Period - run matching

Bug #693764 reported by Marina Rejkuba on 2010-12-23
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
SVMT
Undecided
Unassigned

Bug Description

Where is the period vs. run information obtained from?
Each run should be associated with a single period. Observations can be taken over several periods, but the OBs should belong to a given run and a single period. Currently it seems that some runs appear associated with several periods.

Select VHS & OB
Then select Period 84 - there are 1323 OBs
For Period 85 there are 34 OBs
For period 86 there are 850 OBs

Looking at the runs:
Select 179.A-2010(A) - this run has 230 OBs associated with it. If I select it automatically Period 84 is selected (gray) in the summary on top.
Select run 179.A-2010(B) - this run has 1127 OBs. If I select it, there is NO period indicated in the top summary (Constraints).
Select run 179.A-2010(C) - this run has 850 OBs, and if I select it Period 86 is shown on the top summary (Constraints).

It is not clear to me how is the association between Period - Run and OBs made.

Diego (diego-marcos) wrote :

The period and the runID are obtained from the columns "period" and "prog_id" of the phase2_metadata table.

In the case of run ID 179.A-2010(B) some of the OBs have period 84 (1093) and others period 85 (34).

I'm not an expert on the semantics of data but as far as I know the period stored in data base corresponds to the period when the OB was executed that can be different from the period when the OB was scheduled. For the period 179.A-2010(B) 34 of the OBs were executed on period 85 that is what you see on SVMT. You should review the semantics of the data base in order to avoid future confusions.

Diego.

Changed in svmt:
status: New → Invalid

Diego wrote:
> The period and the runID are obtained from the columns "period" and
> "prog_id" of the phase2_metadata table.
>
> In the case of run ID 179.A-2010(B) some of the OBs have period 84
> (1093) and others period 85 (34).
>
> I'm not an expert on the semantics of data but as far as I know the
> period stored in data base corresponds to the period when the OB was
> executed that can be different from the period when the OB was
> scheduled. For the period 179.A-2010(B) 34 of the OBs were executed on
> period 85 that is what you see on SVMT. You should review the semantics
> of the data base in order to avoid future confusions.
>
> Diego.
>
> ** Changed in: svmt
> Status: New => Invalid
>
>

Hmmm,

Interesting, but I don't think that what you suggest is true.
P84 is between October 1, 2009 - March 31, 2010
P85 is between April 1, 2010 - September 30, 2010
P86 is between October 1, 2010 - March 31, 2011

If period corresponds to teh time when an OB was executed, that would
imply that in 6 months between April - September 2010 we only executed
34 OBs, while in the period when we had science verification, and start
of operations (P84) we executed 1093 OBs.
I know for fact that this cannot be true.... so there must be something
else there.
I'll have to check with Adam/Ignacio how is period associated to run and
to OB in the phase2_metadata table.

Best wishes,
Marina

To post a comment you must log in.
This report contains Public information  Edit
Everyone can see this information.

Other bug subscribers