The current task view is not compatible anymore with the dynamic pipeline. https://alyx.internationalbrainlab.org/admin-tasks/status. We have most of the monitoring already in place in the json fields of the data repository and lab locations tables.
A general rule is to find views that give the magnitude of a given problem is useful for on-call.
Monitoring: get more information about session creation / task creation issues: register the logs in some way or form
Overview per server:
space left, daemon status, ibllib version, last activity
last session (with link to the reports page)
Overview per rig:
json field of the lab location
Task table per server:
for a time elapsed since now (say 3 months)
each row is a task type
columns: number of tasks in each status / proportion
Replicate tasks recap table but without pipeline columns and one local server per row + number of complete tasks over the last three months (n/N complete) + number of sessions with no tasks
Table showing last 5 tasks run on a lab server, with 1st row the last started
Table per task name showing only errored tasks, grouped by task contain line that starts with XError: ...
The current task view is not compatible anymore with the dynamic pipeline. https://alyx.internationalbrainlab.org/admin-tasks/status. We have most of the monitoring already in place in the json fields of the data repository and lab locations tables.
A general rule is to find views that give the magnitude of a given problem is useful for on-call.
Monitoring: get more information about session creation / task creation issues: register the logs in some way or form
Overview per server: space left, daemon status, ibllib version, last activity last session (with link to the reports page)
Overview per rig: json field of the lab location
Task table per server: for a time elapsed since now (say 3 months) each row is a task type columns: number of tasks in each status / proportion