lbcd/log.go
Dave Collins 7c174620f7 indexers: Implement optional tx/address indexes.
This introduces a new indexing infrastructure for supporting optional
indexes using the new database and blockchain infrastructure along with
two concrete indexer implementations which provide both a
transaction-by-hash and a transaction-by-address index.

The new infrastructure is mostly separated into a package named indexers
which is housed under the blockchain package.  In order to support this,
a new interface named IndexManager has been introduced in the blockchain
package which provides methods to be notified when the chain has been
initialized and when blocks are connected and disconnected from the main
chain.  A concrete implementation of an index manager is provided by the
new indexers package.

The new indexers package also provides a new interface named Indexer
which allows the index manager to manage concrete index implementations
which conform to the interface.

The following is high level overview of the main index infrastructure
changes:

- Define a new IndexManager interface in the blockchain package and
  modify the package to make use of the interface when specified
- Create a new indexers package
  - Provides an Index interface which allows concrete indexes to plugin
    to an index manager
  - Provides a concrete IndexManager implementation
    - Handles the lifecycle of all indexes it manages
    - Tracks the index tips
    - Handles catching up disabled indexes that have been reenabled
    - Handles reorgs while the index was disabled
    - Invokes the appropriate methods for all managed indexes to allow
      them to index and deindex the blocks and transactions
  - Implement a transaction-by-hash index
    - Makes use of internal block IDs to save a significant amount of
      space and indexing costs over the old transaction index format
  - Implement a transaction-by-address index
    - Makes use of a leveling scheme in order to provide a good tradeoff
      between space required and indexing costs
- Supports enabling and disabling indexes at will
- Support the ability to drop indexes if they are no longer desired

The following is an overview of the btcd changes:

- Add a new index logging subsystem
- Add new options --txindex and --addrindex in order to enable the
  optional indexes
  - NOTE: The transaction index will automatically be enabled when the
    address index is enabled because it depends on it
- Add new options --droptxindex and --dropaddrindex to allow the indexes
  to be removed
  - NOTE: The address index will also be removed when the transaction
    index is dropped because it depends on it
- Update getrawtransactions RPC to make use of the transaction index
- Reimplement the searchrawtransaction RPC that makes use of the address
  index
- Update sample-btcd.conf to include sample usage for the new optional
  index flags
2016-04-11 17:16:42 -05:00

362 lines
9.4 KiB
Go

// Copyright (c) 2013-2016 The btcsuite developers
// Use of this source code is governed by an ISC
// license that can be found in the LICENSE file.
package main
import (
"fmt"
"os"
"strings"
"time"
"github.com/btcsuite/btcd/addrmgr"
"github.com/btcsuite/btcd/blockchain"
"github.com/btcsuite/btcd/blockchain/indexers"
database "github.com/btcsuite/btcd/database2"
"github.com/btcsuite/btcd/peer"
"github.com/btcsuite/btcd/txscript"
"github.com/btcsuite/btcd/wire"
"github.com/btcsuite/btclog"
"github.com/btcsuite/seelog"
)
const (
// maxRejectReasonLen is the maximum length of a sanitized reject reason
// that will be logged.
maxRejectReasonLen = 250
)
// Loggers per subsystem. Note that backendLog is a seelog logger that all of
// the subsystem loggers route their messages to. When adding new subsystems,
// add a reference here, to the subsystemLoggers map, and the useLogger
// function.
var (
backendLog = seelog.Disabled
adxrLog = btclog.Disabled
amgrLog = btclog.Disabled
bcdbLog = btclog.Disabled
bmgrLog = btclog.Disabled
btcdLog = btclog.Disabled
chanLog = btclog.Disabled
discLog = btclog.Disabled
indxLog = btclog.Disabled
minrLog = btclog.Disabled
peerLog = btclog.Disabled
rpcsLog = btclog.Disabled
scrpLog = btclog.Disabled
srvrLog = btclog.Disabled
txmpLog = btclog.Disabled
)
// subsystemLoggers maps each subsystem identifier to its associated logger.
var subsystemLoggers = map[string]btclog.Logger{
"ADXR": adxrLog,
"AMGR": amgrLog,
"BCDB": bcdbLog,
"BMGR": bmgrLog,
"BTCD": btcdLog,
"CHAN": chanLog,
"DISC": discLog,
"INDX": indxLog,
"MINR": minrLog,
"PEER": peerLog,
"RPCS": rpcsLog,
"SCRP": scrpLog,
"SRVR": srvrLog,
"TXMP": txmpLog,
}
// logClosure is used to provide a closure over expensive logging operations
// so don't have to be performed when the logging level doesn't warrant it.
type logClosure func() string
// String invokes the underlying function and returns the result.
func (c logClosure) String() string {
return c()
}
// newLogClosure returns a new closure over a function that returns a string
// which itself provides a Stringer interface so that it can be used with the
// logging system.
func newLogClosure(c func() string) logClosure {
return logClosure(c)
}
// useLogger updates the logger references for subsystemID to logger. Invalid
// subsystems are ignored.
func useLogger(subsystemID string, logger btclog.Logger) {
if _, ok := subsystemLoggers[subsystemID]; !ok {
return
}
subsystemLoggers[subsystemID] = logger
switch subsystemID {
case "ADXR":
adxrLog = logger
case "AMGR":
amgrLog = logger
addrmgr.UseLogger(logger)
case "BCDB":
bcdbLog = logger
database.UseLogger(logger)
case "BMGR":
bmgrLog = logger
case "BTCD":
btcdLog = logger
case "CHAN":
chanLog = logger
blockchain.UseLogger(logger)
case "DISC":
discLog = logger
case "INDX":
indxLog = logger
indexers.UseLogger(logger)
case "MINR":
minrLog = logger
case "PEER":
peerLog = logger
peer.UseLogger(logger)
case "RPCS":
rpcsLog = logger
case "SCRP":
scrpLog = logger
txscript.UseLogger(logger)
case "SRVR":
srvrLog = logger
case "TXMP":
txmpLog = logger
}
}
// initSeelogLogger initializes a new seelog logger that is used as the backend
// for all logging subsystems.
func initSeelogLogger(logFile string) {
config := `
<seelog type="adaptive" mininterval="2000000" maxinterval="100000000"
critmsgcount="500" minlevel="trace">
<outputs formatid="all">
<console />
<rollingfile type="size" filename="%s" maxsize="10485760" maxrolls="3" />
</outputs>
<formats>
<format id="all" format="%%Time %%Date [%%LEV] %%Msg%%n" />
</formats>
</seelog>`
config = fmt.Sprintf(config, logFile)
logger, err := seelog.LoggerFromConfigAsString(config)
if err != nil {
fmt.Fprintf(os.Stderr, "failed to create logger: %v", err)
os.Exit(1)
}
backendLog = logger
}
// setLogLevel sets the logging level for provided subsystem. Invalid
// subsystems are ignored. Uninitialized subsystems are dynamically created as
// needed.
func setLogLevel(subsystemID string, logLevel string) {
// Ignore invalid subsystems.
logger, ok := subsystemLoggers[subsystemID]
if !ok {
return
}
// Default to info if the log level is invalid.
level, ok := btclog.LogLevelFromString(logLevel)
if !ok {
level = btclog.InfoLvl
}
// Create new logger for the subsystem if needed.
if logger == btclog.Disabled {
logger = btclog.NewSubsystemLogger(backendLog, subsystemID+": ")
useLogger(subsystemID, logger)
}
logger.SetLevel(level)
}
// setLogLevels sets the log level for all subsystem loggers to the passed
// level. It also dynamically creates the subsystem loggers as needed, so it
// can be used to initialize the logging system.
func setLogLevels(logLevel string) {
// Configure all sub-systems with the new logging level. Dynamically
// create loggers as needed.
for subsystemID := range subsystemLoggers {
setLogLevel(subsystemID, logLevel)
}
}
// directionString is a helper function that returns a string that represents
// the direction of a connection (inbound or outbound).
func directionString(inbound bool) string {
if inbound {
return "inbound"
}
return "outbound"
}
// formatLockTime returns a transaction lock time as a human-readable string.
func formatLockTime(lockTime uint32) string {
// The lock time field of a transaction is either a block height at
// which the transaction is finalized or a timestamp depending on if the
// value is before the txscript.LockTimeThreshold. When it is under the
// threshold it is a block height.
if lockTime < txscript.LockTimeThreshold {
return fmt.Sprintf("height %d", lockTime)
}
return time.Unix(int64(lockTime), 0).String()
}
// invSummary returns an inventory message as a human-readable string.
func invSummary(invList []*wire.InvVect) string {
// No inventory.
invLen := len(invList)
if invLen == 0 {
return "empty"
}
// One inventory item.
if invLen == 1 {
iv := invList[0]
switch iv.Type {
case wire.InvTypeError:
return fmt.Sprintf("error %s", iv.Hash)
case wire.InvTypeBlock:
return fmt.Sprintf("block %s", iv.Hash)
case wire.InvTypeTx:
return fmt.Sprintf("tx %s", iv.Hash)
}
return fmt.Sprintf("unknown (%d) %s", uint32(iv.Type), iv.Hash)
}
// More than one inv item.
return fmt.Sprintf("size %d", invLen)
}
// locatorSummary returns a block locator as a human-readable string.
func locatorSummary(locator []*wire.ShaHash, stopHash *wire.ShaHash) string {
if len(locator) > 0 {
return fmt.Sprintf("locator %s, stop %s", locator[0], stopHash)
}
return fmt.Sprintf("no locator, stop %s", stopHash)
}
// sanitizeString strips any characters which are even remotely dangerous, such
// as html control characters, from the passed string. It also limits it to
// the passed maximum size, which can be 0 for unlimited. When the string is
// limited, it will also add "..." to the string to indicate it was truncated.
func sanitizeString(str string, maxLength uint) string {
const safeChars = "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXY" +
"Z01234567890 .,;_/:?@"
// Strip any characters not in the safeChars string removed.
str = strings.Map(func(r rune) rune {
if strings.IndexRune(safeChars, r) >= 0 {
return r
}
return -1
}, str)
// Limit the string to the max allowed length.
if maxLength > 0 && uint(len(str)) > maxLength {
str = str[:maxLength]
str = str + "..."
}
return str
}
// messageSummary returns a human-readable string which summarizes a message.
// Not all messages have or need a summary. This is used for debug logging.
func messageSummary(msg wire.Message) string {
switch msg := msg.(type) {
case *wire.MsgVersion:
return fmt.Sprintf("agent %s, pver %d, block %d",
msg.UserAgent, msg.ProtocolVersion, msg.LastBlock)
case *wire.MsgVerAck:
// No summary.
case *wire.MsgGetAddr:
// No summary.
case *wire.MsgAddr:
return fmt.Sprintf("%d addr", len(msg.AddrList))
case *wire.MsgPing:
// No summary - perhaps add nonce.
case *wire.MsgPong:
// No summary - perhaps add nonce.
case *wire.MsgAlert:
// No summary.
case *wire.MsgMemPool:
// No summary.
case *wire.MsgTx:
return fmt.Sprintf("hash %s, %d inputs, %d outputs, lock %s",
msg.TxSha(), len(msg.TxIn), len(msg.TxOut),
formatLockTime(msg.LockTime))
case *wire.MsgBlock:
header := &msg.Header
return fmt.Sprintf("hash %s, ver %d, %d tx, %s", msg.BlockSha(),
header.Version, len(msg.Transactions), header.Timestamp)
case *wire.MsgInv:
return invSummary(msg.InvList)
case *wire.MsgNotFound:
return invSummary(msg.InvList)
case *wire.MsgGetData:
return invSummary(msg.InvList)
case *wire.MsgGetBlocks:
return locatorSummary(msg.BlockLocatorHashes, &msg.HashStop)
case *wire.MsgGetHeaders:
return locatorSummary(msg.BlockLocatorHashes, &msg.HashStop)
case *wire.MsgHeaders:
return fmt.Sprintf("num %d", len(msg.Headers))
case *wire.MsgReject:
// Ensure the variable length strings don't contain any
// characters which are even remotely dangerous such as HTML
// control characters, etc. Also limit them to sane length for
// logging.
rejCommand := sanitizeString(msg.Cmd, wire.CommandSize)
rejReason := sanitizeString(msg.Reason, maxRejectReasonLen)
summary := fmt.Sprintf("cmd %v, code %v, reason %v", rejCommand,
msg.Code, rejReason)
if rejCommand == wire.CmdBlock || rejCommand == wire.CmdTx {
summary += fmt.Sprintf(", hash %v", msg.Hash)
}
return summary
}
// No summary for other messages.
return ""
}