IO: Fix Root/Shizuku error when reading large directories#1360
Merged
Conversation
Fix services being destroyed before all data has been transferred Similar to `LocalGateway.read(...): Source` where we need to keep the source alive until consumed, we need to do the same when returning a `Flow`. It's up to the caller on when it's consumed. Likely caused by #1345 Previously we had a constant lease on the Shizuku service resource so this never surfaced: https://github.com/d4rken-org/sdmaid-se/pull/1345/files#diff-3105723dbd44154e23041c05d9496670e61e15cbdcd00ea73c99d1984f577f76R135 Root will have the same issue as that changed in #1346 This fixes #1358
d4rken
added
bug
d4rken
added a commit
that referenced
this pull request
Aug 9, 2024
While the parent (LocalGateway or PkgOps) is active and they themself as resource are being kept alive, make sure we keep childs like Shizuku or Root sessions alive too. Closing and opening the service connections (releasing the child resources) while the parent resource is alive, just adds extra churn. Also see #1360 ``` 18:46:35.645 SDMSE:TaskManager:SR D get(205551414): Reviving SharedResource 18:46:35.650 SDMSE:TaskManager:SR D Acquiring shared resource... 18:46:35.653 SDMSE:AppCleaner:SR D get(163473936): Reviving SharedResource 18:46:35.656 SDMSE:AppCleaner:SR D Acquiring shared resource... 18:46:35.664 SDMSE:CSI:FileForensics:SR D get(54980210): Reviving SharedResource 18:46:35.666 SDMSE:CSI:FileForensics:SR D Acquiring shared resource... 18:46:35.670 SDMSE:Gateway:Switch:SR D get(214268849): Reviving SharedResource 18:46:35.673 SDMSE:Gateway:Switch:SR D Acquiring shared resource... 18:46:35.676 SDMSE:Pkg:Ops:SR D get(250454437): Reviving SharedResource 18:46:35.679 SDMSE:Pkg:Ops:SR D Acquiring shared resource... 18:46:35.747 SDMSE:Gateway:Local:SR D get(252054746): Reviving SharedResource 18:46:35.749 SDMSE:Gateway:Local:SR D Acquiring shared resource... 18:46:35.788 SDMSE:Shizuku:Service:Client:SR D get(27747328): Reviving SharedResource 18:46:35.790 SDMSE:Shizuku:Service:Client:SR D Acquiring shared resource... 18:46:45.315 SDMSE:TaskManager D Releasing resource lock for 267d2fe2-b3f0-4f94-90f6-2a1369b22794 18:46:46.328 SDMSE:TaskManager:SR D Shared resource flow completed. 18:46:47.356 SDMSE:AppCleaner:SR D Shared resource flow completed. 18:46:48.376 SDMSE:CSI:FileForensics:SR D Shared resource flow completed. 18:46:49.387 SDMSE:Pkg:Ops:SR D Shared resource flow completed. 18:46:49.390 SDMSE:Gateway:Switch:SR D Shared resource flow completed. 18:46:50.397 SDMSE:Gateway:Local:SR D Shared resource flow completed. 18:46:51.441 SDMSE:Shizuku:Service:Client:SR D Shared resource flow completed. ```
d4rken
added a commit
that referenced
this pull request
Aug 9, 2024
While the parent (LocalGateway or PkgOps) is active and they themself as resource are being kept alive, make sure we keep childs like Shizuku or Root sessions alive too. Closing and opening the service connections (releasing the child resources) while the parent resource is alive, just adds extra churn. Also see #1360 ``` 18:46:35.645 SDMSE:TaskManager:SR D get(205551414): Reviving SharedResource 18:46:35.650 SDMSE:TaskManager:SR D Acquiring shared resource... 18:46:35.653 SDMSE:AppCleaner:SR D get(163473936): Reviving SharedResource 18:46:35.656 SDMSE:AppCleaner:SR D Acquiring shared resource... 18:46:35.664 SDMSE:CSI:FileForensics:SR D get(54980210): Reviving SharedResource 18:46:35.666 SDMSE:CSI:FileForensics:SR D Acquiring shared resource... 18:46:35.670 SDMSE:Gateway:Switch:SR D get(214268849): Reviving SharedResource 18:46:35.673 SDMSE:Gateway:Switch:SR D Acquiring shared resource... 18:46:35.676 SDMSE:Pkg:Ops:SR D get(250454437): Reviving SharedResource 18:46:35.679 SDMSE:Pkg:Ops:SR D Acquiring shared resource... 18:46:35.747 SDMSE:Gateway:Local:SR D get(252054746): Reviving SharedResource 18:46:35.749 SDMSE:Gateway:Local:SR D Acquiring shared resource... 18:46:35.788 SDMSE:Shizuku:Service:Client:SR D get(27747328): Reviving SharedResource 18:46:35.790 SDMSE:Shizuku:Service:Client:SR D Acquiring shared resource... 18:46:45.315 SDMSE:TaskManager D Releasing resource lock for 267d2fe2-b3f0-4f94-90f6-2a1369b22794 18:46:46.328 SDMSE:TaskManager:SR D Shared resource flow completed. 18:46:47.356 SDMSE:AppCleaner:SR D Shared resource flow completed. 18:46:48.376 SDMSE:CSI:FileForensics:SR D Shared resource flow completed. 18:46:49.387 SDMSE:Pkg:Ops:SR D Shared resource flow completed. 18:46:49.390 SDMSE:Gateway:Switch:SR D Shared resource flow completed. 18:46:50.397 SDMSE:Gateway:Local:SR D Shared resource flow completed. 18:46:51.441 SDMSE:Shizuku:Service:Client:SR D Shared resource flow completed. ```
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Fix services being destroyed before all data has been transferred Similar to
LocalGateway.read(...): Sourcewhere we need to keep the source alive until consumed, we need to do the same when returning aFlow. It's up to the caller on when it's consumed.Likely caused by #1345
Previously we had a constant lease on the Shizuku service resource so this never surfaced: https://github.com/d4rken-org/sdmaid-se/pull/1345/files#diff-3105723dbd44154e23041c05d9496670e61e15cbdcd00ea73c99d1984f577f76R135
Root will have the same issue as that changed in #1346
This fixes #1358