neigh: enable garbage collection#1068
Open
dirkmueller wants to merge 1 commit into
Open
Conversation
Itxaka
previously approved these changes
Feb 6, 2017
aplanas
reviewed
Feb 6, 2017
| net.ipv4.ip_local_reserved_ports = 35357 | ||
| # Increase system IP port range to allow for more concurrent connections | ||
| net.ipv4.ip_local_port_range = 27018 64999 | ||
| # ensure STALE arp neighbor entries expire from the cache, otherwise |
| # VIPs of an OpenStack service or the floating IP of a VM | ||
| # might not become reachable | ||
| # gc_thresh1 is the lower threshold that needs to be reached before | ||
| # stale entries are getting garbage collected. the default of 128 means |
dirkmueller
force-pushed
the
fix_hanging_floating_ips
branch
from
3f69a02 to
80901e3
Compare
VIPs and floating ips that move between differnet interfaces might stay for very long times cached incorrectly in the neighbor table until the garbage collection kicks in. by default a STALE (so an entry that used to have an active connection but now doesn't anymore) gets garbage collected after gc_stale_timeout, but *only* if there are more than gc_thresh1 STALE entries in total. The default of 128 means that one has to accumulate 128 stale entries (or trigger a forced cache flush) until this is happening, which for small/low traffic clouds can take an eternity.
dirkmueller
force-pushed
the
fix_hanging_floating_ips
branch
from
80901e3 to
748540c
Compare
vuntz
reviewed
Mar 9, 2017
vuntz
left a comment
vuntz
left a comment
Member
There was a problem hiding this comment.
openstack-ansible is using a different approach: https://git.openstack.org/cgit/openstack/openstack-ansible-openstack_hosts/tree/defaults/main.yml#n46
Does that make sense? Or is your approach better?
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
5 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
VIPs and floating ips that move between differnet interfaces might stay
for very long times cached incorrectly in the neighbor table until the
garbage collection kicks in. by default a STALE (so an entry that used
to have an active connection but now doesn't anymore) gets garbage
collected after gc_stale_timeout, but only if there are more than
gc_thresh1 STALE entries in total. The default of 128 means that one has
to accumulate 128 stale entries (or trigger a forced cache flush) until
this is happening, which for small/low traffic clouds can take an
eternity.