ingestr: CLI tool to copy data between any databases

[michabbb]

Problem Statement

switching from any database to crate is not always easy, tools such https://github.com/bruin-data/ingestr could help a lot also: syncing data to crate is not that easy as well. would be nice to see a tool that is able to sync data from MySQL to crate because at the moment I need tons of bash scripts to make this work (mysqldump -> transform to json -> crash.....)

Possible Solutions

support this project: https://github.com/bruin-data/ingestr 😏

Considered Alternatives

No response

[matriv]

@michabbb Thx for opening this issue, it belongs better to this repo, as it's an integration tool. Please also take a look at: https://github.com/crate/cratedb-flink-jobs where you can use the power of flink to migrate data between many different datasources and CrateDB.

[amotl]

Hi Michael,

thank you for writing in. ingestr indeed looks like an excellent tool, thanks for sharing. Can you elaborate what it would take to make it support CrateDB? We see ingestr might be based on SQLAlchemy, so if all plays well, it might work out of the box already, at least from a connectivity perspective?

With kind regards, Andreas.

[michabbb]

@amotl I don't know the tool, I just saw it today the first time and my very first thought was: where is crate? 😏 So I just wanted to let you know, that's all

[amotl]

All right, thanks. We will look into it!

[michabbb]

@matriv, thanks for the note, but this would force someone to learn Flink and Kafka first, and build the whole infrastructure for it; that´s a monster for someone who sees that the first time, sorry 🙈 so having a tool that is able to sync with only one command line, would be a blessing. that's why I wanted to mention ingestr here 😏

[matriv]

You don't need Kafka if you migrate from let's say MySQL, just two jdbc connections, one for MySQL as a source and one to CrateDB as a sink. Of course, a tool like your suggestion would ease things, I just wanted to point to a current alternative. Thx again for bringing this up.

[amotl]

One variant of our dream tool for conducting effortless data I/O would have a convenient and fluent Python API and command line interface like how ingestr is doing it, with the power of Apache Flink under the hood.

In this spirit, we are occasionally exploring that topic over at those projects, both being descendants or spin-offs of Kotori in one way or another.

• https://github.com/daq-tools/lorrystream
• https://github.com/daq-tools/influxio

As we see @bruin-data is doing an excellent job over there (thanks, @karakanb!), we will surely adjust our efforts, and will certainly start building upon and eventually contributing to ingestr on corresponding "universal I/O" matters, instead of rolling our own adapter/dialect broker.

[karakanb]

Thanks for tagging me folks, and thanks a lot for your kind words on ingestr! ❤️

I didn't know about crate before, but based on what I understand from the repository description if it is PostgreSQL-compatible, that means a regular Postgres ingestion through ingestr can work out-of-the-box with crate as well, I'd encourage you to try it out.

if that doesn't work: ingestr utilizes dlt destinations for ease of use, and we can always develop custom destinations. The better way forward would be -if you'd be willing- to contribute directly to dlt, which means both ingestr, as well as other users of dlt, can benefit from that directly. If, for any reason, contributing to dlt is not an option, I'd happily accept a PR on the ingestr side.

[amotl]

Dear Burak,

thanks for your quick reply, and also for telling us about dlt. We had it on the radar, but now that we learn from you that this is the home of corresponding pipeline elements you are using in ingestr, we will add a focus to improve compatibility with CrateDB, either by contributing, or by adding corresponding documentation in one way or another.

With kind regards, Andreas.