inspect: add inspect mode for debugging crashed tendermint node (#6785)

EDIT: Updated, see [comment below]( https://github.com/tendermint/tendermint/pull/6785#issuecomment-897793175)

This change adds a sketch of the `Debug` mode. 

This change adds a `Debug` struct to the node package. This `Debug` struct is intended to be created and started by a command in the `cmd` directory. The `Debug` struct runs the RPC server on the data directories: both the state store and the block store.

This change required a good deal of refactoring. Namely, a new `rpc.go` file was added to the `node` package. This file encapsulates functions for starting RPC servers used by nodes. A potential additional change is to further factor this code into shared code _in_ the `rpc` package. 

Minor API tweaks were also made that seemed appropriate such as the mechanism for fetching routes from the `rpc/core` package.

Additional work is required to register the `Debug` service as a command in the `cmd` directory but I am looking for feedback on if this direction seems appropriate before diving much further.

closes: #5908
This commit is contained in:
William Banfield
2021-08-24 18:12:06 +00:00
committed by GitHub
parent 6d5ff590c3
commit bc2b529b95
21 changed files with 1200 additions and 116 deletions
+87
View File
@@ -0,0 +1,87 @@
package commands
import (
"context"
"os"
"os/signal"
"syscall"
"github.com/spf13/cobra"
cfg "github.com/tendermint/tendermint/config"
"github.com/tendermint/tendermint/inspect"
"github.com/tendermint/tendermint/state"
"github.com/tendermint/tendermint/state/indexer/sink"
"github.com/tendermint/tendermint/store"
"github.com/tendermint/tendermint/types"
)
// InspectCmd is the command for starting an inspect server.
var InspectCmd = &cobra.Command{
Use: "inspect",
Short: "Run an inspect server for investigating Tendermint state",
Long: `
inspect runs a subset of Tendermint's RPC endpoints that are useful for debugging
issues with Tendermint.
When the Tendermint consensus engine detects inconsistent state, it will crash the
tendermint process. Tendermint will not start up while in this inconsistent state.
The inspect command can be used to query the block and state store using Tendermint
RPC calls to debug issues of inconsistent state.
`,
RunE: runInspect,
}
func init() {
InspectCmd.Flags().
String("rpc.laddr",
config.RPC.ListenAddress, "RPC listenener address. Port required")
InspectCmd.Flags().
String("db-backend",
config.DBBackend, "database backend: goleveldb | cleveldb | boltdb | rocksdb | badgerdb")
InspectCmd.Flags().
String("db-dir", config.DBPath, "database directory")
}
func runInspect(cmd *cobra.Command, args []string) error {
ctx, cancel := context.WithCancel(cmd.Context())
defer cancel()
c := make(chan os.Signal, 1)
signal.Notify(c, syscall.SIGTERM, syscall.SIGINT)
go func() {
<-c
cancel()
}()
blockStoreDB, err := cfg.DefaultDBProvider(&cfg.DBContext{ID: "blockstore", Config: config})
if err != nil {
return err
}
blockStore := store.NewBlockStore(blockStoreDB)
stateDB, err := cfg.DefaultDBProvider(&cfg.DBContext{ID: "state", Config: config})
if err != nil {
if err := blockStoreDB.Close(); err != nil {
logger.Error("error closing block store db", "error", err)
}
return err
}
genDoc, err := types.GenesisDocFromFile(config.GenesisFile())
if err != nil {
return err
}
sinks, err := sink.EventSinksFromConfig(config, cfg.DefaultDBProvider, genDoc.ChainID)
if err != nil {
return err
}
stateStore := state.NewStore(stateDB)
ins := inspect.New(config.RPC, blockStore, stateStore, sinks, logger)
logger.Info("starting inspect server")
if err := ins.Run(ctx); err != nil {
return err
}
return nil
}
+1
View File
@@ -28,6 +28,7 @@ func main() {
cmd.ShowNodeIDCmd,
cmd.GenNodeKeyCmd,
cmd.VersionCmd,
cmd.InspectCmd,
cmd.MakeKeyMigrateCommand(),
debug.DebugCmd,
cli.NewCompletionCmd(rootCmd, true),