writecache: Fix remaining addr2key uses #660

Merged

fyrchik merged 1 commit from ale64bit/frostfs-node:fix/badger-delete-addr-bug into master

2023-08-30 17:12:36 +00:00

ale64bit commented

2023-08-29 12:27:31 +00:00

Member

Signed-off-by: Alejandro Lopez a.lopez@yadro.com

Signed-off-by: Alejandro Lopez <a.lopez@yadro.com>

ale64bit force-pushed fix/badger-delete-addr-bug from 0d0394172f to 3781479a85

2023-08-29 12:28:01 +00:00

Compare

ale64bit requested review from storage-core-committers 2023-08-29 12:28:26 +00:00

ale64bit requested review from storage-core-developers 2023-08-29 12:28:32 +00:00

acid-ant approved these changes 2023-08-29 12:35:41 +00:00

fyrchik approved these changes 2023-08-30 12:14:42 +00:00

fyrchik left a comment

Owner

Are these errors caught by tests?

ale64bit commented

2023-08-30 12:18:19 +00:00

Author

Member

Are these errors caught by tests?

Not currently, AFAIK. I discovered it with a test for the lite cache: https://git.frostfs.info/TrueCloudLab/frostfs-node/pulls/654/files#diff-40a69d51c07d693c799c744120ffaf50e90e53f2. We can generalize that one after it's merged and use it for the other caches as well.

> Are these errors caught by tests? Not currently, AFAIK. I discovered it with a test for the lite cache: https://git.frostfs.info/TrueCloudLab/frostfs-node/pulls/654/files#diff-40a69d51c07d693c799c744120ffaf50e90e53f2. We can generalize that one after it's merged and use it for the other caches as well.

fyrchik commented

2023-08-30 13:19:15 +00:00

Owner

@ale64bit I thought Generic tests are exactly like this (and more comprehensive) -- it seems not.

ale64bit commented

2023-08-30 13:21:51 +00:00

Author

Member

@ale64bit I thought Generic tests are exactly like this (and more comprehensive) -- it seems not.

In this particular case, the test caught it because it deletes twice the same key, which should result first in no error and then in ObjectNotFound error, as the contract requires. But at any rate, I do think that the generic tests should be more comprehensive than they currently are.

> @ale64bit I thought Generic tests are exactly like this (and more comprehensive) -- it seems not. In this particular case, the test caught it because it deletes twice the same key, which should result first in no error and then in `ObjectNotFound` error, as the contract requires. But at any rate, I do think that the generic tests should be more comprehensive than they currently are.