GB . Software . RegressionTest_7_6

Regression Test of M&C 7.6 : October 1, 2007 (00:00 - 11:00)

Jim Braatz

Overview:

These notes describe tests of the M&C 7.6 software on 10/01/07. This was the first full regression test since March. There was time for a thorough set of tests, and everything checked out fine. My tests included a check of the BalanceOnOff procedure Toney developed for balancing the IF system necessary for observing a strong continuum source. It also included HI tests suggested by Paul for automated and regular checks of flux calibration. I ran tests of the standard observing modes at L- and K-bands.

WX: Clear, no wind, ~ 3C

Data: /home/gbtdata/TREG_d071001

Significant items

none

Minor items

cleo issue. I got a popup error as follows: "Error: cleo scan coordinator problem: Possible problems with entered project code Bad username - the user is not in the password file." The Proj ID was "JUNK" before I ran a scan in astrid, so that part is understandable, but the user name part is mysterious. Cleo's scan coordinator checks on the validity of project ID's and warns if they do not follow the rules set down by operations for what constitutes a legal ID. Any ID that starts with a J is used to designate projects that need not be archived. Operators asked that the syntax for J projects be J followed by the user's login name (e.g., JRMADDALE, JBRAATZ, ...)..

GBT Operator Kevin was not seeing astrid plots in his GFM window on startup. An astrid restart on his console fixed the problem.

When running astrid in "monitor only" mode the user will get a message during Peak scans that says "LPC's not updated," but they really are updated, just not by that session. This message led to confusion and could be made clearer.

cleo issue. When running Toney's balance test (the third iteration) I got a cleo popup error saying "SQL Data base error Cannot determine receivers locations couldn't execute "/home/cleoversions/Cleo/utilities/rcvrQuery.sh": no such file or directory" This popup didn't appear during the first couple of iterations of running the same script. The same message appeared a second time during a Peak scan (scan 124). Again on scan 128 (TP Nod). This is a bit worrisome in that it implies a problem with the computing infrastructure and maybe a bit of overzealousness on the part of Astrid. CLEO uses the rcvrQuery.sh script to query the SQL data base to determine what receivers are in what turret locations. The script is executed whenever some one or some process changes the value of the Antenna's gregorianReceiver parameter. The job of the script is to fill in the Antenna screen's menu of what receivers are where. For example, the operator uses the "Greg Rcvr" menu in the Antenna control screen to move a receiver into position. The error message you received indicates that CLEO could not find the script. The location of the script is the same as CLEO itself. Thus, why would the file system on which CLEO resides suddenly disappear? Since it sounds like the operator was not changing receivers, it means that Astrid is altering that gregorianReceiver parameter. My worry is that this may not be a safe thing for Astrid to do since it, not the operator, could rotate the turret inadvertently.

Old items

Astrid popups grab mouse focus, and if the user presses a space bar he may unwittingly respond to the popup in a way that was not intended. For example, a user working in GBTidl may inadvertantly respond to an astrid popup while typing GBTIDL commands.

Starting a new session management window in astrid causes a popup claining the session is no longer authorized, but it is still authorized.

Scan Log for project TREG_d071001

Scan	Source	Description	Comments
			00:00 Mark here for version change
			00:30 Jim B. starting
			L-band
1-5	3C123	AutoPeakFocus
6-10	3C123	AutoPeakFocus	repeat
			Tried to run Toney's BalanceOnOffTest script. I needed to modify the RX in the configuration to L-band, and reset the frequencies. After these mods, the script worked as far as I could tell.
11-15	0502+2516	AutoPeakFocus	didn't specify a source this time
			Slew Test -- ok
			restarted astrid to apply default .sparrow file (I was previously using a custom version that included tip and additional spectrometer plugins)
16 - 22	U11861		This is Paul's integration1 test
23 - 29	U2855		This is Paul's integration2 test
30 - 33	3C147	Peak	fluxcal test
34 - 35	W3OH	Track	L-band, 1 IF, SP
36 - 38	W3OH	Track	L-band, 2 IF, ACS
39	W3OH	Track	Cross pol test
			turn off cross pol with JB_CrossPol-off script
40 - 41	M82	OnOffSameHA	check this against 2007-02-09 regression test result
			K-band
42 - 46	0251+4315	AutoPeakFocus	no source detected
47 - 51	0006+2422	AutoPeakFocus	ok
52 - 56	NGC 23	AutoPeakFocus	ok
57 - 64	NGC 23	Nod	h2o with explicit setting of iftarget=3
65 - 69	NGC 23	AutoPeakFocus	ok
70 - 77	NGC 23	Nod	h2o with default setting of iftarget
			after removing if target from config, the if rack still blanced to the old value
78 - 82	AutoPeakFocus	0503+0203	ok
83 - 90	Nod, 2 IF, K-band	UGC3193	ok
91 -103	"
104	tip	Tip
105 - 109	Mars	AutoPeakFocus
110 - xxx	N1106
136 - 143	M33	RALongMap	ok
144 -	U3789

-- JimBraatz - 28 Sep 2007

Topic RegressionTest_7_6_First . { Edit | Attach | Ref-By | Printable | Diffs | r1.6 | > | r1.5 | > | r1.4 | More }

Revision r1.6 - 01 Oct 2007 - 21:59 GMT - RonMaddalena Parents: SoftwareReportCentral_IntegrationRegressionTestingReports	Content copyright © 1999-2007 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.


NRAO Home > Green Bank \| Wiki Topic: GB > Software > SoftwareReportCentral_IntegrationRegressionTestingReports > RegressionTest_7_6_First

Changes \| Index \| Contents \| Search \| Statistics \| Go