Task #3125 (closed)
BUG: deleting a Screen fails to delete binary data
| Reported by: | cblackburn | Owned by: | jamoore |
|---|---|---|---|
| Priority: | major | Milestone: | OMERO-Beta4.2.1 |
| Component: | Bin-Services | Keywords: | n.a. |
| Cc: | jamoore, jburel | Remaining Time: | 0.0d |
| Sprint: | 2010-10-28 (18) | Resources: | n.a. |
| References: | n.a. | Referenced By: | n.a. |
Description (last modified by cblackburn) (diff)
Deleting a Screen succeeds but only one Thumbnail and Pixels file (120 of each on disk, numbered 2-121) is deleted. two OriginalFiles? (4 on disk numbered 15-18) are deleted:
From the log it looks like only the last id is being returned per plate.
2010-10-18 10:10:33,663 INFO [ ome.services.blitz.impl.DeleteHandleI] (3-thread-1) Binary delete of OriginalFile for /Screen:1: [15, 16] 2010-10-18 10:10:33,664 INFO [ ome.services.blitz.impl.DeleteHandleI] (3-thread-1) Binary delete of Pixels for /Screen:1: [121] 2010-10-18 10:10:33,665 INFO [ ome.services.blitz.impl.DeleteHandleI] (3-thread-1) Binary delete of Thumbnail for /Screen:1: [31]
Also, there's an exception at the end of the log where one of the OriginalFiles? that has been deleted has an attempt made to update it.
References
| Referenced by: | |||
|---|---|---|---|
|
Attachments
-
Blitz-0.log
(5.4 MB) -
added by cblackburn 3 years ago.
Change History
comment:1 Changed 3 years ago by jmoore
The issue seems to be a run-away depth++ somewhere. There's no way that the graph has depth >300:
2010-10-18 10:10:32,908 DEBUG [ ome.services.delete.DeleteState] (3-thread-1) Enter savepoint 07c0605a606a4b5b9efdbf255902ba98: new depth=364 2010-10-18 10:10:32,910 DEBUG [ ome.services.delete.DeleteState] (3-thread-1) Deleted 4 from Screen/ScreenPlateLink/Plate/Well/WellSample/Image/Pixels/Channel/LogicalChannel: root=1 2010-10-18 10:10:32,911 INFO [ org.perf4j.TimingLogger] (3-thread-1) start[1287393032908] time[3] tag[omero.delete.LogicalChannel.4] 2010-10-18 10:10:32,911 DEBUG [ ome.services.delete.DeleteState] (3-thread-1) Released savepoint 07c0605a606a4b5b9efdbf255902ba98 with 1 ids: new depth=363
comment:2 Changed 3 years ago by jmoore
This is the first savepoint which doesn't get released:
2010-10-18 10:10:25,653 DEBUG [ ome.services.delete.DeleteState] (3-thread-1) Enter savepoint 2f4c12631fc542f2beabe5efcdf0c9f7: new depth=2 2010-10-18 10:10:25,653 DEBUG [ ome.services.delete.DeleteState] (3-thread-1) Enter savepoint 5ddd6fe7397e408c80e07df7cc1830a1: new depth=3 2010-10-18 10:10:25,653 DEBUG [ ome.services.delete.DeleteState] (3-thread-1) Enter savepoint 7a958f9294a94948b463ace41d1e68d0: new depth=4 2010-10-18 10:10:25,662 DEBUG [ ome.services.delete.DeleteState] (3-thread-1) Deleted 1 from Screen/ScreenPlateLink/Plate/Well/WellSample: root=1
comment:3 Changed 3 years ago by cblackburn
The binary delete part of this bug may be more general. See #3130
comment:4 Changed 3 years ago by jmoore
- Status changed from new to assigned
- Owner set to jmoore
- Description modified (diff)
- Remaining Time set to 1
Note: See
TracTickets for help on using
tickets.
You may also have a look at Agilo extensions to the ticket.
