Age | Commit message (Collapse) | Author | Files | Lines |
|
Change-Id: Ibd371b845debac895fad97b4e15732d858e00a3c
Signed-off-by: Marek Gradzki <mgradzki@cisco.com>
|
|
Change-Id: I3d65b568c3651e4b2474c25bdf80cf3fff416360
Signed-off-by: Marek Gradzki <mgradzki@cisco.com>
|
|
Change-Id: Ib0b9a2a2369730cd8a21e35e4dce0b62e45e8ad3
Signed-off-by: Marek Gradzki <mgradzki@cisco.com>
|
|
Change-Id: I3160e5df4e66b3b3c732619fc1fc7599459fda76
Signed-off-by: Marek Gradzki <mgradzki@cisco.com>
|
|
Change-Id: Idb0fec8ab45ec76431ab6d05d70a61eee1929628
Signed-off-by: Marek Gradzki <mgradzki@cisco.com>
|
|
Change-Id: I2b886a118ff5167df89a2922802b343ce5978c95
Signed-off-by: Marek Gradzki <mgradzki@cisco.com>
|
|
Fix isses reported by Sonar
thanks to IntelliJ's code inspection.
Change-Id: Ief7b4307d4b90e107f010d7feaa307776afaf947
Signed-off-by: Marek Gradzki <mgradzki@cisco.com>
|
|
Unlike InetAddress.getByName(), Guava's InetAddresses.forString()
never cause DNS services to be accessed.
JDK equivalents whenever you are expecting to handle only IP address string literals,
so there is no blocking DNS penalty for a malformed string.
See:
https://google.github.io/guava/releases/22.0/api/docs/com/google/common/net/InetAddresses.html#forString-java.lang.String-
Change-Id: Id1fc5fa45465b0e0b3aea7d60b9fdb855240d902
Signed-off-by: Marek Gradzki <mgradzki@cisco.com>
|
|
* guava bumped to 22.0 (same as used by ODL)
* use 4K RSA for SSH server (related to bouncy-castle bump in ODL)
- based on https://git.opendaylight.org/gerrit/#/c/60138/2
- also fixes HC2VPP-207
* removed some yang models from netconf-whitelist.xml,
that are no longer present in ODL dependencies,
* increased timeouts in distro tests, because generating 4K RSA key
my last longer (especially on slow jenkins nodes).
* EmptyMountPointService exposed to enable reuse by notification
and restconf providers
Change-Id: I2ff32bd7f0298716210fcfee4918007a42fab171
Signed-off-by: Marek Gradzki <mgradzki@cisco.com>
|
|
Adds option to specify subtree writer that can handle whole subtree
of nodes without having whole subtree specified. Its checking
if node is children at runtime, rather than having pre-computed tree
Change-Id: Ic46f2bd6de84f0dd14865825399f5a90a1f80859
Signed-off-by: Jan Srnicek <jsrnicek@cisco.com>
|
|
Provides maven modules per bgp extension.
Defines common configuration that can be used to define new
extensions(AbstractBgpExtensionModule).
Change-Id: I1c8ff65b6f9aa5474026f4bdf42476bd0559208e
Signed-off-by: Jan Srnicek <jsrnicek@cisco.com>
|
|
Change-Id: I64bcd7e45da5d7431ba436ff6ad1ac62cd1d1051
Signed-off-by: Marek Gradzki <mgradzki@cisco.com>
|
|
Change-Id: I896f1aae3b71e21bd6127aedf22faef07a132daf
Signed-off-by: Marek Gradzki <mgradzki@cisco.com>
|
|
Change-Id: If57a5414153cb983138872c7a1ded7c7066b1b6d
Signed-off-by: Jan Srnicek <jsrnicek@cisco.com>
|
|
Change-Id: I63fa239e34178ab03c4eecd421bf19d80303d30e
Signed-off-by: Marek Gradzki <mgradzki@cisco.com>
|
|
Change-Id: I3a77a35fb039acb1e0a138450883922e9a02683f
Signed-off-by: Marek Gradzki <mgradzki@cisco.com>
(cherry picked from commit 756ffae6576fe50c052950418627be4a1d21f996)
|
|
BGP peer configuration is no longer read from bgp-peer.json file.
Netconf/Restconf is can be used instead.
BGP peer configuration in HC follows openconfig-extensions model (as in ODL BGP):
* http://docs.opendaylight.org/en/stable-boron/user-guide/bgp-user-guide.html#bgp-peering
* http://docs.opendaylight.org/en/stable-boron/user-guide/bgp-user-guide.html#bgp-application-peer-and-programmable-rib
Change-Id: I91aa6c4fc0923edbacf6cd10abd3957569a4f8c6
Signed-off-by: Marek Gradzki <mgradzki@cisco.com>
(cherry picked from commit a3d562afdd96d4c37fe608af99f364e879ee92b6)
|
|
Change-Id: I49a201b8e027d229a60bb76abe9b5688af7b7cdf
Signed-off-by: Marek Gradzki <mgradzki@cisco.com>
|
|
Change-Id: I459425f4ee8f43d0bf196d2c762e741c58667907
Signed-off-by: Marek Gradzki <mgradzki@cisco.com>
|
|
Change-Id: I6294fc266b704568d4e093a50d615f96eac153be
Signed-off-by: Marek Gradzki <mgradzki@cisco.com>
|
|
Use module configuration to enable/disabe BGP.
Also hides BgpExtensionModule module to the user
(until HONEYCOMB-363 is fixed).
Change-Id: I8294d080138a0bcf640420484b76c089aa332d93
Signed-off-by: Marek Gradzki <mgradzki@cisco.com>
|
|
This change is necessary to break dep cycle, that would by introduced
by splitting netconf and data-core modules. Also allows to reduce
use in hc2vpp(some project incude whole minimal-distribution-core
just because of this one interface)
Change-Id: I7744b4dc78f6d6797263b10bbc4ead545a2f6593
Signed-off-by: Jan Srnicek <jsrnicek@cisco.com>
|
|
Separate BGP distribution was temporary workaround
for HONEYCOMB-358 and (partially) HONEYCOMB-360.
BGP northbound interface is disabled by default.
Change-Id: I72a0f83f40ea96bee478be7d6560f2112611d80f
Signed-off-by: Marek Gradzki <mgradzki@cisco.com>
|