Dataflow fail notification bug

LCash

Hi,

Over the last few weeks I've been getting notifications that various dataflows are failing. However, there's no failure metadata and the same version runs successfully moments later. Any idea as to why this is happening? It's not a huge deal but I need to be able to distinguish between true failures that need action and cases such as the below.

jaeW_at_Onyx

is your question "why did the dataflow fail"? or is it "how can i tell if a job that failed actually failed and then seconds later the task reran and it was fine?"

the first one. talk to support. but also it was probably just a hiccup.

the second one. you can't. failure is binary and therefore triggered the email. at the time the first job failed, domo would have no way of knowing that the second attempt would succeed.

you could build different alerts based on "if the dataset hasn't updated in XYZ number of hours or something. but that would be based off of entirely different datasets and NOT focused on whether a job failed.

jaeW_at_Onyx

is your question "why did the dataflow fail"? or is it "how can i tell if a job that failed actually failed and then seconds later the task reran and it was fine?"

the first one. talk to support. but also it was probably just a hiccup.

the second one. you can't. failure is binary and therefore triggered the email. at the time the first job failed, domo would have no way of knowing that the second attempt would succeed.

you could build different alerts based on "if the dataset hasn't updated in XYZ number of hours or something. but that would be based off of entirely different datasets and NOT focused on whether a job failed.

LCash

Thanks @jaeW_at_Onyx. I guess the question was really 'Is this a known bug?' since it's only been happening over the last few weeks. I'll flag with support.