PuppetDB now autoThere are automate-promotes into PEable tasks that we currently perform manually for releases and general keeping-the-lights- on merges to FOSS maintenance of the two PuppetDB Git repos. They all have context switching and cognitive-overhead costs, but there's still more the worst of these are the ones like creating merge-ups in the process of maintaining our LTS branch(es). This requires a few minutes of work to do w create the merge-up, then 20-30 minutes of tests running, followed by a few more minutes to commit the merge-up.r If you can fill that interim period with something productive, the cost is relatively low, but if the cost of context switching is high enough that can easily become a task that "costs" 40 minutes.t improving our CI pipelines.
There are automate-able tasks that we currently perform manually for releases and general keeping-the-lights-on maintenance of the two PuppetDB Git repos. They all have context switching and cognitive-overhead costs, but the worst of these are the ones like creating merge-ups in the process of maintaining our LTS branch(es). This requires a few minutes of work to create the merge-up, then 20-30 minutes of tests running, followed by a few more minutes to commit the merge-up. If you can fill that interim period with something productive, the cost is relatively low, but if the cost of context switching is high enough that can easily become a task that "costs" 40 minutes.
PDB-4520 is different that the other 3 4 tickets, it's just wrong behaviour that could result in us merging broken code.
There are automate-able tasks that we currently perform manually for releases and general keeping-the-lights-on maintenance of the two PuppetDB Git repos. They all have context switching and cognitive-overhead costs, but the worst of these are the ones like creating merge-ups in the process of maintaining our LTS branch(es). This requires a few minutes of work to create the merge-up, then 20-30 minutes of tests running, followed by a few more minutes to commit the merge-up. If you can fill that interim period with something productive, the cost is relatively low, but if the cost of context switching is high enough that can easily become a task that "costs" 40 minutes.
PDB-4520 is different that than the other 4 tickets, it's just wrong behaviour that could result in us merging broken code.