[Update] Multi-node and distributed hypertables #492
Comments
|
Let's talk about the "current status and limitation", as well as how we plan to include. I do worry a bit about that sometimes we try to be very complete, so we end up writing a lot and users end up not reading because it's just too length / in the weeds. |
|
For "Add "quick start" for setting up a new postgres user", I'm not quite sure what that's supposed to encompass vs. the overall "Create new "quick-start / quick configure" guide to setting up multi-node" changes to |
|
@solugebefola I'll just review when you are done. The old instructions just had a /lot/ of details about adding new databases users (and then propagating user creation to DNs as well), certification authentication approach, etc. Was thinking there was a way to just show one simplified perferred approach, rather than multiple. But let's review your draft and see what's needed. |
|
|
bump? Seems like there is only "introduction" changes to add? |
|
@mfreed close? |
New Intro / Getting Started
New Using Distributed Hypertables [@erimatnor]
https://docs.timescale.com/beta-v2.0.0/using-timescaledb/distributed-hypertables
Current Status and Limitations
compress_chunk, no policies) [@gayyappan ]API docs (@pmwkaa)
create_distributed_hypertable[ @k-rus ] [ Improve create_distributed_hypertable #533 ]add_data_node- Improper layout as syntax/arguments should be at top [ @pmwkaa ] [ Update add_data_node() API description #541 ]delete_data_node(no changes required)attach_data_node(no changes required)detach_data_node[ @pmwkaa ][ Add if_attached argument to detach_data_node() api #529 ]distributed_exec[ @k-rus ] [ Fix distributed_exec to correspond to the implementation #569 ]set_replication_factorhttps://docs.timescale.com/beta-v2.0.0/api#set_replication_factor [ @k-rus ]allow_new_chunks(was moved to internal namespace)block_new_chunks(was moved to internal namespace)Add label/badge to all multi-node functionsMisc
timescaledb.remote_data_fetchercan be set torow-by-rowandcursor, where the former is the default since the latter doesn't support parallel execution on the DN. [ @mkindahl @ryanbooz ]The text was updated successfully, but these errors were encountered: