Question regarding HA functionning

semper22

New Member
Sep 19, 2019
19
0
1
29
Hello, I've been testing my HA config lately (3 nodes with 10 VMs distributed) I configured with HA two of my VMs on the node-3; needing to reboot the node-3 server, I migrated these two VM, one on node-1 and one one node-2, I then rebooted the node-3 server via reboot command line.
My question is, is it normal that my Ceph tells me (after the reboot) that there are a lot of active+undersized+degraded files and is it normal that the two VMs are not working ?

Thanks in advance for your answers !
 
Hello, I've been testing my HA config lately (3 nodes with 10 VMs distributed) I configured with HA two of my VMs on the node-3; needing to reboot the node-3 server, I migrated these two VM, one on node-1 and one one node-2, I then rebooted the node-3 server via reboot command line.
My question is, is it normal that my Ceph tells me (after the reboot) that there are a lot of active+undersized+degraded files and is it normal that the two VMs are not working ?

Thanks in advance for your answers !

It would if your crush map isn't configured properly. Sounds like you are at the OSD level instead of Host.

Honestly you should start here.

https://docs.ceph.com/docs/master/rados/operations/crush-map/