12:18 <tpires> Hi all, I have an OVN Central cluster where the leader of the NB started to use 100% of CPU load most of the time.

12:18 <tpires> The database size is around 460MB and I already tried to use 24.03 but no luck, same behavior.

12:18 <tpires> While in 100% CPU the read and write of the NB cluster is impacted. Doing a debugging when there is this increase of CPU load, I can see jsonrpc reply to a member of the cluster.

12:18 <tpires> Do you know anything that can cause such behavior or other steps that I can run?

12:35 <felixhuettner> is there maybe some client that regularly dumps/loads the whole database?

12:44 <tpires> Hi Felix, we were suspecting it and we setup up an ovn-fake-multinode cluster and imported this database there and the behavior still the same. The jsonrpc reply that I mentioned is to a member of the cluster and the size of the reply is about 460MB, I think could be it that is increasing the load but I'm not sure what could be done to fix it.

13:23 <tpires> that is the coverage output

13:23 <tpires> ovs-appctl -t /var/run/ovn/ovnnb_db.ctl coverage/show

13:23 <tpires> Event coverage, avg rate over last: 5 seconds, last minute, last hour, hash=6087dcfb:

13:23 <tpires> hmap_pathological 5.8/sec 3.667/sec 3.6136/sec total: 499941

13:23 <tpires> raft_entry_serialize 0.0/sec 0.000/sec 0.0000/sec total: 58

13:23 <tpires> hmap_expand 79731.0/sec 53154.400/sec 52378.7333/sec total: 7245581653

13:23 <tpires> hmap_reserve 0.0/sec 0.000/sec 0.0000/sec total: 36

13:23 <tpires> lockfile_lock 0.0/sec 0.000/sec 0.0000/sec total: 1

13:23 <tpires> poll_create_node 4.4/sec 4.683/sec 4.5703/sec total: 3477126

13:23 <tpires> poll_zero_timeout 0.6/sec 0.133/sec 0.1447/sec total: 102305

13:23 <tpires> seq_change 0.8/sec 0.433/sec 0.4336/sec total: 365523

13:23 <tpires> pstream_open 0.0/sec 0.000/sec 0.0000/sec total: 4

13:23 <tpires> stream_open 0.0/sec 0.000/sec 0.0000/sec total: 3

13:23 <tpires> unixctl_received 0.0/sec 0.000/sec 0.0000/sec total: 5

13:23 <tpires> unixctl_replied 0.0/sec 0.000/sec 0.0000/sec total: 5

13:23 <tpires> util_xalloc 3428152.4/sec 2285430.383/sec 1059025.8833/sec total: 311602019027

13:23 <tpires> 100 events never hit