Issue Number | 3550 |
---|---|
Summary | [CTGov] CTGov diff report not working |
Created | 2012-10-22 12:13:45 |
Issue Type | Improvement |
Submitted By | Osei-Poku, William (NIH/NCI) [C] |
Assigned To | Kline, Bob (NIH/NCI) [C] |
Status | Closed |
Resolved | 2012-11-07 13:01:11 |
Resolution | Fixed |
Path | /home/bkline/backups/jira/ocecdr/issue.107878 |
BZISSUE::5247
BZDATETIME::2012-10-22 12:13:45
BZCREATOR::William Osei-Poku
BZASSIGNEE::Bob Kline
BZQACONTACT::William Osei-Poku
It appears the CTGovProtocols Imported vs. CWDs report [CIAT/OCCM Staff > CTGov Protocols > 2.CTGovProtocols Imported vs. CWDs] isn't working well. The Summary of changes at the end of the report gives the total processed documents equal to documents with differences but no changes are highlighted.
BZDATETIME::2012-10-22 17:47:59
BZCOMMENTOR::Bob Kline
BZCOMMENT::1
The report job you ran right before filing this issue was for CTGovImport job 3163, which ran Sunday morning. The statistics for the download job which fed into that import job look like this:
!7236 Sun Oct 21 06:38:44 2012: New trials: 0
!7236 Sun Oct 21 06:38:44 2012: Transferred trials: 0
!7236 Sun Oct 21 06:38:44 2012: Updated trials: 0
!7236 Sun Oct 21 06:38:44 2012: Skipped unchanged trials: 25229
!7236 Sun Oct 21 06:38:44 2012: Skipped trials from PDQ/CDR: 3916
!7236 Sun Oct 21 06:38:44 2012: Skipped duplicate trials: 462
!7236 Sun Oct 21 06:38:44 2012: Skipped out of scope trials: 293
!7236 Sun Oct 21 06:38:44 2012: Skipped closed trials: 4476
!7236 Sun Oct 21 06:38:44 2012: Total trials: 34376
!7236 Sun Oct 21 06:38:44 2012: Added NCT IDs for trials: 0
!7236 Sun Oct 21 06:38:44 2012: Removed NCT IDs from trials: 0
!7236 Sun Oct 21 06:38:44 2012: NCT ID problems detected: 0
So it looks like nothing was queued up for import. Not surprising, as I don't think NLM does much work on the weekends. Is this a red herring? If so, can you tell me which job(s) you were reporting on?
BZDATETIME::2012-10-22 18:00:29
BZCOMMENTOR::Bob Kline
BZCOMMENT::2
I believe I found the problem. The recent upgrade of the version control client clobbers the path's location of the real diff utility, replacing it with a broken diff. I modified cdrxdiff.py on Bach to specify the full path location of the version of diff which isn't broken (Cygwin's diff). If you can confirm that this corrects the problem you reported I will propogate this workaround to the other two CDR servers and check the change into version control.
BZDATETIME::2012-10-23 15:27:10
BZCOMMENTOR::William Osei-Poku
BZCOMMENT::3
(In reply to comment #2)
> If you can
> confirm that this corrects the problem you reported I will
propogate this
> workaround to the other two CDR servers and check the change into
version
> control.
Yes. The problem is fixed. Thanks!
BZDATETIME::2012-10-26 09:21:48
BZCOMMENTOR::Bob Kline
BZCOMMENT::4
Fix installed on all three CDR servers.
BZDATETIME::2012-11-07 13:01:11
BZCOMMENTOR::William Osei-Poku
BZCOMMENT::5
(In reply to comment #4)
> Fix installed on all three CDR servers.
Verified. Closing bug. Thank you!
Elapsed: 0:00:00.001636