Moving node to new/bigger harddrive , skipping trash possible?

Node Operators FAQ
,
1

I am moving nodes to a bigger HDD on windows

already know how to do it , and have done it multiple times

But my question is , can i skip the trash folder and not restore it on the new harddrive only this one time ?
Will it have any consequence on the node , or with the database , the filewalker or with anything ?

Thanks

(trash i quite large sometime , and this is precious wasted time , for file and folder that may not even be here when you launch Robocopy for the 2nd or 3rd time)

1 Like
2

Reasons that justify existence of trash don’t disappear just because you are doing migration.

On the other hand:

  • If you go robocopy route, then time it takes is irrelevant. Node stays online.
  • If you go disk cloning route - then you can’t copy selectively.

In either scenarios the point is moot.

1 Like
3

The trash could get audited and it could knock out your audit score.

Using robocopy/rsync… yeah it’s boring but I would just bite the bullet and suck it up and wait for it to finish.

Now, I had a case where my hard disk filled up HARD, and nothing could run anymore. So I went in and deleted one day’s worth of trash just to get the node operational again. And it did drop my audit score a percent or two, but it wasn’t fatal.

Incidentally I am now migrating two 7TB nodes… looks like it’s going to take two weeks or so just for the first rsync pass!

2 Likes
4

yes thats why i asked the question , migrating 6TB nodes , it take already more than one day for just the trash folder , and by the time i launch robocopy/rsync for the second or third time

i have spend more than a day copying entire folders that have been deleted in the mean time

5

You may reduce the amount of syncing data if you would set an allocation below the usage (it will stop an ingress only), you may also disable the scan on startup (to do not spend time on updating databases), save the config and restart the node.

1 Like
6

already do that , this is indeed good practice :wink:

7

You could exclude the whole trash folder or the folders with the oldest date for the first few passes and hope that they will be gone when you do the final pass. The final pass should be done with the trash folder.

That’s what I did recently.

4 Likes
8

What thrash you talking about? Storj is so lazy with bloom filters atm…I only have very little trash from saltlake but nothing from the other 3 sats.

9

I do have processed BF also for US1, SLC and EU1, and the last for for AP1 was in the end of the last month. However, my nodes hosts a little data for AP1 for some reason.

However the trash contains not deleted subfolders only for SLC (also for Stefan, but it’s empty).

$ sudo tree -L 2 /mnt/x/storagenode2/storage/trash/
/mnt/x/storagenode2/storage/trash/
├── abforhuxbzyd35blusvrifvdwmfx4hmocsva4vmpp3rgqaaaaaaa
│   └── 2024-04-20
├── pmw6tvzmf2jv6giyybmmvl4o2ahqlaldsaeha4yx74n5aaaaaaaa
│   ├── 2024-07-09
│   └── 2024-07-13
├── qstuylguhrn2ozjv4h2c6xpxykd622gtgurhql2k7k75wqaaaaaa
├── ukfu6bhbboxilvt7jrwlqk7y2tapb5d2r2tsmj2sjxvw5qaaaaaa
└── v4weeab67sbgvnbwd5z7tweqsqqun7qox2agpbxy44mqqaaaaaaa