Conversation

Notices

1. Phantasm (phnt@fluffytail.org)'s status on Tuesday, 26-Nov-2024 00:38:31 JST Phantasm
   tfw it takes 6 hours to restore a 10GB Pleroma DB the right way. Thanks OVH :)
   •  (mint@ryona.agency)'s status on Tuesday, 26-Nov-2024 00:38:30 JST 

     in reply to
     @phnt Just copy the whole cluster, it should work as long as the system has the same architecture (if not just endianness).
   • Phantasm (phnt@fluffytail.org)'s status on Tuesday, 26-Nov-2024 02:06:32 JST Phantasm

     in reply to

     • 
     @mint I've never tried to do that.
     It should work fine as long as the collation version and the OS are the same. With different collation a reindex should be enough.
     
     In this case it wouldn't have helped, because terrible IO is the culprit and a copy would take about the same time. The CPU is doing basically nothing but waiting for IO to finish.
     I should've kept the database alone instead of doing a full system reinstall, oh well.
      likes this.