Gentle nudge that you probably want to migrate to the banner_activity refinery module, as mentioned above in T298777#8029475 . The existing R script is not maintainable.

Note to reviewers, here's a mixed map of inline data and external data, which should show only a marker. To make the external data succeed, remove the "foo" prefix.

<mapframe width="300" height="400" zoom="10">
[
  {
    "type": "Feature",
    "properties": {
       "marker-symbol": "town-hall", 
       "marker-color": "46ea5f", 
       "marker-size": "medium"
    },
    "geometry": { 
      "type": "Point",
      "coordinates": [-73.9903, 40.6928] 
    }
  },
  {
    "type": "ExternalData",
    "service": "page",
    "title": "fooNeighbourhoods/New York City.map"
  }
]
</mapframe>

In T321722#8409135, @jijiki wrote:

@awight mind if we close this task and revisit if we are having replication issues?

This looks like a database sync error, and seems to be fixed now. Please reopen if it happens again.

Still broken but in a different way: now the shape is returned empty, https://maps.wikimedia.org/geoshape?getgeojson=1&ids=Q41437865

Confirmed still broken:
https://maps.wikimedia.org/#8/45.066/-81.557

Kicking this out of the Tech Wishes projects, since the stale data on codfw is now resolved, and it looks like a more stable, long-term solution is in place.

Amazing, thank you! I can confirm that the postgres replica sync lag graphs have dropped to 0 for codfw.

In T318058#8393557, @Dzahn wrote:

given the reason for access is stated as "Analysing data together with WMDE/WMF Staff", begs the question who is that staff? Because they would naturally be the sponsor, right?

See also the longer-term fix T290149: Configure replication slots on Postgres masters.

@jcrespo Hi, thank you for the note! We're trying to navigate the process for manager approval, we've asked around a bit but currently there is no sponsor for the work...

In T322353#8391959, @Ladsgroup wrote:

just to be safe add a limit (maybe 1MB?) and if it's larger than that, trigger a statsd metric and avoid storing it for now

In T321789#8391863, @jijiki wrote:

@awight If I understand correctly, we are planning to deploy Karthotherian to Kubernetes using node 16?

Hi! There are discussions happening in email but no news to report, yet. Thanks!

In T322353#8382339, @Ladsgroup wrote:

If I can have some ideas on how many articles will gonna get bigger and how much, I can judge if it's going to break PC or not. Generally I support the idea though

Follow up: "local" gets translated, which is incorrect because the option only works if the exact English keyword is entered. Change the text to:

Use language local to each map area — local

The patch feels pretty close, but I'm having trouble with the basic step of getting it to launch locally so will unassign myself in hope that better-educated devs can take over!