Alteryx Server Discussions

Find answers, ask questions, and share expertise about Alteryx Server.

Cancel long running jobs

nraghavak
6 - Meteoroid

Hi,

 

Any jobs running for more than certain number of hours need to identified and sent a notification initially and if they found still running then we plan to cancel those jobs. Also, could be better to know how far the job is completed before a decision is taken whether to continue or to cancel that job

 

I would like to know what is the best way to implement this, been looking at mongodb collection as well as API option. Which one you recommend and does they meet all above criteria?

 

Thanks.

4 REPLIES 4
carlosteixeira
15 - Aurora
15 - Aurora

Hi @nraghavak 

take a look at this link.

 

I created it some time ago. I think this will help you.

 

https://community.alteryx.com/t5/Engine-Works/How-to-Cancel-Manual-Jobs-in-Alteryx-Server/ba-p/90192...

 

Thanks

Carlos A Teixeira
nraghavak
6 - Meteoroid

Hi @carlosteixeira thanks for your solution.

 

Does anyone has tried to check how far the job was completed and if job take longer to finish than expected then reason for this and to get percentage of job completion?

 

Thanks

nraghavak
6 - Meteoroid

Hi all, please share your experience how you handle to know percentage of job completion?. And what is the best way to identify is it waiting for any resources or like into deadlock situation?

 

Thanks.

TomDiroff
Alteryx
Alteryx

Good Day... 2 issues :

1) How to determine a job is running longer than expected

2) How to determine the percentage of the job that is completed

 

1) At a former company we did this by compiling statistics of past runs of each workflow and stored them in a SQL DB. We could then run a workflow that identified the running workflows, compared the current runtime to the average or max or whatever stat you want from the SQL DB and then sent an email to the server admin or Business admin to make a determination if a job should be terminated. This worked great

2) Not currently possible as there is no visibility into a running workflow since  the engine logs are locked and the reason for being at any step of a workflow at any time depends on a number of factors, including those outside of the control of the Alteryx Servers. There are several idea posts to change this but so far have gotten no traction.