https://prefect.io
Join Slack
```[~] -> docker run -it prefecthq/<server:core-0.14.18> pip show prefect Name: prefect Ver...
# ask-community
s
Copy code
[~] -> docker run -it prefecthq/<server:core-0.14.18> pip show prefect
Name: prefect
Version: 0.14.16+95.g377f2fe7a
holy crap! what is the purpose of the version number in the tag if not the prefect version?
k
Hi @Sven Teresniak ! Sorry about that! Will inform the team about this.
Hey, I looked into this and 
<server:core-0.14.18>
is the last server release tagged with 
core-0.14.18
so when you run 
prefect server start
, we know which image is compatible with the latest core. This is not a prefect core image.
s
Okay, but which official docker image I have to chose to be compatible with (flow) code based on PyPI's 
prefect==0.14.18
? Until now I have a Prefect installation based on 
server:core-$PREFECTVERSION
for towel, an 
apollo:core-$PREFECTVERSION
for apollo, the same tag for 
ui
. For dask-master, dask-worker and agent I have my own images based on 
pip install prefect==$PREFECTVERSION
. Of course the 
$PREFECTVERSION
is the same for all above.
but 
pip install prefect==0.14.18
requires 
click>=7
which in turn installs 
click==8.0.0
. However, the 
sever
image contains code where also 
prefect
(the python package) is installed but in a lower version and with 
click v7.1.2
.
I don't get it. Why is Prefect 0.14.16 part of the 
<server:core-0.14.18>
image?
I'm running Prefect in K8S with a static DaskExecutors
z
Hey @Sven Teresniak -- the server image with 
core-0.14.18
should be compatible with flow code on 0.14.18 even though the version of core on the server image is 0.14.16
The Server's Prefect installation is not used to execute your flows
When a new version of Prefect Core is released, the last released version of Server (which is using Prefect 0.14.16) is tagged as compatible with 0.14.18 so when someone runs 
prefect server start
in 0.14.18 they get a consistent Server setup.
Sorry this is confusing, it's a bit tricky to manage docker containers across multiple repositories and making this a bit more intuitive is on my roadmap.
3 Views
Open in Slack
PreviousNext
Powered by

Prefect Community

Bring your towel and join one of the fastest growing data communities. Welcome to our second-generation open source orchestration platform, a completely rethought approach to dataflow automation.

Powered by