Age | Commit message (Collapse) | Author |
|
|
|
Bug 3917
Need to check current state when in operational mode (not config state).
|
|
|
|
|
|
|
|
Moving NAT to a higher priority to be right after firewall and before
interfaces and routing configuration to prevent connections from getting
tracked before firewall and NAT have been setup
|
|
|
|
|
|
Out of an abundance of caution, we don't want to introduce a default
value for a parameter that previously had none without more extensive
testing, as it would affect all systems. We will introduce the default
value in the next release.
|
|
|
|
Previously the script determined the driver that a NIC was using, then
used that to parse the queue names from /proc/interrupts. Now, it
determines the naming convention just by looking at the queue names.
Now the script should work for all NICs that support one of the two
standard naming conventions.
Also added some support for single-queue NICs.
(cherry picked from commit 28e86f46e504eff3ab087aa179a969b1e3383a07)
|
|
This change implements a mechanism for setting the IRQ affinity of NIC
interrupts in a way that is optimally balanced. The affinity settings
are static, and are set up at configuration time. I've added "auto"
as a new legal value to the "interfaces ethernet ethX smp_affinity"
configuration parameter. If this value is set, then the system
attempts to perform automatic IRQ affinity assignment.
At this time, automatic IRQ affinity is supported only for multiqueue
NICs that use the igb or bnx2 drivers. Even with these two,
differences in the organization of queues and their naming convention
necessitate some NIC-specific code. The assignment strategy attempts
to assign all queues of all NICs that share the same queue number to
the same CPU. This has been shown to be optimal for IP forwarding
workloads in 2.6.31 kernels.
Setting the value to "auto" for NICs that use other drivers will be a
no-op.
(cherry picked from commit 565f1d0d9bfe9f6b4174c322f3f6bb78f52b453e)
|
|
|
|
Packets forwarded by NICs using the igb driver are corrupted when
TSO is enabled. It appears that a train of back-to-back packets from the
same flow coalesced by LRO are not correctly re-generated by TSO. This
workaround prevents the problem by disabling TSO.
This problem does not occur in the Kenwood code base, so this workaround
should NOT be merged forward to Kenwood.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
available to the partial commit status api (bug fix 4919). this approach will
be revised in next
release.
|
|
available to the partial commit status api (bug fix 4919). this approach will be revised in next
release.
|
|
|
|
accounting file.
|
|
|
|
* use new API to get active 'original + commited' values
* split 'dns forwarding' and 'dns dynamic' into different priorities
|
|
|
|
|
|
returnOrigPlusComValue()
no term multinode leaf support yet.
|
|
from commit and via .changes
file used for commit active node set.
|
|
|
|
Also restore "--valid-addr" flag to vyatta-interfaces.pl since it is used outside
this package.
(cherry picked from commit d32ceb43ab64530c6eca81ec83fe4b374dd510d0)
|
|
Changed the the set-time and commit-time check of ethernet interface address
values. These checks need to prevent configuring both DHCP and static
IPv4 addresses on the same interfac. The previous checks were comparing
against the running configuration tree, not the proposed config tree. Now
the set-time check is purely a syntax check, and the commit-time check
only checks for both DHCP and static IPv4 addresses in the proposed config.
The system now allows DHCP and static IPv6 addresses to be configured on
the same interface.
(cherry picked from commit 84fb2f920765c6d02f35ef1654dc0f8c2bef648c)
|
|
|
|
Necessary to be able to distguish operators in config file to
password sync.
|
|
|
|
Bug 4918
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
This reverts commit 0d0761a6081ec04f6f27f5785f1ff5b7d8b38ffa.
The fix is worse than the problem. Since the cli doesn't do deletes in
reverse, the addition priorities cause the parent to be deleted before
the group members.
|
|
|
|
failed to be deleted
|
|
|
|
Bonding management needs easy way to find hardware address.
Related to Bug 4758
|
|
|