Project

General

Profile

Bug #25251

Retired file / parentage bug

Added by Marc Mengel about 2 months ago.

Status:
New
Priority:
Normal
Assignee:
-
Target version:
-
Start date:
11/25/2020
Due date:
% Done:

0%

Estimated time:
Duration:

Description

So doing parentage queries of the sort that POMS does in datasets which have retired ancestors sometimes gives odd results; example below from gm2; in particular
isparentof: (ischildof:(pred) and pred)
gives a different answer than
isparentof: (snapshot_id x)
where snapshot_id x is a current snapshot of the (ischildof:(pred) and pred)
I think this is due to retired files.

<bel-kwinith>$ samweb count-files "(ischildof:(defname: gm2pro_daq_raw_run3_Prod_N) and offline.requestid 5205A)" 
78142
4:37
<bel-kwinith>$  samweb create-definition mengel-d2-2020-11-20 "ischildof:(defname: gm2pro_daq_raw_run3_Prod_N ) and offline.requestid 5205A" 
Dataset definition 'mengel-d2-2020-11-20' has been created with id 288953
<bel-kwinith>$ samweb take-snapshot mengel-d2-2020-11-20
256873
<bel-kwinith>$ samweb count-files "snapshot_id 256873" 
78142
4:38
okay, so we have the count of the defninition, and the count of the snapshot from the definition and they match.  So far so good.
4:38
Now we get a little fancy:
4:38
<bel-kwinith>$ samweb count-files "isparentof:(ischildof:(defname: gm2pro_daq_raw_run3_Prod_N) and offline.requestid 5205A)" 
78153
4:39
This is the mystery count that makes me think there are weird multiple parents...
4:39
But:
4:39
<bel-kwinith>$ samweb count-files "isparentof:(snapshot_id 256873)" 
78142

4:39
If I use the snapshot, it all goes away.

Also available in: Atom PDF