Hi Dev Team,
I was running a simple hello world submission (submission_id= 7817048) using both preprocessing and training in one image. I didn't receive any email about the submission and the log file was only available in the submission table after I terminate the submission. Seems like the prepreocessing has finished but training was not yet started.
I am wondering in this case, what's the best way to monitor my submission. Did I do something wrong that makes my submission stuck?
In addition, I received results in 5 mins after submitted through the Training Express Lane. Am I supposed to use the express lane for all my training?
Thank you for your time!
Created by Yiqiu Shen ashen Thanks Thomas! Your explanation is very helpful! :) Dear Yiqui,
Apologies for the delay in response.
> I am wondering in this case, what's the best way to monitor my submission.
The best way is the dashboard in Sect. 3.6 (scroll down to the table in the "Time Quota" section). (For the express lane it's Sect. 3.6.1.)
>Did I do something wrong that makes my submission stuck?
No.
>In addition, I received results in 5 mins after submitted through the Training Express Lane. Am I supposed to use the express lane for all my training?
No. The 'express lane' only gives you a quick indication of whether your code is correct error free. The a actual computational results are not meaningful.
>Any advances on the creation of the remaining dashboard pages?
That is still pending.
For a quick test of end-to-end functionality, the express lane is the right tool to use. If you don't see any progress on your submissions, then there is probably a problem with the system.
The most definitive way to check for a problem in the system is to look at the "Last Updated" column in the dashboard. It should always show a time stamp that's just a minute or so old. If it hasn't been updated in hours then our server agent has stopped. (We have an open issue to create an alarm based on such a check. The server agent already alerts us by sending an email when it's about to quit due to an anomalous condition.)
Best,
Thomas
Drop files to upload
Submission stucked (11 hours) on EVALUATION_IN_PROGRESS page is loading…