OpenVZ Forum - RDF feed
https://new-forum.openvz.org/index.php
'nf_conntrack: table full, dropping packet' @ High packet rate openvz kernel despite unlimited connt
https://new-forum.openvz.org/index.phpindex.php?t=rview&goto=48633&th=11253#msg_48633
see this error almost constantly in kernel log and syslog:
nf_conntrack: table full, dropping packet
I have increased nf_conntrack_max values, by current usage i always well
below the limit (9999999)
if (nf_conntrack_max && unlikely(atomic_read(&net->ct.count) >
nf_conntrack_max)) { unsigned int hash = hash_conntrack(orig); if
(!early_drop(net, hash)) { atomic_dec(&net->ct.count); if
(net_ratelimit()) printk(KERN_WARNING "nf_conntrack: table full,
dropping" " packet.\n"); return ERR_PTR(-ENOMEM); } }
I then set to nf_conntrack_max to 0 and I still get the dropped packets,
which is expected given the first line of the code above.
I have not seen this on other Linux Kernels, although I have not tested
this exact case on non openvz kernel (the vms are doing the traffic). Do
you think this could be openvz specific? Should I boot this to kernel
list? Below is info demonstrating my issue.