TODO

Version 49 (Etienne Pallier, 09/23/2016 06:28 pm)

1 1 Paul Carensac
h1. TODO
2 1 Paul Carensac
3 1 Paul Carensac
List of tasks by application, and general tasks (organization, tools, ...)
4 1 Paul Carensac
5 1 Paul Carensac
{{>toc}}
6 1 Paul Carensac
7 1 Paul Carensac
---
8 1 Paul Carensac
9 1 Paul Carensac
h2. %{margin-left:0px; font-weight:bold; font-size:25px;  display:block; color:red;}General tasks%
10 1 Paul Carensac
11 43 Etienne Pallier
* Observation_manager/Tasks : créer une classe mère héritant de Task, dont les classes filles (plan_vis et plan_ir) vont hériter
12 3 Etienne Pallier
13 39 Etienne Pallier
* Architecture projet (organisation dossiers):
14 39 Etienne Pallier
15 39 Etienne Pallier
 * src/ : laisser seulement les applications, et mettre tous les autres dossiers (fixtures/, images/, ...) dans un dossier MISC/)
16 39 Etienne Pallier
17 44 Etienne Pallier
* RUN :
18 44 Etienne Pallier
19 44 Etienne Pallier
 * S'assurer que tous les prérequis sont présents et démarrés avant de lancer Pyros (sinon, générer un message d'erreur expliquant le problème):
20 44 Etienne Pallier
 
21 44 Etienne Pallier
  * RabbitMQ
22 44 Etienne Pallier
  * Mysql (sauf si on a choisi sqlite)
23 44 Etienne Pallier
24 44 Etienne Pallier
25 44 Etienne Pallier
26 45 Etienne Pallier
* TESTS (intégration continue):
27 1 Paul Carensac
28 45 Etienne Pallier
 * Même remarque que pour le "RUN" ci-dessous concernant le check des prérequis
29 38 Etienne Pallier
 * Inclure "trial comet" dans les tests pour s'assurer que Comet est bien installé (+ "rm -r _trial_temp/" ensuite pour faire le ménage après le test)
30 37 Etienne Pallier
 * Mettre en place une exécution systématique des tests à chaque commit (unitaires + fonctionnels)
31 36 Etienne Pallier
(pour faire plus simple, on peut imaginer que le serveur de test linux fasse un "git pull" toutes les 30mn, puis une exécution des tests, et l'envoi d'un email en cas de problème)
32 36 Etienne Pallier
33 35 Etienne Pallier
34 35 Etienne Pallier
* Installation script:
35 35 Etienne Pallier
36 35 Etienne Pallier
 * Si l'environnement virtuel existe déjà, demander s'il doit être recréé (par défaut "Non")
37 48 Etienne Pallier
 * (NEW) Renommer le script install_requirements.sh en install.sh
38 48 Etienne Pallier
 * (NEW) Faciliter l'installation du logiciel complet avec CONDA, qui gère toutes les dépendances du logiciel, même les dépendances non Python (mysql, rabbitmq, ...), et c'est du multi-plateforme : (cf https://jakevdp.github.io/blog/2016/08/25/conda-myths-and-misconceptions)
39 35 Etienne Pallier
40 25 Etienne Pallier
* Pyros : 
41 1 Paul Carensac
42 30 Etienne Pallier
 * Doit pouvoir être démarré indifféremment AVANT (par défaut) les devices, ou APRES (actuellement, le AVANT est TODO)
43 25 Etienne Pallier
 * De manière générale, doit être le plus possible "tolérant aux pannes"
44 25 Etienne Pallier
 * Doit être le plus générique et donc paramétrable possible
45 1 Paul Carensac
46 25 Etienne Pallier
47 23 Etienne Pallier
* Django : 
48 12 Etienne Pallier
49 11 Etienne Pallier
 * Upgrade to Django 1.10
50 41 Etienne Pallier
 * Remplacer le serveur web de "dev" (manage runserver, sur port 8000) par un vrai serveur web pour la "prod" (Apache pour les fichiers statiques sur port 80 + serveur d'application Python pour le code Python, par exemple gunicorn)
51 11 Etienne Pallier
(cf https://projects.irap.omp.eu/projects/pyros/wiki/Project_Development#Serveur-Web)
52 23 Etienne Pallier
53 20 Etienne Pallier
* Comet installation : Include in the install_requirements.sh script (by using stable official version for Python 3 (when available) on Linux (and Mac) and Windows 10
54 1 Paul Carensac
55 1 Paul Carensac
* git : Create a "dev" branch (do not write anymore on the "master" branch)
56 1 Paul Carensac
57 25 Etienne Pallier
* Simulators : 
58 1 Paul Carensac
59 25 Etienne Pallier
 * on doit se rendre compte qu'ils sont "vivants" (ils seront progressivement remplacés par les vrais devices, mais resteront toujours utilisables à leur place)
60 25 Etienne Pallier
61 25 Etienne Pallier
  * Le Monitoring doit les interroger régulièrement sur leur statut (check_status, DONE)
62 25 Etienne Pallier
  * Il doit stocker les statuts dans la BD (TODO)
63 25 Etienne Pallier
  * Ces statuts doivent être affichés au fur et à mesure sur la page web du "device" correspondant (Devices, Site, Weather) (TODO)
64 25 Etienne Pallier
65 25 Etienne Pallier
 * quels sont les devices bloquants et non bloquants ?
66 25 Etienne Pallier
67 27 Etienne Pallier
  * Bloquants: si indisponibles (en panne ou pas démarrés), Pyros met en pause le sous-système d'observation
68 26 Etienne Pallier
Seules 3 fonctions restent toujours actives: alertes, monitoring, et serveur web (pour les routines et la surveillance)
69 25 Etienne Pallier
70 25 Etienne Pallier
   * Telescope
71 25 Etienne Pallier
   * Camera VIS (???)
72 28 Etienne Pallier
   * PLC (weather + observation conditions + site)
73 25 Etienne Pallier
74 27 Etienne Pallier
  * Non bloquants: si indisponibles (en panne ou pas démarrés), Pyros se contente d'en prendre note, et continue son fonctionnement normal tout en restant dans l'attente de leur redémarrage pour les réintégrer
75 25 Etienne Pallier
76 25 Etienne Pallier
   * Camera VIS (???)
77 25 Etienne Pallier
   * Camera NIR
78 25 Etienne Pallier
79 23 Etienne Pallier
80 22 Etienne Pallier
* Users: gérer les 3 profils (admin, expert, user)
81 23 Etienne Pallier
82 49 Etienne Pallier
* Scheduler : le rendre complètement autonome, indépendant du projet, afin qu'on puisse facilement le sortir du projet, et le proposer comme un composant générique utilisable dans n'importe quel autre logiciel ; il devra donc gérer ces communications avec les autres modules via des interfaces
83 49 Etienne Pallier
84 14 Etienne Pallier
* Doc :
85 17 Etienne Pallier
86 14 Etienne Pallier
 * TEST:
87 14 Etienne Pallier
  
88 31 Etienne Pallier
  * Unit tests : test of each module
89 31 Etienne Pallier
  * Functional tests (with Celery and simulators): complete workflow
90 14 Etienne Pallier
91 14 Etienne Pallier
 * LAUNCH:
92 14 Etienne Pallier
93 14 Etienne Pallier
  * Start simulators
94 14 Etienne Pallier
  * Start Celery
95 14 Etienne Pallier
  * Start pyros
96 14 Etienne Pallier
97 14 Etienne Pallier
 * USE:
98 14 Etienne Pallier
99 15 Etienne Pallier
  * Administration of the database: http://localhost:8000/admin
100 1 Paul Carensac
  * Interacting with Pyros: http://localhost:8000
101 15 Etienne Pallier
  
102 15 Etienne Pallier
   * Watch the environment: Devices, Site, Weather
103 15 Etienne Pallier
   * Watch the schedule: Schedule
104 15 Etienne Pallier
   * Watch the data processing (workflow): System (Dashboard)
105 16 Etienne Pallier
   * Submit a Routine Request (and get results): Routines
106 21 Etienne Pallier
   * Watch alerts: Alerts
107 15 Etienne Pallier
   * Simulate an Alert: Alerts (TODO: Un user "admin" doit pouvoir déclencher une alerte type depuis la page web Alerts)
108 14 Etienne Pallier
   * Manual operations on the Telescope : Devices/Telescope (TODO)
109 1 Paul Carensac
   * Manage users : Users
110 22 Etienne Pallier
111 14 Etienne Pallier
112 1 Paul Carensac
113 1 Paul Carensac
---
114 1 Paul Carensac
115 2 Paul Carensac
h2. %{margin-left:0px; font-weight:bold; font-size:25px;  display:block; color:red;}Applications tasks%
116 2 Paul Carensac
117 2 Paul Carensac
h3. Dashboard
118 2 Paul Carensac
119 2 Paul Carensac
 * Create the backoffice views as the modules are integrated in pyros
120 2 Paul Carensac
 * Think about a system of permissions
121 2 Paul Carensac
122 2 Paul Carensac
h3. Scheduler
123 2 Paul Carensac
124 2 Paul Carensac
 * views
125 2 Paul Carensac
126 2 Paul Carensac
  * Link the main page to the current schedule instead of the simulation page
127 2 Paul Carensac
  * Show user sequences in the schedules (with links)
128 2 Paul Carensac
  * Give accces to old schedules (day / days before / all end-night plannings / all plannings)
129 2 Paul Carensac
  * Give access to the refused sequences of a schedule, and the reasons of rejects
130 2 Paul Carensac
131 2 Paul Carensac
 * scheduler
132 2 Paul Carensac
133 46 Jeremy Barneron
  * Verification when there are not fields in the scheduler tables (index out of range)
134 2 Paul Carensac
  * Change the system to determine night start/end (they must be given in parameter, only if first_schedule is True)
135 2 Paul Carensac
  * Store the reasons of rejects (create a new attribute, in shs ?)
136 2 Paul Carensac
  * What is the 'flag' attribute ? (@AK)
137 2 Paul Carensac
  * Do not create the execute_sequence tasks if it's not the night (- 120 seconds)
138 2 Paul Carensac
  * Priority and quota computing
139 2 Paul Carensac
  * Quotas evolution
140 2 Paul Carensac
  * Blank space filling
141 2 Paul Carensac
  * At the end of a scheduling send, it to the IC ?
142 2 Paul Carensac
143 2 Paul Carensac
h3. Alert Manager
144 2 Paul Carensac
145 2 Paul Carensac
 * Web :
146 2 Paul Carensac
147 2 Paul Carensac
  * Print if there is an alert in progress in the main page
148 2 Paul Carensac
  * Link the alerts to their status and results
149 2 Paul Carensac
150 2 Paul Carensac
 * Connect to a real VOEvent broker
151 2 Paul Carensac
152 2 Paul Carensac
 * Determine the communication with FSC for strategy change
153 2 Paul Carensac
154 2 Paul Carensac
 * VOEvents :
155 2 Paul Carensac
156 2 Paul Carensac
  * Extract the good fields (see AK Q&A 07/01/2016)
157 2 Paul Carensac
  * Fill the request & alerts objects
158 2 Paul Carensac
  * Use strategies to build a request
159 2 Paul Carensac
  * Possibility to change the default strategy
160 2 Paul Carensac
  * Handle VOEvent updates
161 2 Paul Carensac
  * Be careful to not create 2 alerts for a same GRB, seen by 2 different satellites 
162 2 Paul Carensac
163 2 Paul Carensac
164 2 Paul Carensac
h3. Analyzer
165 2 Paul Carensac
166 2 Paul Carensac
 * Apply the calibrations in the right function
167 2 Paul Carensac
 * Apply the analyses only if it's a GRB
168 2 Paul Carensac
 * Implement the analyses
169 2 Paul Carensac
 * Send analyses to FSC
170 2 Paul Carensac
171 2 Paul Carensac
172 2 Paul Carensac
h3. Majordome
173 2 Paul Carensac
174 2 Paul Carensac
 * TaskManager
175 2 Paul Carensac
176 2 Paul Carensac
  * When a sequence is cancelled, give back the quota to the user
177 2 Paul Carensac
  * In case of alert, do not stop the ongoing plan, and make the instruments abort
178 2 Paul Carensac
179 2 Paul Carensac
 * execute_sequence
180 2 Paul Carensac
181 2 Paul Carensac
  * Add the PLC checks at start (to see if we do the slew)
182 2 Paul Carensac
  * Use the global telescope (instead of creating one here)
183 2 Paul Carensac
  * Give first_schedule as false when a scheduling is launched
184 2 Paul Carensac
  * Remove the default countdown (1, for tests)
185 2 Paul Carensac
186 2 Paul Carensac
 * system_pause
187 2 Paul Carensac
188 2 Paul Carensac
  * Abort the isntruments
189 2 Paul Carensac
  * Stop the execution tasks
190 2 Paul Carensac
191 2 Paul Carensac
 * system_restart
192 2 Paul Carensac
193 2 Paul Carensac
  * Start a scheduling
194 2 Paul Carensac
195 2 Paul Carensac
 * change_obs_conditions
196 2 Paul Carensac
197 2 Paul Carensac
  * Change sequences status (if  needed)
198 2 Paul Carensac
  * If some status changed, re-launch a scheduling
199 2 Paul Carensac
200 2 Paul Carensac
h3. Monitoring
201 2 Paul Carensac
202 2 Paul Carensac
 * views
203 2 Paul Carensac
204 2 Paul Carensac
  * Move the dashboard here
205 2 Paul Carensac
  * Print the instrument status
206 2 Paul Carensac
  * Print PLC informations (with the evolution)
207 2 Paul Carensac
  * In the dashboard screens, put scroll on each screen to see the old logs
208 2 Paul Carensac
209 2 Paul Carensac
 * Monitoring task
210 2 Paul Carensac
 
211 2 Paul Carensac
  * Uncomment the scheduling at the beginning
212 2 Paul Carensac
  * Implement night start/end computation
213 2 Paul Carensac
  * Initialize communication with the instruments
214 2 Paul Carensac
  * Configure intruments at start
215 2 Paul Carensac
  * Send software versions to the IC
216 2 Paul Carensac
  * Initialize connection with PLC
217 2 Paul Carensac
  * After the starting actions, loop to wait for the instruments configuration to be finished
218 2 Paul Carensac
  * Ask PLC for status
219 2 Paul Carensac
  * Ask filter wheel for status
220 2 Paul Carensac
  * Store the instruments & PLC status
221 2 Paul Carensac
  * Send all status to IC
222 2 Paul Carensac
  * Analyse PLC status (obs conditions, ...)
223 1 Paul Carensac
  * Create tasks of obs condition changes
224 33 Paul Carensac
  * At the beginning of the loop, ping the DISABLED devices. If they answer, change their status (TBD), and do their initialization (ex: If it's a camera, set its temperature to -150°)
225 2 Paul Carensac
226 2 Paul Carensac
h3. Observation Manager
227 2 Paul Carensac
228 2 Paul Carensac
 * execute_plan :
229 2 Paul Carensac
230 2 Paul Carensac
  * Uncomment the instruments_ready waiting function
231 1 Paul Carensac
  * Uncomment the observation_ending waiting function
232 42 Paul Carensac
  * Try to remove code duplication (plan_execution_vis & plan_execution_nir)
233 2 Paul Carensac
  * Determine what needs to be done at the end of an observation
234 2 Paul Carensac
235 2 Paul Carensac
 * create_calibrations :
236 2 Paul Carensac
237 2 Paul Carensac
  * Make the calibration images
238 2 Paul Carensac
  * Generate super images
239 2 Paul Carensac
  * Send them to the IC
240 2 Paul Carensac
241 2 Paul Carensac
h3. Routine Manager
242 2 Paul Carensac
243 2 Paul Carensac
 * Web
244 2 Paul Carensac
245 2 Paul Carensac
  * Put the goods fields (for coordinates etc)
246 2 Paul Carensac
  * Only propose the objects that matches the conditions (ex: scientific programs of the user only)
247 2 Paul Carensac
  * Do all the needed checks
248 2 Paul Carensac
  * Add automatic computation of JD1/JD2
249 2 Paul Carensac
  * Add checkbox for JD / GD
250 2 Paul Carensac
  * Add options : copy my sequence on x days, and authorise report
251 2 Paul Carensac
  * Add ETC-IS simulation
252 2 Paul Carensac
  * Add help for new users (and for it the first time an account come on the page)
253 2 Paul Carensac
 
254 2 Paul Carensac
 * Do more checks at unserialization
255 2 Paul Carensac
256 2 Paul Carensac
 * views
257 2 Paul Carensac
258 2 Paul Carensac
  * When saving, do more checks on coordinates, jd1/2 etc
259 2 Paul Carensac
  * Uncomment filter for alerts removing
260 2 Paul Carensac
  * When submitting, use the monitoring to determine sequences status
261 2 Paul Carensac
  * When submitting, modify the first_schedule to False, when scheduling
262 2 Paul Carensac
  * When unsubmitting, uncomment the check for EXED and EXING removing
263 2 Paul Carensac
  * When unsubmitting, uncomment the scheduling and change the first_schedule to False
264 2 Paul Carensac
265 2 Paul Carensac
266 2 Paul Carensac
h3. User Manager
267 2 Paul Carensac
268 2 Paul Carensac
 * Password recovery
269 2 Paul Carensac
 * Profile page
270 2 Paul Carensac
 * User validation by administrator / commission
271 2 Paul Carensac
 * Handle permissions and access
272 2 Paul Carensac
273 2 Paul Carensac
h3. Common
274 2 Paul Carensac
275 1 Paul Carensac
 * Change the 'first_schedule' to False at the end of RequestBuilder.validate()
276 32 Paul Carensac
277 32 Paul Carensac
h3. Devices
278 32 Paul Carensac
279 34 Paul Carensac
 * When a socket connection fails (exception thrown), set the device to DISABLED (or that kind of status) in the DB, and create a system_pause task (majordome)