Outstanding job system work

Issues with * are important for prod release:

[x] * authState should be dynamic so we don't need redirects #2227
[x] * Consider how to implement job_cmd as fixed process #2226
[x] * Test performance of get_simulation_frame vs get_data_file #2225
[ ] Should websocket_free toss exception? #2224
[x] * cookieRedirect is broken #2223
[x] Sbatch: Error PKDict has no attribute 'shifterImage' #2216
[x] simulationFrame requested right before runSimulation #2211
[ ] test.sh needs to run sbatch, docker, local in dev (not build) #2210
[x] Conflict. The container name "snip" is already in use by container #2206
[ ] asynchronous test driver #2204
[ ] run codes in subprocess #2201
[ ] * Report not generated on top of simulation screen #2199
[x] * sbatch: invalid enum "JobRunMode" value for field "jobRunMode" #2198
[ ] job-supervisor db files can be pretty big #2196
[x] job_agent.pid_file should not be removed if not "this" agent #2195
[x] Sbatch: Error getting raw data file #2181
[ ] sbatch needs to interrogate queue #2178
[ ] sbatch run_dirs can be mucked with #2176
[ ] sbatch creds can't change #2175
[ ] sbatchLogin needs to be more secure #2174
[x] Agent start timeout #2171
[ ] NERSC: agents should log hostname #2167
[x] Sbatch: GUI remembers the username and password? #2166
[ ] update GUI on NERSC shifter failure #2165
[x] run parallel simulation not working on alpha #2164
[x] docker unable to open in-json #2163
[x] agent restarts on websocket failure #2162
[x] redirect loop for server upgrade #2160
[x] ignore src user in upgrade #2159
[ ] Remove invalid run dirs #2158
[x] nersc release program #2154
[x] mpi cores #2153
[x] Need driver dependent supervisor_uri #2152
[x] Lag between canceling a sim and all frames coming back #2149
[x] improve job supervisor/agent logging #2135
[x] downloadDataFile fails with not found (404) when cancel #2127
[x] Docker agent idle timeout #2105
[x] job_supervisor tries to find run dir if not db #2101
[x] Job execution speed #2091
[ ] using nginx proxy for remote agent files #2081
[x] look at agent startup time #2080
[ ] remove lib files #2078
[x] start job_supervisor in service.http #2079
[ ] get_data_file reaches back to simulation_db #2065
[ ] When a simulation is stopped for a resource constraint, notify the user #2057
[x] when compute_job != report, minimizing one report, cancels all #1999
[ ] reaching inside other run directories #1950
[ ] serial selected, job run in serial queue #1874
[ ] Elegant bunchCompLSC - lattice time not updating #1865
[ ] elegant: sometimes the simulation progress bar behaves badly #1836
[x] runner_test should assert sirepo versions #1556
[x] Automate multi-host Docker execution #1442
[x] Stop job after timeout #1435
[ ] Process disk use and run-time monitor #1312
[ ] Need structure to server API requests and responses #422
[ ] deleteSimulation should cancel jobs #400
[ ] Server should clear run directories if out of sync with models #343
[x] Run simulation in a particular version container #327
[x] version not saved in simulation run directory #266
[x] Need to query running simulations #60
[x] curl installer radiasoft/download #75
[x] Remove runner #2516

radiasoft / sirepo

Outstanding job system work #2148