Task #8881 (closed)
Opened 12 years ago
Closed 12 years ago
Bug: Delete timeout
| Reported by: | saloynton | Owned by: | jamoore |
|---|---|---|---|
| Priority: | major | Milestone: | OMERO-4.4 |
| Component: | Services | Version: | n.a. |
| Keywords: | n.a. | Cc: | bpindelski |
| Resources: | n.a. | Referenced By: | n.a. |
| References: | n.a. | Remaining Time: | 0.0d |
| Sprint: | 2012-07-17 (19) |
Description
Created a dataset using the web client. Switched to Insight. Deleting the dataset takes very long. When "Cancel" pressed, the dataset is deleted and removed in both the web client and Insight.
- OMERO.insight-4.3.3-2485-a00641e-ice33-b63-mac
- OS X 10.6.8
Attachments (1)
Change History (16)
Changed 12 years ago by saloynton
comment:1 Changed 12 years ago by jburel
- Sprint changed from 2012-05-22 (15) to 2012-06-05 (16)
comment:2 Changed 12 years ago by jburel
- Sprint changed from 2012-06-05 (16) to 2012-06-19 (17)
Moved from sprint 2012-06-05 (16)
comment:3 Changed 12 years ago by jburel
- Sprint changed from 2012-06-19 (17) to 2012-07-03 (18)
Moved from sprint 2012-06-19 (17)
comment:4 Changed 12 years ago by jburel
- Component changed from Insight to Services
- Owner changed from jburel to jmoore
Could you retest that? It seems to be a server issue.
comment:5 Changed 12 years ago by jmoore
Jean-Marie, I haven't had a chance to evaluate this in more detail, but I did notice that it looks like you are passing instances from the unencryptedEntry (ServiceFactoryPrx) to the callbacks but you are also passing the return value from getClient which is the secure (or "encrypted") client. I don't know specifically what this could cause, but it's probably not a good idea.
comment:6 Changed 12 years ago by jburel
This looks like a problem introduced while adding support for multi-groups.
I will review that.
comment:7 Changed 12 years ago by jburel
883aaf40850af725048dbb57922d97c41a4a3009 Modifies the implementation of the `getClient method.
comment:8 Changed 12 years ago by jburel
I might have to replace the delete call-back.
comment:9 Changed 12 years ago by jmoore
If we need to add an actual callback we can. Alternatively, we can use the CmdCallback. Let me know before you undertake anything too large.
comment:10 Changed 12 years ago by bpindelski
I think I found the ultimate way to trigger this bug:
Start Web and Insight. Log in to both using the same user (here, user-1). Create a dataset in Web, then switch to Insight and refresh the data tree. Dataset will appear. Delete it. Everything should work ok. Go back to Web and click the Refresh icon to reload the data tree.
Go back to Web, create another dataset. Switch to Insight, and again, delete the newly created dataset. Go back to web and instead of clicking the refresh icon, delete the dataset that has already been deleted in Insight. Web will happily carry on, but from now on Insight will be in the "infinite spinner" state.
comment:11 Changed 12 years ago by jmoore
- Remaining Time set to 0.5
- Status changed from new to accepted
Blazej, could you tell me what server you reproduced this on and when? I might be able to tell from the server log along what happened?
comment:12 Changed 12 years ago by jmoore
Jean-Marie, I'm starting to think this isn't related to the callback at all or multiple deletes. I'm now seeing this issue under the eclipse debugger for simple deletes. Is it possible endActivity is never being called?
comment:13 Changed 12 years ago by jburel
- Sprint changed from 2012-07-03 (18) to 2012-07-17 (19)
Moved from sprint 2012-07-03 (18)
comment:14 Changed 12 years ago by bpindelski
Retested with RC2. Seems to be fixed now; can't hit the error when following the outlined procedure.
comment:15 Changed 12 years ago by jmoore
- Remaining Time changed from 0.5 to 0
- Resolution set to fixed
- Status changed from accepted to closed
Closing with the hope that we can merge https://github.com/openmicroscopy/openmicroscopy/pull/216
Moved from sprint 2012-05-22 (15)