X-Git-Url: https://git.arvados.org/arvados.git/blobdiff_plain/e502060ffe4f68d33e2cca8f8d7544ce40d53eb7..b49229f98012d7c08ce02b8d28dbcc165c8a6c53:/services/keep-balance/collection.go

diff --git a/services/keep-balance/collection.go b/services/keep-balance/collection.go
index a7a484672a..c4ddc90c41 100644
--- a/services/keep-balance/collection.go
+++ b/services/keep-balance/collection.go
@@ -1,10 +1,14 @@
+// Copyright (C) The Arvados Authors. All rights reserved.
+//
+// SPDX-License-Identifier: AGPL-3.0
+
 package main
 
 import (
 	"fmt"
 	"time"
 
-	"git.curoverse.com/arvados.git/sdk/go/arvados"
+	"git.arvados.org/arvados.git/sdk/go/arvados"
 )
 
 func countCollections(c *arvados.Client, params arvados.ResourceListParams) (int, error) {
@@ -32,23 +36,37 @@ func EachCollection(c *arvados.Client, pageSize int, f func(arvados.Collection)
 	}
 
 	expectCount, err := countCollections(c, arvados.ResourceListParams{
-		IncludeTrash: true,
+		IncludeTrash:       true,
+		IncludeOldVersions: true,
 	})
 	if err != nil {
 		return err
 	}
 
+	// Note the obvious way to get all collections (sorting by
+	// UUID) would be much easier, but would lose data: If a
+	// client were to move files from collection with uuid="zzz"
+	// to a collection with uuid="aaa" around the time when we
+	// were fetching the "mmm" page, we would never see those
+	// files' block IDs at all -- even if the client is careful to
+	// save "aaa" before saving "zzz".
+	//
+	// Instead, we get pages in modified_at order. Collections
+	// that are modified during the run will be re-fetched in a
+	// subsequent page.
+
 	limit := pageSize
 	if limit <= 0 {
 		// Use the maximum page size the server allows
 		limit = 1<<31 - 1
 	}
 	params := arvados.ResourceListParams{
-		Limit:        &limit,
-		Order:        "modified_at, uuid",
-		Count:        "none",
-		Select:       []string{"uuid", "unsigned_manifest_text", "modified_at", "portable_data_hash", "replication_desired"},
-		IncludeTrash: true,
+		Limit:              &limit,
+		Order:              "modified_at, uuid",
+		Count:              "none",
+		Select:             []string{"uuid", "unsigned_manifest_text", "modified_at", "portable_data_hash", "replication_desired"},
+		IncludeTrash:       true,
+		IncludeOldVersions: true,
 	}
 	var last arvados.Collection
 	var filterTime time.Time
@@ -62,7 +80,7 @@ func EachCollection(c *arvados.Client, pageSize int, f func(arvados.Collection)
 			return err
 		}
 		for _, coll := range page.Items {
-			if last.ModifiedAt != nil && *last.ModifiedAt == *coll.ModifiedAt && last.UUID >= coll.UUID {
+			if last.ModifiedAt == coll.ModifiedAt && last.UUID >= coll.UUID {
 				continue
 			}
 			callCount++
@@ -74,9 +92,9 @@ func EachCollection(c *arvados.Client, pageSize int, f func(arvados.Collection)
 		}
 		if len(page.Items) == 0 && !gettingExactTimestamp {
 			break
-		} else if last.ModifiedAt == nil {
+		} else if last.ModifiedAt.IsZero() {
 			return fmt.Errorf("BUG: Last collection on the page (%s) has no modified_at timestamp; cannot make progress", last.UUID)
-		} else if len(page.Items) > 0 && *last.ModifiedAt == filterTime {
+		} else if len(page.Items) > 0 && last.ModifiedAt == filterTime {
 			// If we requested time>=X and never got a
 			// time>X then we might not have received all
 			// items with time==X yet. Switch to
@@ -117,7 +135,7 @@ func EachCollection(c *arvados.Client, pageSize int, f func(arvados.Collection)
 			// avoiding that would add overhead in the
 			// overwhelmingly common cases, so we don't
 			// bother.
-			filterTime = *last.ModifiedAt
+			filterTime = last.ModifiedAt
 			params.Filters = []arvados.Filter{{
 				Attr:     "modified_at",
 				Operator: ">=",
@@ -136,7 +154,8 @@ func EachCollection(c *arvados.Client, pageSize int, f func(arvados.Collection)
 			Attr:     "modified_at",
 			Operator: "<=",
 			Operand:  filterTime}},
-		IncludeTrash: true,
+		IncludeTrash:       true,
+		IncludeOldVersions: true,
 	}); err != nil {
 		return err
 	} else if callCount < checkCount {