Tags:
create new tag
view all tags

Code Release 13

Summary

NCR/ECR Brief Description Status Rating Last Test Date
NCR 125 Tmin > Tmax causes reboot 'All-up' test successful thumbs up Low 2006-11-13
NCR 137 No such uvotmode   Low  
NCR 147 Window table load time between exposures not taken into account      
NCR 149 FW move to blocked doesn't check coarse   Low  
NCR 150 All exposures of uvotmode disallowed if first gets disallowed 'All-up' test successful thumbs up High 2006-11-17
NCR 151 "Total Exposure Time Is Zero" error 'All-up' test successful thumbs up   2006-11-16
NCR 152 Memory dumps cause slew aborts see PJS1    
NCR 153 Slewabort doesn't flush the slew timeout 'All-up' test successful thumbs up   2006-11-20
NCR 154 HV task can appear busy when it is finished Combined with NCR 155   2006-11-20
NCR 155 HV stop ramp too late 'All-up' test successful thumbs up   2006-11-20
NCR 156 Reducing exposure time in 0xA Expected feature    
NCR 157 State table problem on 25 Apr 2006. 'All-up' test successful thumbs up   2006-11-20
NCR 158 Starcat CRC Problem. Tested singly - PJS1 smile   May 10 2006
NCR 159 RTS can hang DPU task 'All-up' test successful thumbs up   2006-11-16
NCR 160 Task body RTS_TYPE has delay of 1 s in the accept Cancelled    
NCR 162 Unreported ICB errors 'All-up' test successful thumbs up   2006-11-17
 
ECR 208 Allow 2 lines of settling exposure definitions 'All-up' test successful thumbs up   2006-11-16
ECR 221 If field too bright for FC from AT table, go to next time band in AT table PJS1 'All-up' test successful thumbs up   2006-05-19
ECR 222 If time left after the scaled uvotmode finishes, keep observing in that filter if not too bright 'All-up' test successful thumbs up   2006-11-16
ECR 223 Implement PtNULL 'All-up' test successful thumbs up High 2006-11-14
ECR 224 Scaling doesn't account for "reduced time" exposures Cancelled   2006-06-20
ECR 227 Cat search settled position 'All-up' test successful thumbs up   2006-11-16

Release 13 NCRs

MSSL NCR 125

Heater set command Tmin > Tmax causes reboot.

Testing Notes

Ada code patch

MSSL NCR 137; SERS S-0102, S-0196

No such mode (uvotmode) 0xA reported by UVOT at 2005-04-05T02:33:07. Again with 0x1005 at 2006-01-14T18:34:07.

MSSL NCR 147 Window load overhead

The window table load time between exposures is not taken into account in the damage calculation. Practically, this means that the count rate limit and total count limit should not be set so that count_rate_limit * max_exp_time > total_count_limit. The total count limit should never be allowed to come into action.

MSSL NCR 149 FW move to blocked doesn't check coarse

FW move to blocked doesn't check coarse. Procs might need ifwdatum before /istate safe.

MSSL NCR 150; SERS anomaly S-0200 All exposures of uvotmode disallowed if first gets disallowed

I exposed uvotmode 0x1007 with a fom obs sec of 100 and it did no exposures. If this because the first exposure of the 3 is 10% which is 10s and "under" the min time of 10s? Then the next one should be 80s but doesn't get done either. It seems that the ICU is stopping observing altogether when any exposure is less than 10s rather than trying the next one. This is not what I expected. Not so bad though. Also I believe there is no error message. It just switches to idle.

Testing Notes

Ada code patch

RTS code patch

MSSL NCR 151 Total Exposure Time Is Zero error

2006-02-17T15:35:00 - bug due to a slew_warning coming in and causing a reset of variable total_requested_exp_time back to zero at just the wrong moment! Could have been worse - that variable is checked every exposure time calc for zero values, not just when it's evaluated, and without that it would have caused an Ada exception.

Testing Notes

Ada code patch

MSSL NCR 152; SERS anomalies S-0194, S-0204 Memory dumps cause slew aborts

Increase the delay in the memory dump task so that all the EEPROM dumps in 85 min instead of the current 64 min.

MSSL NCR 153; SERS anomaly S-0204 Slewabort doesn't flush the slew timeout

There is a need to flush the current timeout (maybe make it available to the RTSs). Slewabort should do a dummy waitfor to flush the waitfor slew.

Analysis indicates this is, in fact, a race condition which occurs when a /isslewabort is received just as the rts chains from process_slew_warn to do_slew and effectively leads to the discarding of the /isslewabort .

Testing Notes

Ada code patch

MSSL NCR 154; SERS anomaly S-0202 HV task can appear busy when it is finished

It looks like detanalog task PERFORM_RAMP_TYPE leaves a small but finite time between the DCS.DECLARE_EVENT and the "accept START". This may occasionally leave time for the "select" to return BUSY.

Combined with NCR 155

MSSL NCR 155; SERS anomaly S-0203 HV stop ramp too late

Ihvstopramp was too late. A HV stop ramp (in ICU software) actually stopped the ramp 2.4s after the first of two stops. The RTSs currently allow for only 1s. This is probably just an extreme case. It may be possible to help the code by planning the SAA exposure to be 1 min before to 1 min after the real SAA and to increase the accuracy of the TAKO SAA.

Testing Notes

RTS code patch

MSSL NCR 156 "Reducing exposure time in 0xA"

At 2006-02-28 21:03:02.33 the ICU reported "Reducing Requested Exp Time" in a blocked idle 0xA uvotmode. The observation was interrupted by SAA.

This happens when the code calculating the maximum exposure time is interrupted by the rts abort flag being set. In these circumstances, it returns a failsafe exposure time of zero and this leads to the message.

This should therefore not be classified as a 'bug' but a 'feature' ?

MSSL NCR 157

State table problem on 25 Apr 2006 at 17:44:00.86.

Testing Notes

RTS code patch

MSSL NCR 158

TABLE_START_POINTER and TABLE_END_POINTER in task body STARCAT_TASK are given values too early during task instantiation before STARCAT_OFFSET gets its value in UVOTMAN.INIT. Workaround is to load RAM with the correct numbers after a power on or watchdog reboot.

MSSL NCR 159

RTS can hang DPU task (do many =GET_PT_CONFIG=s).

Testing Notes

Ada code patch

MSSL NCR 160

Task body RTS_TYPE has a delay 1 (1 s) in the accept which may delay the telecommand task (via DCS.RUN_RTS) violating the prime implied requirement that telecommands must be processed in 300 ms. Violating this requirement will cause trouble with ACS, time, slew request and safety telecommands.

Cancelled.

MSSL NCR 162

ERROR_CODE_TEMP in VALIDATE_HV_PARAMS is not initialized so has an indeterminate value (in this case it happened to be B which is Illegal MID). When HV_SET fails with ICB errors when called from TASKMAN, the last value of error_code is used giving a misleading error message. The code should be modified to send ICB_ERROR as the ERR code.

Also: DETSAFETY.CONFIGURE does 2 ICB.PUTs but returns only the value of the second.

ICU_MEM_MANAGER does ICB.PUTs but doesn't care about the "OK" value which isn't so bad because it's only used for engineering and we'll be in contacy be looking for errors realtime anyway.

TMPSU.SET_MOTOR_TEMP_CONTROL gets an OK from the ICB.PUT but always returns TRUE.

TMPSU.SET_HEATER_CONFIG returns TRUE when it should check "OK".

DETANALOG.SET_FLOOD_LED_BIAS_CURRENT returns TRUE when it should check "OK".

Ada code patch

RTS code patch

Testing Notes

Release 13 ECRs

MSSL ECR 208

Allow 2 lines of settling exposure definitions. If the first is too bright and shortened to less than 2s, the code through the next filter but if the allowed exposure time is greater than or equal to 2s in the first filter, it does only that exposure in the first filter. If there are scientific reasons why UVW2 is a better filter, this ECR is not necessary.

Testing Notes

Ada code patch

RTS code patch

MSSL ECR 221

If the field is too bright for the finding chart in the AT table, go to the next time band in the AT table. Alternatively, and probably better, if there's any finding chart in the band, start at V and then try all other filters in order. That goes for the first FC band too which is currently overridden to V.

Testing Notes

Ada code patch

RTS code patch

MSSL ECR 222

UVOTmode exposure scaling assumes the observe_seconds in the fonextobsinfo is perfectly accurate Extend the view time (internally to the ICU) as SAA is sometimes "late". Saves unnecessary filter wheel wear. If there is time left after the scaled uvotmode finishes, keep observing in that filter if not too bright.

Testing Notes

Ada code patch

MSSL ECR 223

Implement PtNULL where the current filter is used.

Testing Notes

Ada code patch

MSSL ECR 224

Scaling doesn't account for "reduced time" exposures

Cancelled.

MSSL ECR 227

Do the catalogue search based on the settled position for AT and PT exposures but not for the settling exposure! For the first finding chart, consider saving time by only doing the search around the settled position if the search around the target position returned "can't use white filter".

Testing Notes

Ada code patch

RTS code patch




Other Items


Documentation

Release the User Manual with ICU s/w release 13

RTS

Add an appendix to the User Manual describing the RTS language and table specification.


KEY

  • On List
  • Notes on work so far
  • Problem encounted requiring investigation or item requiring attention
  • Tested successfully singly smile
  • Completed (on date shown) thumbs up


Last revision: r46 - 2006-11-20 - 12:28:57 - HowardHuckle

Edit | Attach | Watch | Print version | History: r46 < r45 < r44 < r43 < r42 | Backlinks | Raw View | More topic actions
Topic revision: r46 - 2006-11-20 - HowardHuckle
 
This site is powered by the TWiki collaboration platform Powered by PerlCopyright © 2008-2019 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback