Files
seaweedfs/weed/command/shell.go
T
Chris LuandDevin 85c1b990b1 shell: bound the initial master connection wait behind -connectTimeout (#11922)
* shell: bound the initial master connection wait behind -connectTimeout

weed shell waited on WaitUntilConnected with a background context, so
an unreachable master hung the command forever. The wait now runs on
a 30s (configurable) timeout and reports the failure instead.

Generated with [Devin](https://devin.ai)

Co-Authored-By: Devin <158243242+devin-ai-integration[bot]@users.noreply.github.com>

* shell: stop the master reconnect loop when RunShell returns

KeepConnectedToMaster ran on context.Background(), so once RunShell
returned — including the new connect-timeout early return — the
reconnect goroutine kept dialing forever, and repeated RunShell calls
accumulated one each. Give it its own cancellable context, separate
from connectCtx whose deadline must not cut reconnects under a live
shell.

Generated with [Devin](https://devin.ai)

Co-Authored-By: Devin <158243242+devin-ai-integration[bot]@users.noreply.github.com>

* shell: bound filer discovery by -connectTimeout too

The discovery WithClient and ListClusterNodes ran on the unbounded
background context, so a master disconnect after the initial connect —
or a persistently failing ListClusterNodes — could block startup past
the deadline. Reusing connectCtx bounds the whole startup phase.

Generated with [Devin](https://devin.ai)

Co-Authored-By: Devin <158243242+devin-ai-integration[bot]@users.noreply.github.com>

* shell: report filer discovery failures instead of silently losing the filer target

* shell: give filer discovery its own budget after master connects

Discovery shared connectCtx, so a connection that landed near the
-connectTimeout deadline left ListClusterNodes almost no time and no room
for WithClient to retry. A fresh timeout keeps discovery bounded without
starving it.

Generated with [Devin](https://devin.ai)

Co-Authored-By: Devin <158243242+devin-ai-integration[bot]@users.noreply.github.com>

* shell: cover the -connectTimeout master wait with a test

Generated with [Devin](https://devin.ai)

Co-Authored-By: Devin <158243242+devin-ai-integration[bot]@users.noreply.github.com>

* shell: note filer discovery in the -connectTimeout help text

Generated with [Devin](https://devin.ai)

Co-Authored-By: Devin <158243242+devin-ai-integration[bot]@users.noreply.github.com>

---------

Co-authored-by: Devin <158243242+devin-ai-integration[bot]@users.noreply.github.com>
2026-10-21 20:17:51 -07:00

84 lines
2.5 KiB
Go

package command
import (
"fmt"
"os"
"time"
"github.com/seaweedfs/seaweedfs/weed/pb"
"github.com/seaweedfs/seaweedfs/weed/security"
"github.com/seaweedfs/seaweedfs/weed/shell"
"github.com/seaweedfs/seaweedfs/weed/util"
)
var (
shellOptions shell.ShellOptions
shellInitialFiler *string
shellCluster *string
shellDebug *bool
shellConnectTimeout *time.Duration
)
func init() {
cmdShell.Run = runShell // break init cycle
shellOptions.Masters = cmdShell.Flag.String("master", "", "comma-separated master servers, e.g. localhost:9333")
shellOptions.FilerGroup = cmdShell.Flag.String("filerGroup", "", "filerGroup for the filers")
shellInitialFiler = cmdShell.Flag.String("filer", "", "filer host and port for initial connection, e.g. localhost:8888")
shellCluster = cmdShell.Flag.String("cluster", "", "cluster defined in shell.toml")
shellDebug = cmdShell.Flag.Bool("debug", false, "print informational logs to stderr")
shellConnectTimeout = cmdShell.Flag.Duration("connectTimeout", 30*time.Second, "timeout for the initial master connection and filer discovery")
}
var cmdShell = &Command{
UsageLine: "shell",
Short: "run interactive administrative commands",
Long: `run interactive administrative commands.
Generate shell.toml via "weed scaffold -config=shell"
`,
}
func runShell(command *Command, args []string) bool {
util.LoadSecurityConfiguration()
shellOptions.GrpcDialOption = security.LoadClientTLS(util.GetViper(), "grpc.client")
shellOptions.Directory = "/"
util.LoadConfiguration("shell", false)
viper := util.GetViper()
cluster := viper.GetString("cluster.default")
if *shellCluster != "" {
cluster = *shellCluster
}
if *shellOptions.Masters == "" {
if cluster == "" {
*shellOptions.Masters = "localhost:9333"
} else {
*shellOptions.Masters = viper.GetString("cluster." + cluster + ".master")
}
}
filerAddress := *shellInitialFiler
if filerAddress == "" && cluster != "" {
filerAddress = viper.GetString("cluster." + cluster + ".filer")
}
shellOptions.FilerAddress = pb.ServerAddress(filerAddress)
shellOptions.Debug = *shellDebug
shellOptions.ConnectTimeout = *shellConnectTimeout
if shellOptions.Debug {
fmt.Fprintf(os.Stderr, "master: %s filer: %s\n", *shellOptions.Masters, shellOptions.FilerAddress)
}
if err := shell.RunShell(shellOptions); err != nil {
// The command already printed the error; a piped run has to fail the
// script wrapping it too.
SetCommandExitStatus(1)
}
return true
}