MyFavMirrors/minio
1
0
Fork 0
mirror of https://github.com/minio/minio.git synced 2025-11-07 12:52:58 -05:00
Code Issues Packages Projects Releases Wiki Activity

initialize the disk healer early on (#19143)

Browse Source 
This PR fixes a bug that perhaps has been long introduced,
with no visible workarounds. In any deployment, if an entire
erasure set is deleted, there is no way the cluster recovers.
This commit is contained in:
Harshavardhana
2024-02-27 23:02:14 -08:00
committed by GitHub
parent 0aae0180fb
commit 9a012a53ef
15 changed files with 59 additions and 122 deletions
Download Patch File Download Diff File Expand all files Collapse all files

22
cmd/background-newdisks-heal-ops.go
View File

@@ -376,26 +376,8 @@ func getLocalDisksToHeal() (disksToHeal Endpoints) {
var newDiskHealingTimeout = newDynamicTimeout(30*time.Second, 10*time.Second)
func healFreshDisk(ctx context.Context, z *erasureServerPools, endpoint Endpoint) error {
disk, format, _, err := connectEndpoint(endpoint)
if err != nil {
return fmt.Errorf("Error: %w, %s", err, endpoint)
}
defer disk.Close()
poolIdx := globalEndpoints.GetLocalPoolIdx(disk.Endpoint())
if poolIdx < 0 {
return fmt.Errorf("unexpected pool index (%d) found for %s", poolIdx, disk.Endpoint())
}
// Calculate the set index where the current endpoint belongs
z.serverPools[poolIdx].erasureDisksMu.RLock()
setIdx, _, err := findDiskIndex(z.serverPools[poolIdx].format, format)
z.serverPools[poolIdx].erasureDisksMu.RUnlock()
if err != nil {
return err
}
if setIdx < 0 {
return fmt.Errorf("unexpected set index (%d) found for %s", setIdx, disk.Endpoint())
}
poolIdx, setIdx := endpoint.PoolIdx, endpoint.SetIdx
disk := getStorageViaEndpoint(endpoint)
// Prevent parallel erasure set healing
locker := z.NewNSLock(minioMetaBucket, fmt.Sprintf("new-drive-healing/%d/%d", poolIdx, setIdx))