Yeah. Perhaps that chunk of code… is a bit too enthusiastic?
Welp that 3 TB collected trash is right going back. That will take some time.
That trashing and untrashing is like an emotion rollercoaster. Hope it ends soon
Maybe Storj heard of some competitor problems and just wanted to be 110% sure they’ve preserved every bit. Always nice to pick up some new customers when their current vendor has them spooked…
Haha, i saw that exact same Video yesterday and immediately started googling. But i didnt realy found something. So i couldn’t verify it. Do you maybe have a source?
This situation is being handled very unprofessional, and doesn’t help build trust in Storj.
and testing should only be done on Saltlake, and these tests concern hashstore or also piecestore?
Perhaps when trying to repair on production just use very small bloom filters to identify the problem.
Second step increase bloom filters slowly.
But please dont continue deleting and restoring big amounts of data.
interestingly the restore process moved on, then stopped for like 3 hours (on some nodes even up to 6 hours) and then moved on. Why is that so? Is it expected?
(Value on the left is GB)
Satellites are independent. So it was actually 3 restores.
My nodes received the restore commands a bit closer together.
US1 - 2025-02-28T16:37:39Z INFO pieces:trash restore trash started
Eu1 - 2025-02-28T16:46:31Z INFO pieces:trash restore trash started
Ap1 - 2025-02-28T17:09:57Z INFO pieces:trash restore trash started
Yes and no. All restore requests were received at the same time. There was no gap there. And I looked and there is no log entry for restore at the time the second wave started.
Hm, was there a new bloom filter processed during the stop?
No, because then the Trash size would increase again. Today at aprox. 13:30 3 Nodes recieved a bloomfilter for US
So obviously this was an illusion. With the last Bloom filter the totals all over my nodes are back as they were on January 1st. Growth = ZERO.
They want to replace old data, from the old free model with new paid data
So as long as there is free data left in the network you can expect the grow to be slower
They don’t even know how much data that is.
But suddenly they know it:
All, just wanted to post something here. We recognize there is a concern with the restoration from trash that is happening. There is an unknown error in the bloom filter generation process that is causing pieces to incorrectly be deleted. We are working on root causing the issue so that we can resolve it but in the meantime out of an abundance of caution we have been restoring data from trash to preserve pieces.
they are deleted friday-yesterday again, will they be restored again?


