Kostas Dermentzis

I'm pretty sure these queries happen only during rollback. Rollbacks are not that frequent and spending a few seconds for them should be fine.

The `new_committee` table has an entry for this. Fyi there is a typo there `nominator` instead of `numerator`

> We're running some queries which result in Postgres crashing. I would need more information to help here. What kind of queries and why they consume so much memory.

As the draft release docs say https://github.com/input-output-hk/cardano-db-sync/releases ``` This release adds support for [node 8.0.0](https://github.com/input-output-hk/cardano-node/releases/tag/8.0.0) Possibly it also supports newer node versions, but this requires testing. ``` So db-sync is...

I think accidentally the header size was used instead of the block body size

state-maching property base testing has been in https://github.com/input-output-hk/cardano-db-sync/pull/1269, for empty or almost empty blocks. They can be extended with the ledger generators

db-sync currently uses the default schema of the db. One solution is to change it before starting db-sync ``` alter database set search_path to ; ``` then db-sync will use...

I did some benchmarks and it seems these queries until epoch 329 are execute 82171515 times (equal to the number of inputs at this moment) and they cause an average...

Have you set the schema dir correctly? Does your ${HOME}/db-sync-schema/ contains the migration files that seem to be missing?

The `--schema-dir` path shouldn't be empty. It should point to the schema files.