Daily Hivemind backups

in HiveDevs5 months ago

Hivemind is the technology that empowers Hive RPC API nodes' backend. Since I run a full node for a while, I also maintain a Hivemind instance.

The initial sync of Hivemind takes a good amount of time. (It was around 6-7 days since the last time I did full sync from hived, instead of a backup file.)

@gtg provides these backups (not daily but he updates when things change) at https://gtg.openhive.network/get/hivemind/ already.

I also wanted to provide daily Hivemind backups to make sure we have multiple options when somebody needs to fire up a new Hivemind instance.

Where to download


holds daily database dumps, including the last two days.

It's refreshed every day at 05:00, UTC. When a backup in progress, the file extension will be in_progress, so don't download it if the file has an extension as in_progress.
You can either wait or download the previous backup to move on.

How to restore


$ pg_restore -j 8 -U <db_user> -d <db_dump> <dump_filename.dump>

-j stands for thread count. I have aligned it with my core count in my restore actions. It takes around 50~ minutes in my setup.

Vote for my witness


If you appreciate what I do, consider casting a vote on Hivesigner or on PeakD.

Sort:  

-j stands for thread count. I have aligned it with my core count in my restore actions. It takes around 50~ minutes in my setup.

Change it to -j $(nproc) and it will dynamically pull the correct amount of processors.

Thanks, I always forget about this.