Skip to main content
NetApp Knowledge Base site will be down for 3 hours between Oct 26, 23:59 PST and Oct 27, 02:59 PST, for system maintenance and infrastructure update.
NetApp Stage KB

Altavault Cloud Reclamation(aka garbage collection) FAQ

  1. Last updated
  2. Save as PDF
Views:
Visibility:
Public
Votes:
0
Category:
altavault
Specialty:
DP
Last Updated:

 

Applies to

  • AltaVault
  • Cloud Replication
  • Garbage Collection

Answer

  • Cloud Reclamation is the process of removing unused data from the cloud in response to the deletion of files on the front-end SMB or NFS shares.  
  • For efficiency, the cloud reclamation process runs periodically when sufficient number of files are ready to be processed.
  • After the files are deleted, the slabs they reference are added to a list
  • AltaVault uses 5% of total files or 5% of total data in the local cache
  • The cloud reclamation service runs to determine how much of each slab is still referenced by other files stored on the AltaVault.If no files reference the slab, in both the cloud and the local copy, the slab is deleted if it exists.
  • If the slab still contains data segments that are still being referenced, then the slab cannot be deleted.
  • However, if more than 50% of the slab is no longer required, and it is in the local cache, then it is compacted and re-enqueued for replication to the cloud where it overwrites the larger version.
  • To reduce the amount of messages being logged, AltaVault no longer prints messages from garbage collection acting on individual slab actions.

Messages:

Oct  8 01:27:10 altavault1 rfsd[19024]: [gc.INFO] (14306) bitmaps set up for 251656 slabs, using 32211968 bytes

Oct  8 02:08:43 altavault1 rfsd[19024]: [scan.INFO] (14306) processed 15403 labelmaps, 13166362825 labels in 1098.926 secs (00:18:18), 11981119.365 labels/sec, 10318004471 (9.6GB) lm bytes; 96682492516524 (87.9TB) expanded bytes

Oct  8 02:08:43 altavault1 rfsd[19024]: [gc.INFO] (14306) 251656 total slabs, (0 resized)

Oct  8 02:08:43 altavault1 rfsd[19024]: [gc.INFO] (14306) deleting metadata from cloud

Oct  8 02:08:51 altavault1 rfsd[19024]: [gc.INFO] (14306) enqueuing slabs for reclamation

Oct  8 02:08:51 altavault1 rfsd[19024]: [gc.INFO] (14306) slab: 0x000001560592 (128b) 0/1 0% :

Oct  8 02:08:51 altavault1 rfsd[19024]: [gc.INFO] (14306) slab: 0x000001560597 (128b) 16/250 6.4% : 0300e0a240000000020000000000000000000000800000018000000000008c00

Oct  8 02:08:51 altavault1 rfsd[19024]: [gc.INFO] (14306) slab: 0x0000015605a0 (128b) 77/223 34.5% : 000000007bc0000000000000000000000000000640c007ffe6fffff03fbffffb

Oct  8 02:08:51 altavault1 rfsd[19024]: [gc.INFO] (14306) slab: 0x0000015605a4 (128b) 29/228 12.7% : 0000000c2000000000000000c7f80000203800004001c041c380000000040000

Oct  8 02:08:51 altavault1 rfsd[19024]: [gc.INFO] (14306) slab: 0x0000015605a9 (128b) 68/162 42% : 0000000303e401ff80000007db9f04fcffffde3f00000000

  • If slabs are determined to be 0% referenced, they are deleted:

Oct  8 02:08:51 altavault1 rfsd[19024]: [gc.INFO] (14306) slab: 0x0000015605ab (128b) 0/1 0% :

Oct  8 02:08:51 altavault1 rfsd[19024]: [gc.INFO] (32480) removed slab 0x15605ab

  • If slabs are less than 50% referenced, and are in the local cache, they are compacted:

Oct  8 02:08:51 altavault1 rfsd[19024]: [gc.INFO] (14306) slab: 0x0000015605e2 (128b) 22/220 10% : 000000000900400000080120008015000200000b885808000000000010000200

Oct  8 02:08:51 altavault1 rfsd[19024]: [gc.INFO] (32480) compacted slab 0x15605e2

Oct  8 02:12:34 altavault1 rfsd[19024]: [replicator.INFO] (23497) replicated slab 0000015605E2.slab

  • At the end of the Cloud Reclamation cycle, a summary message is printed:

Oct  5 15:09:02 altavault1 rfsd[19024]: [gc.INFO] (7198) compacted 5790 slabs, removed 3960 slabs, (0 slabs already unlinked), 947652 segments, 14313056733 bytes (13.3GB) recovered, 4160.11s elapsed

  • The amount of data that can be recovered is dependent on the amount of deduplication that has occurred, the resulting percentage of unused data in the 4 MB slab objects, and if the data is still in the local cache. 
  • Optimal reclamation performance occurs when the average eviction age, the age of the data that is being deleted from the local cache to make room for new data, is greater than the expiration age of the backups, ensuring it is local and eligible for compaction. The average evicted age can be viewed in the WebUI under Reports.

Additional Information

additionalInformation_text

 

NetApp provides no representations or warranties regarding the accuracy or reliability or serviceability of any information or recommendations provided in this publication or with respect to any results that may be obtained by the use of the information or observance of any recommendations provided herein. The information in this document is distributed AS IS and the use of this information or the implementation of any recommendations or techniques herein is a customer's responsibility and depends on the customer's ability to evaluate and integrate them into the customer's operational environment. This document and the information contained herein may be used solely in connection with the NetApp products discussed in this document.