[Isocops] nightly carnage

fair-play at comcast.net fair-play at comcast.net
Fri Jun 18 09:05:25 EDT 2010


Geoff, 


I am to blame for that. I chose 8 because it was the first digit in the series files that didn't have an existing file and I was afraid that using one of the assigned ones would break something. Guess it didn't work out that way. I see digits 0 - 5 defined in the manual but didn't check far enough to see that the higher ones were also used. 


I don't think we want to change a lot for this - I should just change the series number to something that won't cause a problem. I do see that 6 and 7 are currently linked (one of the reasons I thought they might be used). But if it's not a problem to blast one of those I could use one of them. 


Ken 

----- Original Message ----- 
From: "Geoffrey B. Crew" <gbc at space.mit.edu> 
To: "Nathan Schwadron" <nschwadron at me.com> 
Cc: "Ken Fairchild" <fair-play at comcast.net>, isocops at lists.sr.unh.edu 
Sent: Friday, June 18, 2010 8:42:54 AM GMT -05:00 US/Canada Eastern 
Subject: Re: [Isocops] nightly carnage 

He indeed installed a new ephemeris, the problem is that the nightly 
build uses this one. 

Unless code has been changed (which I doubt) only orbit series 
6 and 7 are available for this sort of thing. 

The build testsuite uses series 8 and 9 in a number of places 
(i.e. all the ones that broke last night). 

0..5 are used operationally. (2..5 only when an ephemeris changes, 
so these are briefly available--i.e. between times when FDG gives 
us something.) 

So: 
a) disable all the tests that broke, or 
b) reinstall 8 back to the test ephemeris.... 
c) change the tests to use series 0 
d) other options 

Your choice. 

-- 

Geoff (gbc at space.mit.edu) 

On Fri, Jun 18, 2010 at 08:30:37AM -0400, Nathan Schwadron wrote: 
> Hi Geoff 
> 
> This was a test file for starting to test new potential orbits. We were just verifying that Ken could install new ephemerides. I guess he didn't commit the change; thus the error. 
> 
> Nathan 
> 
> On Jun 18, 2010, at 8:25 AM, Geoffrey B. Crew wrote: 
> 
> > On Fri, Jun 18, 2010 at 08:03:50AM -0400, Geoffrey B. Crew wrote: 
> >>> I say Yea. If someone has a reason for a Nay, say it soon. 
> >> 
> >> Will do, but the nightly carnage needs fixing first. 
> >> 
> >> There are two problems--Paul forgot to commit everything in his 
> >> xhihb upgrade and something bad in spice land which I'll look into. 
> > 
> > On the former, you'll want to commit the offending file and 
> > verify that that directory rebuilds properly. There are a 
> > large number of compiler warning errors which Paul should 
> > be a good do-be and fix. 
> > 
> > On the latter, 
> > 
> > $ l $IBEX_ANC/isoc/IBEX_series* 
> > ... 
> > -rwxrwxrwx 1 fairchild isoc 1112064 Jun 17 13:56 /home/gbc/IBEX/anc/isoc/IBEX_series_8.bsp 
> > 
> > I'm not sure what this is or why it is here, but this is the spice problem. 
> > 
> > Series 8 is reserved for a special testing ephemeris which, now that it is 
> > not what it was yesterday, should indeed break a number of tests.... 
> > 
> > Any idea how that file got there? 
> > 
> > -- 
> > 
> > Geoff (gbc at space.mit.edu) 
> > 
> > Please restrict discussions on this email list to non-ITAR sensitive topics. 
> > ______________________________________________ 
> > Isocops mailing list 
> > Isocops at lists.sr.unh.edu 
> > http://lists.sr.unh.edu/mailman/listinfo/isocops 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.sr.unh.edu/mailman/private/isocops/attachments/20100618/411e27df/attachment.html 


More information about the Isocops mailing list