apache.

the class HostRegexTableLoadBalancer method balance.

@Override
public long balance(SortedMap<TServerInstance, TabletServerStatus> current, Set<KeyExtent> migrations, List<TabletMigration> migrationsOut) {
    long minBalanceTime = 20 * 1000;
    // Iterate over the tables and balance each of them
    TableOperations t = getTableOperations();
    if (t == null)
        return minBalanceTime;
    Map<String, String> tableIdMap = t.tableIdMap();
    long now = System.currentTimeMillis();
    Map<String, SortedMap<TServerInstance, TabletServerStatus>> currentGrouped = splitCurrentByRegex(current);
    if ((now - this.lastOOBCheck) > this.oobCheckMillis) {
        try {
            // Check to see if a tablet is assigned outside the bounds of the pool. If so, migrate it.
            for (String table : t.list()) {
                LOG.debug("Checking for out of bounds tablets for table {}", table);
                String tablePoolName = getPoolNameForTable(table);
                for (Entry<TServerInstance, TabletServerStatus> e : current.entrySet()) {
                    // pool names are the same as table names, except in the DEFAULT case.
                    // If this table is assigned to a pool for this host, then move on.
                    List<String> hostPools = getPoolNamesForHost(e.getKey().host());
                    if (hostPools.contains(tablePoolName)) {
                        continue;
                    }
                    String tid = tableIdMap.get(table);
                    if (null == tid) {
                        LOG.warn("Unable to check for out of bounds tablets for table {}, it may have been deleted or renamed.", table);
                        continue;
                    }
                    try {
                        List<TabletStats> outOfBoundsTablets = getOnlineTabletsForTable(e.getKey(), Table.ID.of(tid));
                        if (null == outOfBoundsTablets) {
                            continue;
                        }
                        Random random = new Random();
                        for (TabletStats ts : outOfBoundsTablets) {
                            KeyExtent ke = new KeyExtent(ts.getExtent());
                            if (migrations.contains(ke)) {
                                LOG.debug("Migration for out of bounds tablet {} has already been requested", ke);
                                continue;
                            }
                            String poolName = getPoolNameForTable(table);
                            SortedMap<TServerInstance, TabletServerStatus> currentView = currentGrouped.get(poolName);
                            if (null != currentView) {
                                int skip = random.nextInt(currentView.size());
                                Iterator<TServerInstance> iter = currentView.keySet().iterator();
                                for (int i = 0; i < skip; i++) {
                                    iter.next();
                                }
                                TServerInstance nextTS = iter.next();
                                LOG.info("Tablet {} is currently outside the bounds of the regex, migrating from {} to {}", ke, e.getKey(), nextTS);
                                migrationsOut.add(new TabletMigration(ke, e.getKey(), nextTS));
                                if (migrationsOut.size() >= this.maxTServerMigrations) {
                                    break;
                                }
                            } else {
                                LOG.warn("No tablet servers online for pool {}, unable to migrate out of bounds tablets", poolName);
                            }
                        }
                    } catch (TException e1) {
                        LOG.error("Error in OOB check getting tablets for table {} from server {}", tid, e.getKey().host(), e);
                    }
                }
            }
        } finally {
            // this could have taken a while...get a new time
            this.lastOOBCheck = System.currentTimeMillis();
        }
    }
    if (migrationsOut.size() > 0) {
        LOG.warn("Not balancing tables due to moving {} out of bounds tablets", migrationsOut.size());
        LOG.info("Migrating out of bounds tablets: {}", migrationsOut);
        return minBalanceTime;
    }
    if (migrations != null && migrations.size() > 0) {
        if (migrations.size() >= maxOutstandingMigrations) {
            LOG.warn("Not balancing tables due to {} outstanding migrations", migrations.size());
            if (LOG.isTraceEnabled()) {
                LOG.trace("Sample up to 10 outstanding migrations: {}", Iterables.limit(migrations, 10));
            }
            return minBalanceTime;
        }
        LOG.debug("Current outstanding migrations of {} being applied", migrations.size());
        if (LOG.isTraceEnabled()) {
            LOG.trace("Sample up to 10 outstanding migrations: {}", Iterables.limit(migrations, 10));
        }
        migrationsFromLastPass.keySet().retainAll(migrations);
        SortedMap<TServerInstance, TabletServerStatus> currentCopy = new TreeMap<>(current);
        Multimap<TServerInstance, String> serverTableIdCopied = HashMultimap.create();
        for (TabletMigration migration : migrationsFromLastPass.values()) {
            TableInfo fromInfo = getTableInfo(currentCopy, serverTableIdCopied, migration.tablet.getTableId().toString(), migration.oldServer);
            if (fromInfo != null) {
                fromInfo.setOnlineTablets(fromInfo.getOnlineTablets() - 1);
            }
            TableInfo toInfo = getTableInfo(currentCopy, serverTableIdCopied, migration.tablet.getTableId().toString(), migration.newServer);
            if (toInfo != null) {
                toInfo.setOnlineTablets(toInfo.getOnlineTablets() + 1);
            }
        }
        migrations = EMPTY_MIGRATIONS;
    } else {
        migrationsFromLastPass.clear();
    }
    for (String s : tableIdMap.values()) {
        Table.ID tableId = Table.ID.of(s);
        String tableName = tableIdToTableName.get(tableId);
        String regexTableName = getPoolNameForTable(tableName);
        SortedMap<TServerInstance, TabletServerStatus> currentView = currentGrouped.get(regexTableName);
        if (null == currentView) {
            LOG.warn("Skipping balance for table {} as no tablet servers are online.", tableName);
            continue;
        }
        ArrayList<TabletMigration> newMigrations = new ArrayList<>();
        getBalancerForTable(tableId).balance(currentView, migrations, newMigrations);
        if (newMigrations.isEmpty()) {
            tableToTimeSinceNoMigrations.remove(s);
        } else if (tableToTimeSinceNoMigrations.containsKey(s)) {
            if ((now - tableToTimeSinceNoMigrations.get(s)) > ONE_HOUR) {
                LOG.warn("We have been consistently producing migrations for {}: {}", tableName, Iterables.limit(newMigrations, 10));
            }
        } else {
            tableToTimeSinceNoMigrations.put(s, now);
        }
        migrationsOut.addAll(newMigrations);
        if (migrationsOut.size() >= this.maxTServerMigrations) {
            break;
        }
    }
    for (TabletMigration migration : migrationsOut) {
        migrationsFromLastPass.put(migration.tablet, migration);
    }
    LOG.info("Migrating tablets for balance: {}", migrationsOut);
    return minBalanceTime;
}

Example 9 with TabletStats

use of org.apache.accumulo.core.tabletserver.thrift.TabletStats in project accumulo by apache.

the class TabletServerResource method getTserverDetails.

/**
 * Generates details for the selected tserver
 *
 * @param tserverAddress
 *          TServer name
 * @return TServer details
 */
@Path("{address}")
@GET
public TabletServerSummary getTserverDetails(@PathParam("address") @NotNull @Pattern(regexp = SERVER_REGEX) String tserverAddress) throws Exception {
    boolean tserverExists = false;
    for (TabletServerStatus ts : Monitor.getMmi().getTServerInfo()) {
        if (tserverAddress.equals(ts.getName())) {
            tserverExists = true;
            break;
        }
    }
    if (!tserverExists) {
        return null;
    }
    double totalElapsedForAll = 0;
    double splitStdDev = 0;
    double minorStdDev = 0;
    double minorQueueStdDev = 0;
    double majorStdDev = 0;
    double majorQueueStdDev = 0;
    double currentMinorAvg = 0;
    double currentMajorAvg = 0;
    double currentMinorStdDev = 0;
    double currentMajorStdDev = 0;
    total = new TabletStats(null, new ActionStats(), new ActionStats(), new ActionStats(), 0, 0, 0, 0);
    HostAndPort address = HostAndPort.fromString(tserverAddress);
    historical = new TabletStats(null, new ActionStats(), new ActionStats(), new ActionStats(), 0, 0, 0, 0);
    List<TabletStats> tsStats = new ArrayList<>();
    try {
        ClientContext context = Monitor.getContext();
        TabletClientService.Client client = ThriftUtil.getClient(new TabletClientService.Client.Factory(), address, context);
        try {
            for (String tableId : Monitor.getMmi().tableMap.keySet()) {
                tsStats.addAll(client.getTabletStats(Tracer.traceInfo(), context.rpcCreds(), tableId));
            }
            historical = client.getHistoricalStats(Tracer.traceInfo(), context.rpcCreds());
        } finally {
            ThriftUtil.returnClient(client);
        }
    } catch (Exception e) {
        return null;
    }
    List<CurrentOperations> currentOps = doCurrentOperations(tsStats);
    if (total.minors.num != 0)
        currentMinorAvg = (long) (total.minors.elapsed / total.minors.num);
    if (total.minors.elapsed != 0 && total.minors.num != 0)
        currentMinorStdDev = stddev(total.minors.elapsed, total.minors.num, total.minors.sumDev);
    if (total.majors.num != 0)
        currentMajorAvg = total.majors.elapsed / total.majors.num;
    if (total.majors.elapsed != 0 && total.majors.num != 0 && total.majors.elapsed > total.majors.num)
        currentMajorStdDev = stddev(total.majors.elapsed, total.majors.num, total.majors.sumDev);
    ActionStatsUpdator.update(total.minors, historical.minors);
    ActionStatsUpdator.update(total.majors, historical.majors);
    totalElapsedForAll += total.majors.elapsed + historical.splits.elapsed + total.minors.elapsed;
    minorStdDev = stddev(total.minors.elapsed, total.minors.num, total.minors.sumDev);
    minorQueueStdDev = stddev(total.minors.queueTime, total.minors.num, total.minors.queueSumDev);
    majorStdDev = stddev(total.majors.elapsed, total.majors.num, total.majors.sumDev);
    majorQueueStdDev = stddev(total.majors.queueTime, total.majors.num, total.majors.queueSumDev);
    splitStdDev = stddev(historical.splits.num, historical.splits.elapsed, historical.splits.sumDev);
    TabletServerDetailInformation details = doDetails(address, tsStats.size());
    List<AllTimeTabletResults> allTime = doAllTimeResults(majorQueueStdDev, minorQueueStdDev, totalElapsedForAll, splitStdDev, majorStdDev, minorStdDev);
    CurrentTabletResults currentRes = doCurrentTabletResults(currentMinorAvg, currentMinorStdDev, currentMajorAvg, currentMajorStdDev);
    TabletServerSummary tserverDetails = new TabletServerSummary(details, allTime, currentRes, currentOps);
    return tserverDetails;
}