David Bitner

@drnextgis I think that depends on the use case. If you are using with titiler-pgstac or anything else that is using the query hash to look up a query, clearing...

hey @drnextgis, just want to give a heads up that I have seen this. I was out-of-the-office for a while. I'll be looking at this and a few other things...

@drnextgis I've added an option to the pypgstac loader and a new command in #226 that I think should ameliorate the issue that you are having, while still making sure...

@drnextgis If you are running multiple workers, you should try to ensure that the multiple workers are not accessing the same partitions as the partition maintenance tasks are locking and...

The issue if you get too aggressive with pre-generating partitions is the risk that if you create many empty partitions it can just make the query planning process slower for...

We definitely sacrifice some ingest performance for the sake of read performance. In Postgres, adding additional indexes is not always the best thing -- it adds additional overhead that actually...

Yep, that is correct. I would note that as with all settings, if you set it using `SET pgstac.use_queue='t';` Only queries that were "queued" during that session will be deferred....

You'll need to use the UPDATE statement as @constantinius mentions. I'll get a fix in for any instances that have this exact setting for eo_cloudcover, but I will make sure...

@m-mohr yes, exactly, so you may want to update the definition for eo:cloud_cover on your instance to be the non-referenced schema json rather than the $ref. I think that something...

@hrodmn I'm not sure that those are quite appropriate for the Loader class as the Loader is designed to take a file to ingest it. This is definitely something that...