Return to all command-based Ceph gathering
Using the Rados module was very problematic, specifically because it had no sensible timeout parameters and thus would hang for many seconds. This has poor implications since it blocks further keepalives. Instead, remove the Rados usage entirely and go back completely to using manual OS commands to gather this information. While this may cause PID exhaustion more quickly it's worthwhile to avoid failure scenarios when Ceph stats time out. Closes #137
This commit is contained in:
2
debian/control
vendored
2
debian/control
vendored
@ -8,7 +8,7 @@ X-Python3-Version: >= 3.2
|
||||
|
||||
Package: pvc-daemon-node
|
||||
Architecture: all
|
||||
Depends: systemd, pvc-daemon-common, python3-kazoo, python3-psutil, python3-apscheduler, python3-libvirt, python3-psycopg2, python3-dnspython, python3-yaml, python3-distutils, python3-rados, python3-gevent, ipmitool, libvirt-daemon-system, arping, vlan, bridge-utils, dnsmasq, nftables, pdns-server, pdns-backend-pgsql
|
||||
Depends: systemd, pvc-daemon-common, python3-kazoo, python3-psutil, python3-apscheduler, python3-libvirt, python3-psycopg2, python3-dnspython, python3-yaml, python3-distutils, python3-gevent, ipmitool, libvirt-daemon-system, arping, vlan, bridge-utils, dnsmasq, nftables, pdns-server, pdns-backend-pgsql
|
||||
Suggests: pvc-client-api, pvc-client-cli
|
||||
Description: Parallel Virtual Cluster node daemon (Python 3)
|
||||
A KVM/Zookeeper/Ceph-based VM and private cloud manager
|
||||
|
Reference in New Issue
Block a user