IP_Conntrack table Full Errors
In the process of modifing VM's on two of our production servers, the OVM Manager went into a state where all VM's were in an unknown state. I noticed that there were the following messages in /var/log/messages file:
Feb 18 21:13:02 corpapv3 kernel: ip_conntrack: table full, dropping packet.
Feb 18 21:13:02 corpapv3 kernel: printk: 15 messages suppressed.
I issues the command:
sysctl net.ipv4.netfilter.ip_conntrack_max=1048576
On both nodes, and the unknown status on the OVM Manager cleared up and I was able to continue. I have checked the /etc/sysctl.conf file and we have the setting:
net.ipv4.netfilter.ip_conntrack_max=1048576
Feb 18 21:13:02 corpapv3 kernel: ip_conntrack: table full, dropping packet.
Feb 18 21:13:02 corpapv3 kernel: printk: 15 messages suppressed.
I issues the command:
sysctl net.ipv4.netfilter.ip_conntrack_max=1048576
On both nodes, and the unknown status on the OVM Manager cleared up and I was able to continue. I have checked the /etc/sysctl.conf file and we have the setting:
net.ipv4.netfilter.ip_conntrack_max=1048576
0