Describe the endofnight script: how is it run, maitained, etc
Describe the checks it performs and summary statistics it reports.
Describe the viewer at https://data.desi.lbl.gov/desi/users/dkirkby/endofnight/
Describe the directory layout at nersc...
JPEG images showing the focal plane at the end of the night, stored as fvc-back-YYYYMMDD.jpg and fvc-front-YYYYMMDD.jpg. These images are captured by the end-of-night park procedure performed by the lead observer, then processed to remove instrument signatures and highlight relevant details. There are two images: a back-illuminated image where only the fiber tips are visible, and a front-illuminated image where the focal plane itself is visible.
Stored as calib-YYYYMMDD.csv and derived from any calibration rows added during this observing night (but many nights do not generate any new rows). See the calibration database schema for details.
Stored as moves-YYYYMMDD.csv.gz and derived from the rows added to the move table during this observing night. Many of these columns are directly copied from the moves database, which is documented here.
| Column | Unit | Description |
|---|---|---|
| time_recorded | hr | Hours relative to local noon before this observing night |
| pos_id | - | Positioner ID string, e.g. M01234 |
| location | - | 1000 * PETAL_LOC + DEVICE_LOC |
| ctrl_enabled | - | 0 = disabled, 1 = enabled (non-functional devices are always disabled) |
| move_cmd | - | String describing the move performed |
| log_note | - | Log note attached to this move |
| exposure_id | - | Unique exposure identifier |
| exposure_iter | - | Uteration within this exposure (0=blind, 1=correction) |
| mflags | - | Bitmask of status flags documented here (renamed to avoid collision with the flags method) |
| pos_t,p | deg | Internal theta and phi angles recorded in the moves database |
| req_t,p | deg | Requested angles stored in the log_note as "req_posintTP=(...,...)" if available or NaN |
| fvc_t,p | deg | Angles derived from an FVC spot if available or NaN. Uses an immediately following FVC feedback if present. |
| req_dt,dp | deg | Requested change of angles obtained as the sum of the angles appearing in the move_val1,2 text columns |
| act_dt,dp | deg | Actual change of angles obtained as the change in fvc_t,p from the previous to current row |
| ptl_x,y | mm | PTL coords from the moves DB converted to FP coords with nominal petal alignments |
| obs_x,y | mm | FP coords from the moves DB, derived from FVC spots |
| req_x,y | mm | Requested petal coords from log note converted to FP coords with nominal petal alignments |
| pred_x,y | mm | Predicted FP coords obtained by transforming pos_t,p using desimeter and nominal petal alignments |
| blocked | - | Log note and flags indicate that this move was blocked by a comms error or to avoid a collision |
Notes:
- obs-ptl includes effects of petal misalignments implemented in PlateMaker (since ptl_x,y are calculated for nominal alignments)
- obs-req is due to positioning errors and turbulence (or just turbulence for non-functional robots)
- act/req != 1 indicates a potential motor problem or collision
- fvc_t,p will be NaN for the blind (exposure_iter=1) positioning performed before each cosmic split since there is no accompanying FVC image.
Stored as fp-YYYYMMDD.ecsv and...