ExpiredObjectDeleteMarker with multiple deleted markers

[restive sedge]

Hello,

I have a bucket with active versioning. Now we have a lot of objects with 1000 revisions, all (deleted) -> deleted markers.

I have this lifecycle policy active on this bucket:

{
    "Rules": [
        {
            "ID": "delete markers",
            "Filter": {},
            "Status": "Enabled",
            "Expiration": {
                "ExpiredObjectDeleteMarker": true
            }
        },
        {
            "ID": "delete noncurrent versions",
            "Filter": {},
            "Status": "Enabled",
            "NoncurrentVersionExpiration": {
                "NoncurrentDays": 1
            }
        }
    ]
}

Still nothing is deleted, I guess the issue is, that we have not only ONE deleted marker, but around 1000 of them.

Is there an option to do this with a lifecycle policy or do I have to manually delete this markers?!

[somber orchid]

Think this KB explains it.
https://kb.netapp.com/hybrid/StorageGRID/Object_Mgmt/Delete_markers_persist_in_an_object_versioning_enabled_bucket_on_StorageGRID
In short, if you're on 11.8 you have support for ExpiredObjectDeleteMarker. If you're not on 11.8 and upgrade this new feature will only work on newly created markers. For 11.7 and earlier your left with manual deletion (using some s3 client) or scripting.

[restive sedge]

Ok, thank you for the information. I'll do some testing & plan the update 🙁

[restive sedge]

I have another question. I have a bucket with data created in 2026 (we are on 11.9 since 2025) - so all objects and the bucket were created after 11.9 release update. I configured the above mentionend lifecycle policy. Currently we have around 424.12 TB of data inside the bucket. The policy is active and we still see very little deletions. Is there any way we can increase the speed of deletions or change the priority? We see a decrease of data of 0.01TB in over 24h. This will take years to complete.

Object versions: 10447576
Delete markers: 11080444
Total entries: 21528020

Found something about async deletion queue, but can't find the current number. (how many deletions are currently in the queue)

[somber orchid]

Typically I would suggest going the script route to get things done quickly, relying on the lifecycle process is often more time consuming. That being said, what you describe sounds veeery slow. There's a KB available that describe how to adjust "delete settings" in case those are causing latencies. I'm not suggesting you tweak things (better to have a support case open to investigate), but I guess that KB can be used to do the opposite or at least check so that the default settings are in place and nothing have been tweaked already that might explain the slow deletes.
https://kb.netapp.com/hybrid/StorageGRID/Object_Mgmt/StorageGRID_experiencing_high_latency_during_batch_deletes