Tasq

![tasq](Tasq) is the best way to simply keep track of your ![task](tasks) and monitor their progress. The same interface is also present in ![render](Render).

Here is how you can monitor your tasks.

Task status

The Tasks tab displays a list of all your tasks and prints some information about them. You can change the way they are sorted.

Task Table

Among the most relevant information are Sate and Execution time. The latter is the sum of the execution time on each ![instance](instance), e.g. the time that will finally be invoiced. The former contains the status of the task. Here are the possible values and their meaning:

Submitted indicates that the task was submitted and is yet to be processed by the orchestrator.
Fully dispatched indicates that all the instances have been dispatched to a physical machine. They are now booting and getting ready for your computations.
Partially dispatched means that only part of the instances has been dispatched.
{{alert type="info" content="This may be due to the fact that you have reached your running instances limit."}}
Fully executing means that all the instances have effectively started their computations. The percentage displayed is the number of instances that have finished.
{{alert type="info" content="most of the time it isn't a measure of progress. Indeed, if all the instances have similar loads, or if you have only one task, all the instances should start and finish at the same time so it will jump from 0% to 100%."}}
Partially executing means that some instances have started while some are still waiting to be dispatched.
Downloading results signifies that the task has finished and is now downloading all its results files to your ![bucket](bucket).
Other values are Pending cancel, Cancelled, Success, Failure, Pending delete and they mean exactly what you think they mean.
{{alert type="info" content="However Success only means that the process finished normally on Qarnot's side of things. The program you started may have failed anyway. If needed, see the #[/debugging-your-payloads/fetching-logs](Debugging your payloads) section."}}

By clicking on an element of the list, you can have very useful details about the task. In addition to general information such as the task's UUID, the ![profile](profile), ![constant](constants) and links toward your ![bucket](buckets):

Task Details

Logs

The central pane displays the task's standard output on the 1st tab and its standard error on the 2nd tab.

{{alert type="info" content="These are reported in a best effort. They may not be exact, and sometimes the API can't follow the output, so [...] may appear. As explained in the #[/debugging-your-payloads/fetching-logs](Fetching logs) page, you may want to log all the outputs into a file, and optionally use snapshots to follow them and make sure you get all the logs."}}

The 3rd tab, JSON, shows the raw answer given by the ![api](API). The format isn't really user-friendly, yet the syntax is rather straight forward. Some interesting information can be retrieved. See the next paragraph to learn more about them.

Keeping track of your computations with the web UI

Any feedback on this page ?