Ticket #1224 (closed: fixed)
WISH Redhat crashes
Reported by: | Martyn Gigg | Owned by: | Martyn Gigg |
---|---|---|---|
Priority: | major | Milestone: | Iteration 28 |
Component: | Mantid | Keywords: | |
Cc: | Blocked By: | ||
Blocking: | Tester: | Michael Whitty |
Description
Laurent has reported crashes on his Redhat machine in the WISH cabin that happen a lot but not reproducibly.
Again I don't know when exactly it happened, but I had that bug probably 7 or 8 times working with the users yesterday night. It seems to be correlated to: 1- Doing some data processing and generating a number of workspaces 2- Delete them and graphics associated with them 3- reprocess with the same output names Alternatively, sometimes it happens just by replacing workspaces.
Attachments
Change History
comment:3 Changed 10 years ago by Martyn Gigg
(In [4440]) Fix an issue with symbol visibility on machines using ld (linux/mac). Dlopen was not able to find some symbols defined but not referenced in the core Mantid libraries due to the default behaviour for dynamic executables being to only export the symbols referenced at link time. This was causing issues with Group/UnGroup notifications. Re #1224
comment:4 Changed 10 years ago by Nick Draper
- Milestone changed from Iteration 26 to Iteration 27
Bulk move of tickets to iteration 27, if your ticket is essential for Iteration 26 then move it back.
comment:5 Changed 10 years ago by Martyn Gigg
- Status changed from new to accepted
- Component set to Mantid
comment:6 Changed 10 years ago by Martyn Gigg
- Status changed from accepted to verify
- Resolution set to fixed
No reports for quite some time regarding odd crashes so I'll close this.
comment:7 Changed 10 years ago by Brett Tully
- Status changed from verify to verifying
- Tester set to Brett Tully
comment:8 Changed 10 years ago by Brett Tully
- Status changed from verifying to reopened
- Resolution fixed deleted
I modified the reduction test script to loop through the reduction multiple times. While an unusual use of Mantid, it gives a reproducible crash on 1.1.9157 RHEL 5. The two scripts (attached) differ by the number of outer and inner loops conducted.
Run example_reduction_BTloop.py successively:
- Attempt 1: Segmentation Fault on the second run of the script
- Attempt 2: Segmentation Fault on the second run of the script
- Attempt 3: Very slow execution on the second run of the script, normal on the third and Segmentation Fault on the fourth run.
Run example_reduction_BTloop-2.py successively:
- Attempt 1: Segmentation on the fifth run of the script.
comment:9 Changed 10 years ago by Martyn Gigg
- Milestone changed from Iteration 27 to Iteration 28
The example scripts are not a usual way of running the SANS reduction so I'm moving this to iteration 28.
comment:11 Changed 10 years ago by Martyn Gigg
comment:12 Changed 10 years ago by Martyn Gigg
- Status changed from accepted to verify
- Resolution set to fixed
comment:13 Changed 9 years ago by Michael Whitty
- Status changed from verify to verifying
- Tester changed from Brett Tully to Michael Whitty
comment:14 Changed 9 years ago by Michael Whitty
- Status changed from verifying to closed
tested on mantidlx1 v 11062. ran scripts several times without crash.
comment:15 Changed 5 years ago by Stuart Campbell
This ticket has been transferred to github issue 2071
(In [4378]) A fix for CylinderAbsorption that could still try to run if the sample volume was zero, which meant it crashed. Also a couple of improvements to the MantidDock regarding workspace groups could possibly have caused a crash. Re #1224