<bgilbert>
coreos-installer and Butane containers have been updated for the OpenSSL fix ✔️
ravanelli has joined #fedora-coreos
smarnv has joined #fedora-coreos
ravanelli has quit [Ping timeout: 255 seconds]
jpn has quit [Ping timeout: 248 seconds]
c4rt0__ has joined #fedora-coreos
c4rt0_ has quit [Ping timeout: 260 seconds]
c4rt0 has quit [Ping timeout: 260 seconds]
c4rt0 has joined #fedora-coreos
Betal has joined #fedora-coreos
jpn has joined #fedora-coreos
ravanelli has joined #fedora-coreos
c4rt0 has quit [Remote host closed the connection]
c4rt0__ has quit [Remote host closed the connection]
jpn has quit [Ping timeout: 260 seconds]
<njha>
Does anyone know about tooling that can run software before / after zincati reboots a node? I specifically want to set my rook/ceph OSDs to `noout` when I shut down or reboot a node, so it doesn't try to re-replicate all my data when Zincati does an update.
<njha>
I know of fleetlock but from skimming the source code I think the only check it does is if the Kubernetes node is Ready
<njha>
also as kind of a general question, is FCoS even an appropriate choice for us when all our nodes do storage + compute + mesh network (""hyperconverged"" i suppose)
<njha>
we are about to write our own tool to also make sure ceph is healthy before rebooting anything, but someone else has got to have run into this right 🤔
ravanelli has quit [Remote host closed the connection]