Skip to main content
March 13, 2025
Question

Question about "key-value pair mapping" best practice for Process deployment

  • March 13, 2025
  • 4 replies
  • 0 views

Hi,

I am currently interested about best practices for process deployment, and I've read a thread (please see the link below) about a critical issue.

I've read a reply explaining that "key-value pair mapping" is a best practice for connecting Interface/SAIL with process to avoid deployment issues.

Could you explain me what this best practice is about please ? or just give me a litlle example ?

Deployment issue :
[mention:92d5c8846e574db9a3de2720dbfefcc8:46448885d0e64133bbfbf0cd7b0fd6f7]

Process Best practice :
[mention:d6e94ce985b94acfbda1889c02640965:f586769b0822468ab7f3a94d480ed9b0] 

Regards

4 replies

mathieud0001
March 13, 2025

I think they are referring to passing a map of key/values to your interface which means you can avoid having to remap PVs and facilitates a process upgrade.

I do not really recommend this practice as it can create other issues. If you have short lived processes, this should not be much of an issue. And if you do have some long running processes, I would actually recommend simply starting a new instance from the old instance as an alternative approach.

March 13, 2025

Thank you for your reply

What do you mean by "starting a new instance from the old instance" ?
to manually edit those instances from the instances designer ?

stefanhelzle0001
March 13, 2025

I am not sure whether it makes sense to follow "best practices" from the ancient times. While I have to admit, that I do not understand which exact problem such a "key-value pair mapping" would solve, I use records and have zero issues when deploying.

March 13, 2025

Yes Stefan, you are probably right, it is very old, and I use records too with no issues.But I was just curious about that best practice.
But I maybe should avoid reading these old posts :-)