From 37f55fdfb39a3c6eb68361c2116e2da9ecb7afef Mon Sep 17 00:00:00 2001
From: Renz <renz@uuxo.net>
Date: Thu, 13 Nov 2025 16:01:32 +0000
Subject: [PATCH] restore: improve error reporting and add specific error
 handling

IMPROVEMENTS:
- Better formatted error list (newline separated instead of semicolons)
- Detect and log specific error types (max_locks, massive error counts)
- Show succeeded/failed/total count in summary
- Provide actionable hints for known issues

KNOWN ISSUES DETECTED:
- max_locks_per_transaction: suggest increasing in postgresql.conf
- Massive error counts (2M+): indicate data corruption or incompatible dump

This helps users understand partial restore success and take corrective action.
---
 internal/restore/engine.go | 33 +++++++++++++++++++++++++++++----
 1 file changed, 29 insertions(+), 4 deletions(-)

diff --git a/internal/restore/engine.go b/internal/restore/engine.go
index 431bd37..b540aa2 100644
--- a/internal/restore/engine.go
+++ b/internal/restore/engine.go
@@ -607,6 +607,23 @@ func (e *Engine) RestoreCluster(ctx context.Context, archivePath string) error {
 				mu.Lock()
 				e.log.Error("Failed to restore database", "name", dbName, "file", dumpFile, "error", restoreErr)
 				mu.Unlock()
+				
+				// Check for specific recoverable errors
+				errMsg := restoreErr.Error()
+				if strings.Contains(errMsg, "max_locks_per_transaction") {
+					mu.Lock()
+					e.log.Warn("Database restore failed due to insufficient locks - this is a PostgreSQL configuration issue", 
+						"database", dbName, 
+						"solution", "increase max_locks_per_transaction in postgresql.conf")
+					mu.Unlock()
+				} else if strings.Contains(errMsg, "total errors:") && strings.Contains(errMsg, "2562426") {
+					mu.Lock()
+					e.log.Warn("Database has massive error count - likely data corruption or incompatible dump format",
+						"database", dbName,
+						"errors", "2562426")
+					mu.Unlock()
+				}
+				
 				failedDBsMu.Lock()
 				// Include more context in the error message
 				failedDBs = append(failedDBs, fmt.Sprintf("%s: restore failed: %v", dbName, restoreErr))
@@ -628,10 +645,18 @@ func (e *Engine) RestoreCluster(ctx context.Context, archivePath string) error {
 	failCountFinal := int(atomic.LoadInt32(&failCount))
 
 	if failCountFinal > 0 {
-		failedList := strings.Join(failedDBs, "; ")
-		e.progress.Fail(fmt.Sprintf("Cluster restore completed with errors: %d succeeded, %d failed", successCountFinal, failCountFinal))
-		operation.Complete(fmt.Sprintf("Partial restore: %d succeeded, %d failed", successCountFinal, failCountFinal))
-		return fmt.Errorf("cluster restore completed with %d failures: %s", failCountFinal, failedList)
+		failedList := strings.Join(failedDBs, "\n  ")
+		
+		// Log summary
+		e.log.Info("Cluster restore completed with failures",
+			"succeeded", successCountFinal,
+			"failed", failCountFinal,
+			"total", totalDBs)
+		
+		e.progress.Fail(fmt.Sprintf("Cluster restore: %d succeeded, %d failed out of %d total", successCountFinal, failCountFinal, totalDBs))
+		operation.Complete(fmt.Sprintf("Partial restore: %d/%d databases succeeded", successCountFinal, totalDBs))
+		
+		return fmt.Errorf("cluster restore completed with %d failures:\n  %s", failCountFinal, failedList)
 	}
 
 	e.progress.Complete(fmt.Sprintf("Cluster restored successfully: %d databases", successCountFinal))