On 06/18/2018 05:03 AM, Satish Patel wrote: > Thanks you for comments, I told you half story, full story is we have two kea > instance for HA and they are pointed to galera DB cluster for centralize > Shared DB. > > Both node active and inserting records in cluster DB and everything works but > every few hour node-2 kea is dying with mentioned error, everything same node > dying. Node-1 never had any issue. > > Do you think this is clustering issue they both trying to do same thing and > one of node dying? I have check network and every possible thing but nothing > wrong there, also I have increase max_packet_size in db but no luck :( >
I have galera cluster too and also observe routine cases of 'server has gone away' across sql apps, which is being dealt with by application layer checks to 'reconnect' to the db. Have not investigated the underlying cause other than tcpdump but I think this is squarely a galera / mysql issue. MIke- _______________________________________________ Kea-users mailing list Kea-users@lists.isc.org https://lists.isc.org/mailman/listinfo/kea-users
