Aswin Suryanarayanan [Tue, 1 Aug 2017 13:28:08 +0000 (18:58 +0530)]
Bug 7545 - FIP-FIP traffic
between vm in same n/w in same compute is not working in stateful SG
mode.
The patch fixes the FIP-FIP use case.
Add a dummy table to do a ct_lookup if the packet was already send to
the netfilter from the pipeline.
Change-Id: Ibc0d30aa1315d67938ebcbaddcfa2593113f9686
Signed-off-by: Aswin Suryanarayanan <asuryana@redhat.com>
Vinh Nguyen [Thu, 3 Aug 2017 19:14:37 +0000 (12:14 -0700)]
BUG 8876 - Fix to remove ELAN interface flows when external network deleted
- Problem: when deleting external network, the underlying ietf
Interface might have been removed from the config DS prior
to the call to delete the ELAN interface. As a result, the call
interfaceManager.getInterfaceInfo returns null and the
subsequent call to delete the ELAN interface fails to delete the
flow entries for external network's ELAN interface.
- Fix: invoke interfaceManager.getInterfaceInfoFromOperationalDataStore
when delete External Network's ELAN interface
Change-Id: I764f3516d34a6d79aacff8dbedabb55219799ac9
Signed-off-by: Vinh Nguyen <vinh.nguyen@hcl.com>
Sam Hague [Mon, 7 Aug 2017 17:08:23 +0000 (17:08 +0000)]
Revert "Bug 8882 - With conntrack SNAT communication with PNF fails "
This reverts commit
0a1048ddc55621b2b675b7cef3e1b6c808693af1.
Change-Id: Ia0522b4c5f30dc11995bbe5e043d66f41f490cb7
Signed-off-by: Sam Hague <shague@redhat.com>
(cherry picked from commit
c8483f3ffc7a63fb72ae67c82ea9fbe73ac93900)
Sridhar Gaddam [Mon, 7 Aug 2017 06:55:35 +0000 (12:25 +0530)]
Bug 8936: Fix IPv6 IPAM issues when an IPv6 subnet is added to IPv4 network
This patch addresses the following issues in ipv6service.
1. In an existing IPv4 network (which has a DHCP Port and some VMs)
when you add an IPv6 subnet, the VMs/DHCP Port is not configured
with the IPv6 address.
2. Similarly, in a multi-node setup when an IPv6 subnet is added to
an existing IPv4 network and associated to the router, sometimes
the necessary punt flows in Table_45 are not configured.
Change-Id: I0c91d40ae692f32567bc77f778d77af278e0b663
Signed-off-by: Sridhar Gaddam <sgaddam@redhat.com>
eaksahu [Mon, 3 Jul 2017 12:15:38 +0000 (17:45 +0530)]
Bug 8789 - Designated DHCP DPN is missing.
When l2gw connection is creaated and first dpn interface creation
happens at about same time there is a race in dhcp designated listener
which thinks that it has to update mcast only if there are no dpns in
elan.
when an elan dpn is removed, the dpns tep ip should not be removed from
mcast mac if that tep ip is the designated dpn tep ip.
when l2gw is deleted clear its connections also.
Change-Id: I6a216cecc609ba75f10ebfc750e492348e447122
Signed-off-by: eaksahu <a.k.sahu@ericsson.com>
Jaime Caamaño Ruiz [Fri, 14 Jul 2017 17:10:51 +0000 (19:10 +0200)]
Add chain egress classifier support
On scenarios where origin and destination of traffic are on the same
node, there is no tunnel id metadata to help steer the traffic back
to the destination on chain egress, thus SFC cannot take care of
this by itself. In a general case, the classifier is in a better
position to steer traffic to destination as it is aware on how it
was intercepted on the first place.
For netvirt classifier and the scenario mentioned above, the
classifier will copy the REG6 to NSH C4 on interception. On chain
egress, SFC will send the encapsulated packet back to the origin node
(only when C2 is 0, which will only happen when origin and destination
node are the same). There, the classifier will restore the
de-encapsulated packet to the egress dispatcher table with the proper
REG6 value taken from NSH C4. REG6 contains the the egress ifindex plus
the nex egress service in sequence from the classifier egress service
to handle the packet. This effectively restore the packet processing
where it left off on interception.
Change-Id: I1c494f24c5af530849faabcbfa8efcaa9cd4344a
Signed-off-by: Jaime Caamaño Ruiz <jaime.caamano.ruiz@ericsson.com>
Signed-off-by: Sam Hague <shague@redhat.com>
David Suarez [Wed, 2 Aug 2017 18:38:31 +0000 (20:38 +0200)]
Fix some logs
Fix some logs.
Change-Id: I9a72aa7ec1563305596370528f70653eae2ff9ae
Signed-off-by: David Suarez <david.suarez.fuentes@gmail.com>
David Suarez [Fri, 4 Aug 2017 18:46:48 +0000 (20:46 +0200)]
Replace size()==0 by isEmpty()
Replace size()==0 by isEmpty(). It improves readability and avoid "=0".
Change-Id: I60f975639d32ba7f2c317779b7af5beb2f92ed1a
Signed-off-by: David Suarez <david.suarez.fuentes@gmail.com>
Aswin Suryanarayanan [Fri, 28 Jul 2017 12:19:23 +0000 (17:49 +0530)]
Bug 8882 - With conntrack SNAT communication with PNF fails
Change set_field to load action and set only the relevent bits. Thus the
lport tag is preserved.
Depends on [1] https://git.opendaylight.org/gerrit/#/c/60870
Change-Id: I769254de971deed02cf510ee62db08c3d887070e
Signed-off-by: Aswin Suryanarayanan <asuryana@redhat.com>
Jaime Caamaño Ruiz [Mon, 31 Jul 2017 14:56:09 +0000 (16:56 +0200)]
BUG 8914: Fix NPE in sfc.classifier-impl
NPE was due ACLs with no RedirectToSfc action. Such ACLs should be
ignored.
Change-Id: If4eb944308db5432919da1c5df8b00469a5df561
Signed-off-by: Jaime Caamaño Ruiz <jaime.caamano.ruiz@ericsson.com>
eupakir [Wed, 2 Aug 2017 13:47:58 +0000 (19:17 +0530)]
Bug 8346 - Conflicting modification for vpnNextHops.
The VpnInstanceListener used to wait for the vpn Interface Operational
data to clean up, before cleaning up data stores related to it.
There was a max wait time calculate, but some time interface op cleanup is
exceeding this wait time. As a result, the vpn instance related cleanup is
done before or at the same time as the vpn interface op cleanup,
resulting in these exceptions.
This fix remove the timed wait for the vpn instance, and cleans up the vpn
data based on notifications fired when both a vpn instance delete is
orchestrated and its footprint is cleared
Change-Id: I54d6369f30aa6d2a0e98df56b80cfcd6b4295978
Signed-off-by: eupakir <kiran.n.upadhyaya@ericsson.com>
Signed-off-by: Sam Hague <shague@redhat.com>
David Suarez [Wed, 2 Aug 2017 14:51:15 +0000 (16:51 +0200)]
Cleanup
- Replace ArraList by List
- Add override
- Declaration of constants
- Replace size() by isEmpty()
Change-Id: Iefc721ded2ef21e7da9fc59b36e1195e47d77f08
Signed-off-by: David Suarez <david.suarez.fuentes@gmail.com>
(cherry picked from commit
e84605a35e3b12b601e8d510dcd2d961bb925416)
eupakir [Tue, 1 Aug 2017 08:25:24 +0000 (13:55 +0530)]
Bug 8917 - CSIT Sporadic failures - Arp learning suite - missing flow on
compute node
Remote FIB Flow installation was failing due to OptimisticLockException.
The DJCs involved in this operation is now modified in the L3 Application
to retry on failure.
Change-Id: Ifdb4d3735892de28fb73b6b563878d059d8298ee
Signed-off-by: eupakir <kiran.n.upadhyaya@ericsson.com>
Valentina Krasnobaeva [Thu, 20 Jul 2017 14:01:30 +0000 (16:01 +0200)]
Bug 8875 - fix in handleNeutronPortUpdated
This fixes commit "Bug 6770 - Fixes DjC for NPortCL + snmaps serialized
+ listed changes"
19af79aecc2f217c01faeda721606a77bd1b2504.
NeutronPortChangeListener: handleNeutronPortUpdated method receives
portoriginal and portupdate objects as arguments. So, to check, if
neutron port was really updated, compare portoriginal and portupdate
SubnetIDs.
Change-Id: I7509c61bfa1135e21fa8e0a49ca8e0f1c68d4c72
Signed-off-by: Valentina Krasnobaeva <valentina.krasnobaeva@6wind.com>
Acked-by: Philippe Guibert <philippe.guibert@6wind.com>
eceghkl [Thu, 18 May 2017 05:43:35 +0000 (11:13 +0530)]
Bug 8439: Handling interface update event for ECMP extra routes.
* Up/Down event from Interface state was not handled before this fix.
Only add and delete event were handled
* With this fix, up/down events are handled. These events are processed
similar like add/delete events. Vpntoextraroute, extraroute-rds and
fibentries are updated accordingly which will result in the bucket
removal/addition from the select group. For the last extraroute of a
particular prefix, the corresponding flow(table 21) and select group
are removed
Change-Id: Ieb205ff9e626c67c888ecdf3d5bfbcc679fdeba8
Signed-off-by: eceghkl <manu.b@ericsson.com>
karthikeyan [Tue, 25 Jul 2017 11:39:55 +0000 (17:09 +0530)]
Bug 8884: SNAT traffic is getting dropped when router gateway is removed
Problem Description:
====================
Issue 1
+++++++
SNAT traffic is getting dropped for second router when gateway is removed
from first router.
Consider the following topology.
1) 2 Internal Network + 2 Subnets
2) 2 Routers with each subnet associated to it.
3) One External network and both router associated to same External
Network.
4) One External BGPVN.
when both Routers, if the same NAPT switch is elected, it will have a
common NAPT PFIB TABLE which forwards the outgoing packet. In such
scenario, when the gateway is removed for one of the routers,
this flow gets removed effecting the other routers traffic.
Solution for Issue 1:
+++++++++++++++++++++
Changes done to prevent this NAPT PFIB table flow(47->21) deletion if
there exists
any other router having same DPN as NAPT Switch.
Issue 2:
++++++++
Label of external-fixed-ip is freed from vpnservice pool when NAPT
switchover happens for SNAT.
Solution for Issue 2:
+++++++++++++++++++++
Changes done to prevent release of label allocated for external-fixed-ip
when Switch-Over Happens.
Change-Id: I0d34ee1140f0d2b800d82271b6f0604bbb8bb9c1
Signed-off-by: karthikeyan <karthikeyan.k@altencalsoftlabs.com>
eceghkl [Thu, 29 Jun 2017 05:28:53 +0000 (10:58 +0530)]
Bug 8632: ECMP LB group not updated during tunnel events
* Tunnel State is read from interface operational DS before installing
select group
* A few NPE also fixed in this review
Change-Id: Id61c20059fcf4dff16f7590b0c439330bb2c46c3
Signed-off-by: eceghkl <manu.b@ericsson.com>
Michael Vorburger [Tue, 11 Jul 2017 20:42:48 +0000 (02:12 +0530)]
Remove not needed Thread.sleep() from aclservice component tests
Change-Id: Id3582518e28cc5f523027f806b77aa46cbd32204
Signed-off-by: Michael Vorburger <vorburger@redhat.com>
Shashidhar Raja [Wed, 26 Jul 2017 05:06:11 +0000 (10:36 +0530)]
Bug 8838: aclservice NPE's
This commit fixes NPE observed in processInterfaceUpdate().
Similar fix added in applyAce() and removeAce().
Also, unwanted check taken out from updateRemoteAclTableForMultipleAcls().
Change-Id: I895af6b8c59a85400b714c007dab5b03630d93cf
Signed-off-by: Shashidhar Raja <shashidharr@altencalsoftlabs.com>
Victor Pickard [Tue, 16 May 2017 20:34:35 +0000 (16:34 -0400)]
Bug 8476 Add support for logical switch replication mode
With OVS 2.6, the Logical Switch table, in the hardware_vtep schema, has
a new field, replication_mode. This bug will be used to add support to
netvirt, ovsdb, and genius for setting the replication_mode of logical
switches. The default will be for ODL to set the replication_mode
to "source_node" when creating logical switches.
Refer to [1] for more details on replication_mode.
[1] http://openvswitch.org/support/dist-docs/vtep.5.html
This patch depends on genius patch:
https://git.opendaylight.org/gerrit/#/c/57210/
Change-Id: I8581fa3bbc1af7c8d13bed2d66e351d020366ea3
Signed-off-by: Victor Pickard <vpickard@redhat.com>
(cherry picked from commit
e25b3264f15af87200b94870dd0d6e9a6f4db14b)
Freeingo Sure [Wed, 26 Jul 2017 03:32:35 +0000 (11:32 +0800)]
Bug 8892 - LOG.error format error(VpnManagerImpl.java:176)
Change-Id: Ic1c88364cb04a5226f4cf3d7d32c8125db5508f7
Signed-off-by: Freeingo Sure <xiao.fenggang@zte.com.cn>
karthikeyan [Tue, 18 Jul 2017 12:44:21 +0000 (18:14 +0530)]
Bug 8864: DNAT to SNAT traffic is getting failed on same DPN for VXLAN
Problem Description:
=========================
DNAT to SNAT traffic is getting failed on same hypervisor for External
VXLAN Provider Type
Solution:
===========
Have added the follwing flow install/remove code to handle this problem
for external VXLAN provider type (EVPN_RT5)
Install the flow L3_FIB_TABLE (table=21)-> INBOUND_NAPT_TABLE (table=44)
(FIP VM on DPN1 is responding back to external fixed Ip on DPN1 itself.
ie. same Hypervisor){DNAT to SNAT Intra DC traffic}
Change-Id: Id6670876a42ab0a7e89abee705bb0848a642e8df
Signed-off-by: karthikeyan <karthikeyan.k@altencalsoftlabs.com>
Vishal Thapar [Tue, 25 Jul 2017 04:24:21 +0000 (09:54 +0530)]
BUG 8880: Trunk port flows not created for subports
Fix for BUG 8020 broke the assumption for trunk sub-ports i.e. Interface
will be already created for Subports at time of NeutronPort creation.
https://git.opendaylight.org/gerrit/#/c/54410/ adds an if condition that
will always be false for VM ports, so we don't create interface at Port
creation. This breaks trunk logic. This also has side effect of increasing
rae conditions by adding too much to do once VM is actually spawned and is
not a good design.
Change-Id: I90e5c60f6d5afe5242cdd0a93d5c35d00a25ab85
Signed-off-by: Vishal Thapar <vishal.thapar@ericsson.com>
Tom Pantelis [Sat, 22 Jul 2017 16:49:53 +0000 (12:49 -0400)]
Remove derivation from controller config-parent
Netvirt does not use the config subsystem so does not need to
derive from controller config-parent. I removed the local
config-parent and changed poms to derive from the local
binding-parent.
Change-Id: Ib6ab82d43a021247808f17cefb8d836e5b00b36e
Signed-off-by: Tom Pantelis <tompantelis@gmail.com>
Signed-off-by: Sam Hague <shague@redhat.com>
Aswin Suryanarayanan [Mon, 17 Jul 2017 07:27:28 +0000 (12:57 +0530)]
Bug 8853 - In conntrack SNAT , FIB flows are not created for existing
Non-NAPT switch ports.
When a port is added to the VpnDpnList it is not checked whether it is a
first/last port and handle the fib flows.
VpnFootprint service Api is now used to handle the last port and
add/delete the fib flows.
Change-Id: Iea809bad856846aaad623bd4c49f4b0f0642982b
Signed-off-by: Aswin Suryanarayanan <asuryana@redhat.com>
eswanit [Thu, 20 Jul 2017 14:30:13 +0000 (20:00 +0530)]
Bug 8753 : Changes to create bgpvpn without VPN target
When bgpvpn is created without VPN targets, it throws
an error as "GetVPN failed due to null" in VPN config.
This is fixed by adding a check of, if VPN target is
null, then also create VPN by avoiding the exception.
Change-Id: I08179cc7a7e9013be7008e0514b98e9e0edc3bd9
Signed-off-by: eswanit <swati.udhavrao.niture@ericsson.com>
Noel de Prandières [Mon, 10 Jul 2017 14:50:53 +0000 (16:50 +0200)]
Bug 8841 : add fib-show option to select entries per address-family or prefix
An enhancement of fib-show is done here to show optionally prefixes
based on address-families or/and prefix/subnet value given in input.
Available values for address-families are : ipv4, l2vpn, ipv6.
Available values for prefix/subnet are IPv4 or IPv6 entries, with or without
prefix length.
Change-Id: I8d1c61149bae7442221dea5b0d04aae0de59f9b5
Signed-off-by: Noel de Prandières <prandieres@6wind.com>
Tarun Thakur [Tue, 4 Jul 2017 09:50:16 +0000 (15:20 +0530)]
Bug 8683: Aclservice releaseId IdManager Exception
Issue: When VM port is deleted, ACL service calls IDManager to release ID
which in turn throws exception that Id Entries does not exist.
Root cause:
When OVS is restarted, then ACL pools are deleted and created freshly
due to which IDs allocated to SG flows of VM port before OVS restart are
not available in IdManager ACL pools after OVS restart.
Fix:
On OVS restart, Do not delete ACL pools for OVS if VM ports are associated with OVS.
Change-Id: I0a0d36d4d8f25d7157d73292f9a027f67fe6dc0c
Signed-off-by: Tarun Thakur <tarun.t@altencalsoftlabs.com>
Stephen Kitt [Wed, 19 Jul 2017 15:25:19 +0000 (17:25 +0200)]
Move statistics into org.opendaylight.netvirt
Now that the BluePrint plugin only considers org.opendaylight.netvirt,
we need to move all NetVirt’s packages into that namespace.
Change-Id: I953e9df0a9c650161083239acb1889e529429f1a
Signed-off-by: Stephen Kitt <skitt@redhat.com>
Signed-off-by: Sam Hague <shague@redhat.com>
Stephen Kitt [Fri, 16 Jun 2017 16:36:38 +0000 (18:36 +0200)]
Clean up lambdas and streams
* Use method references where appropriate.
* Use lambdas instead of anonymous classes where appropriate.
* Simplify lambdas as far as possible.
* Use forEach() directly on collections instead of streaming them.
* Simplify streaming matchers (!anyMatch -> noneMatch in particular).
Change-Id: I38c4ec0b45d355d495b9db563cb363414059f1aa
Signed-off-by: Stephen Kitt <skitt@redhat.com>
Aswin Suryanarayanan [Tue, 18 Jul 2017 12:50:47 +0000 (18:20 +0530)]
Bug 8857 - SNAT Conntrack - VM MAC is used as Source MAC for all
outbound traffic
The vm mac is replaced with router mac address
Change-Id: Ia49aad53178d5f5b67b915dce72e453a23b3e9d2
Signed-off-by: Aswin Suryanarayanan <asuryana@redhat.com>
Philippe Guibert [Thu, 6 Jul 2017 14:12:12 +0000 (15:12 +0100)]
Bug 8810 : BGP Manager / support for EVPN on OAM submodule missing
Despite EVPN is supported on VPNs, there is no statistics coming from
BGP. The changes include the following:
- evpn summary information ( list of neighbors speaking epvn)
- evpn total list of entries
Change-Id: I18b89b6ebe5815e1e3b827e96e0269852fd88a5f
Signed-off-by: Philippe Guibert <philippe.guibert@6wind.com>
Philippe Guibert [Thu, 6 Jul 2017 14:13:30 +0000 (15:13 +0100)]
Bug 8809 : BGP Manager / neighbor summary for VPNv6 wrongly parsed
Instead of filtering the BGP neighbor IP with IPv6 address, the
filtering is done against IPv4 neighbor addresses. There is no need to
match with IPv6 neighbors, since it is related to underlays tunnels, and
not overlays information like VPNv6.
Change-Id: I6f2da4c1ca84a842a34a4c7beab02e4899322803
Signed-off-by: Philippe Guibert <philippe.guibert@6wind.com>
Philippe Guibert [Thu, 6 Jul 2017 14:39:53 +0000 (15:39 +0100)]
Bug 8811 : BGP Manager / bgp-nbr shell config with afi vpnv6 / evpn
This commit brings some changes related to VPNv6 and EVPN capabilities,
since it is becoming possible to create neighbours with VPNv6 or EVPN.
Change-Id: I821475d5d44790fb8ec0f86b6eb3a4d4577f1e57
Signed-off-by: Philippe Guibert <philippe.guibert@6wind.com>
Philippe Guibert [Fri, 7 Jul 2017 08:09:41 +0000 (09:09 +0100)]
Bug 8818 : BGP Manager. VPNv6 prefixes injected to QBGP
This issue is related to VPNv6 prefixes under the format prefix/length.
Indeed, the prefix/length format is not recognized by bgp manager when
analysing the prefix. Then it fails and considers it as IPv4.
Subsequently, an inconsistency is detected by QBGP, and it is rejected.
The fix consists in striping the length part so that the prefix can be
analysed correctly.
Signed-off-by: Philippe Guibert <philippe.guibert@6wind.com>
Change-Id: Ic396fe795eb6704c92b099c74631c96774034427
Philippe Guibert [Fri, 7 Jul 2017 13:55:38 +0000 (14:55 +0100)]
Bug 8822 : Bgp Manager / bgp-cache dumps unknown address-families
The fix consists in parsing the various afi/safis of neighbour contained
in address-family list, and identify as VPNv6 or EVPN address-families.
Change-Id: Ifffa01aef4f6f8e27b6750d1c7dde6c3bca8e008
Signed-off-by: Philippe Guibert <philippe.guibert@6wind.com>
Philippe Guibert [Thu, 30 Mar 2017 09:51:16 +0000 (10:51 +0100)]
Spec to support dualstack VMs in L3VPN
This specification brings the ability for a BGPVPN configured VPN to
distribute IPv4 and IPv6 addresses within the same VPN, and by using two
distinct routers to respectively handle IPv4 and IPv6 routing issues.
Change-Id: Ia76e88a8eff6135d58b52ec69e610952cf8a3621
Signed-off-by: Philippe Guibert <philippe.guibert@6wind.com>
Signed-off-by: Sam Hague <shague@redhat.com>
karthikeyan [Tue, 4 Jul 2017 04:30:51 +0000 (10:00 +0530)]
Bug 8770: Stale NAT entries are not getting removed for external FLAT
Problem Description:
=====================
Stale NAT entries are not getting removed for external FLAT/VLAN network
on following flows
Stale FIB Entry =>L3_FIB_TABLE(21)-> INBOUND_NAPT_TABLE(44)
Stale Tunnel Entry =>INTERNAL_TUNNEL_TABLE(36)-> OUTBOUND_NAPT_TABLE(46)
Stale LFIB Entry =>NAPT_PFIB_TABLE(47)->L3_FIB_TABLE(21)
Solution:
=============
As part of hairpinning feature support, external vpn name was generated
from external-subnet Id instead of external-network Id. To accomplish this
changes have done the changes in all the places to install and remove the
NAT related flows for external FLAT/VLAN use cases.
Change-Id: I0bd0c79ca638cf6664eaa4237334c6524f2d2c1f
Signed-off-by: karthikeyan <karthikeyan.k@altencalsoftlabs.com>
David Suarez [Sun, 16 Jul 2017 14:26:03 +0000 (16:26 +0200)]
Add a private constructor to this utility class
Add a private constructor to this utility class so that it cannot be
instantiated.
Change-Id: Iba69fb65160e8fd905c477cec8c89f4d59c1c538
Signed-off-by: David Suarez <david.suarez.fuentes@gmail.com>
(cherry picked from commit
44946c0fd5283147460f56213104ac10d65aacd0)
Stephen Kitt [Mon, 17 Jul 2017 08:26:19 +0000 (08:26 +0000)]
Revert "Ship aaa-cli-jar.jar in the distribution"
This reverts commit
22ec593f55cf8bb1cf3674dd8b4a15f11667a3d7, which breaks the build of the Karaf distribution (on Jenkins).
Change-Id: I5eb91fe277bcb75c2706d01761fee216ac6c4323
Signed-off-by: Stephen Kitt <skitt@redhat.com>
Stephen Kitt [Thu, 6 Jul 2017 15:04:35 +0000 (17:04 +0200)]
natservice-impl: propagate upstream non-null annotations
Change-Id: I458244f842dd1266ae8560032151f6f8224b79ba
Signed-off-by: Stephen Kitt <skitt@redhat.com>
Siva Kumar Perumalla [Tue, 11 Jul 2017 17:32:34 +0000 (23:02 +0530)]
Bug 8835: Java NullPointerException in display-bgp-config command
Change-Id: I8df4de8620c1147d4d9f30152339e2db840f5787
Signed-off-by: Siva Kumar Perumalla <sivakumar.perumalla@ericsson.com>
eswanit [Tue, 11 Jul 2017 16:59:59 +0000 (22:29 +0530)]
Bug 8567: Addition of new flows after addng extra route
When the extra routes are configured on VMs, ping between
them works and these extra route IP entries appear in list
of adjacencies and new flows are added in table 21.
This is fixed by adding a check of, if an extra route IP
is present in any of the adjacencies' listed subnets,
then do not execute ProcessArpLearning.
Change-Id: I49f691b48b7f838bdc42d14ee81daff1e96d7e16
Signed-off-by: eswanit <swati.udhavrao.niture@ericsson.com>
Achuth Maniyedath [Wed, 12 Jul 2017 13:55:54 +0000 (19:25 +0530)]
New Yang model container for Neutron DHCP Port service.
Updated spec with correct yang.
Change-Id: I8ee02c72db01dcb840683584a1bacc24f20448c9
Signed-off-by: Achuth Maniyedath <achuth.m@altencalsoftlabs.com>
Stephen Kitt [Thu, 6 Jul 2017 14:27:53 +0000 (16:27 +0200)]
Enforce non-null collection returns in NatUtil
Collection-returning methods shouldn’t return null; this fixes
NatUtils appropriately and propagates the non-null guarantees.
Change-Id: Ibf7a1c87ba37098df1abb1d1dbbd8517a9c7e454
Signed-off-by: Stephen Kitt <skitt@redhat.com>
karthikeyan [Wed, 12 Jul 2017 06:53:35 +0000 (12:23 +0530)]
Bug 8844: CSIT Job: NullPointerException from NAT feature
Problem Description:
====================
Retrieving router-name from router-id NAT DS
"odl-nat/router-id-name/routerIds" sometimes returning with empty as
resulting Nullpointer exceptions are been thrown from NAT module. Since
there is no check to handle for null values.
Solution:
============
As part of fixing this Nullpointer exception, we have added the
appropriate check to validate the null values in cll the aller methods.
Change-Id: I6b7b057632fb430ae61af00a31a212684edb1e9a
Signed-off-by: karthikeyan <karthikeyan.k@altencalsoftlabs.com>
cgowdru [Wed, 28 Jun 2017 09:10:10 +0000 (14:40 +0530)]
SNAT performance improvement for Controller-Based SNAT
Description : Updated log statements and reduced number of retry
threads required for processing.
Change-Id: I41a4d03088efbe2c8c2e5c000e68b665c215453e
Signed-off-by: cgowdru <chetan.arakere@altencalsoftlabs.com>
kaoru sueda [Fri, 14 Jul 2017 05:48:34 +0000 (14:48 +0900)]
Bug 8796: Fix of issue that vxlan port is created with remote_ip set to the node itself
Issue:
If start L2GW node (Open vSwitch HWVTEP emulator) without setting HA
configuration and create L2GW GATEWAY and L2GW CONNECTION, after
update (set) HA configuration for this L2GW Node, vxlan port which
its remote_ip is set to the node itself will be created.
Cause of this issue:
If there are multiple tep ips of the node itself in the tep ip list
which is generated from L2GW devices information,
the information of the node itself will be set in RemoteMcastMac.
Fix:
Do not add the node information to tep ip list if the same tep ip
exists in the list which is generated from L2GW devices information.
Change-Id: I8ca27a32248a9439a1f9d56077d6071fe4de30ac
Signed-off-by: kaoru sueda <k-sueda@zj.nes.jp.nec.com>
Stephen Kitt [Thu, 13 Jul 2017 14:34:30 +0000 (16:34 +0200)]
Ship aaa-cli-jar.jar in the distribution
This is a port of https://git.opendaylight.org/gerrit/49278 and
https://git.opendaylight.org/gerrit/54388
Change-Id: I1c3cf5efd60a017bc33f63c5872e21f1eef9eda3
Signed-off-by: Stephen Kitt <skitt@redhat.com>
eaksahu [Mon, 3 Jul 2017 12:51:33 +0000 (18:21 +0530)]
Bug 8791 - L2gateway delete is not clearing l2gwCo
L2gw delete was not clearing associated L2gateway Connection.
CLearing associated l2gwConnection as part of the review.
Change-Id: Ie8b6a9668c7a943041bb2cc7f20bbcab7e68b51d
Signed-off-by: eaksahu <a.k.sahu@ericsson.com>
eaksahu [Mon, 3 Jul 2017 12:38:35 +0000 (18:08 +0530)]
Bug 8790 - Local Macs getting cleared
Local Macs are getting cleared from HA parent node , in case only one of
the HA child node is diconnected.
It should be cleared if both the HA nodes are disconnected .
Change-Id: I222bc69aa7e4aa477870014a793e0ae5d73de26c
Signed-off-by: eaksahu <a.k.sahu@ericsson.com>
Stephen Kitt [Fri, 7 Jul 2017 14:32:39 +0000 (16:32 +0200)]
Fix dpnId handling in SRISCListener
In SubnetRouteInterfaceStateChangeListener::add, we call
vpnSubnetRouteHandler.onInterfaceUp() whether we get a dpnId or not.
But onInterfaceUp() immediately returns if dpnId is BigInteger.ZERO,
so there’s no point in doing so, and having both calls in the same try
blocks makes the whole process clearer.
Change-Id: Icfe7a1273305ec4825afa16484ee0a8e8f685d47
Signed-off-by: Stephen Kitt <skitt@redhat.com>
Jaime Caamaño Ruiz [Mon, 10 Jul 2017 10:41:15 +0000 (12:41 +0200)]
BUG 8828: Fix NPE when no remote IP on interface
Fix NPE When new netvirt classifier fetches the remote IP address of an
egress interface that does not have one (for example, when it is not
a tunnel interface).
Change-Id: Ie76ca25847113d04d440c6497768e9f363ac9569
Signed-off-by: Jaime Caamaño Ruiz <jaime.caamano.ruiz@ericsson.com>
Vishal Thapar [Tue, 4 Jul 2017 09:45:49 +0000 (15:15 +0530)]
Filter notifications for unwanted interfaces
Change-Id: I3e6a82c97bdd8fd467cf5f39cd59fdb88c121223
Signed-off-by: Vishal Thapar <vishal.thapar@ericsson.com>
Sam Hague [Fri, 14 Jul 2017 19:29:18 +0000 (19:29 +0000)]
Revert "Spec to support dualstack VMs in L3VPN"
This reverts commit
590dcb2a96af99052e20465be3ae4c75742b2ef5.
Change-Id: I4457dc6d0f76e527ab894f5ef0bd38b768b8e963
Signed-off-by: Sam Hague <shague@redhat.com>
Michael Vorburger [Tue, 11 Jul 2017 21:40:26 +0000 (03:10 +0530)]
Fix ActionNxResubmit in FlowEntryObjectsStateless for aclservice
This fixed a test failure I'm seeing locally in aclservice's
AclServiceStatelessTest>AclServiceTestBase.newInterface
(and which should fail on Jenkins as well; if it does not, I do not
understand how it can work on the build; we should find out more why).
The test failure message shown is confusing; with
https://git.opendaylight.org/gerrit/#/c/60219/ it becomes much clearer.
Change-Id: Ib69703c13b9bbfae12eaca7fab6a073cdeea4c54
Signed-off-by: Michael Vorburger <vorburger@redhat.com>
Sam Hague [Mon, 10 Jul 2017 19:35:55 +0000 (15:35 -0400)]
Use right version for statistics pom
Change-Id: I71099836cb81f55c9a590a565b4d244e7ed47d3a
Signed-off-by: Sam Hague <shague@redhat.com>
Sam Hague [Mon, 10 Jul 2017 18:58:15 +0000 (14:58 -0400)]
Fix cherry-pick of 56902
Change-Id: I5b4ab5f5d0d80fcd31ff844734fb8931dda404dc
Signed-off-by: Sam Hague <shague@redhat.com>
Sam Hague [Mon, 10 Jul 2017 18:56:35 +0000 (14:56 -0400)]
Fix cherry-pick of 56875
Change-Id: I91c94af8343b5c0e0ec3841edd92f1a5693dd939
Signed-off-by: Sam Hague <shague@redhat.com>
Sam Hague [Mon, 10 Jul 2017 19:02:49 +0000 (15:02 -0400)]
Fix cherry-pick of 58749
Change-Id: Ie134f9b5925b3c1e40ff89519ec58e20bbe9d8b2
Signed-off-by: Sam Hague <shague@redhat.com>
Josh [Thu, 6 Jul 2017 07:52:42 +0000 (10:52 +0300)]
Fix funny character in ebgp.yang (slanted quotation mark)
Change-Id: If8b5aa13c1884685f0d283ad2e622be42aa29f22
Signed-off-by: Josh <jhershbe@redhat.com>
(cherry picked from commit
e9421c8ae44ce642277d37f80c6d4f00f3eb8011)
kiranvasudeva [Tue, 27 Jun 2017 08:56:36 +0000 (14:26 +0530)]
Adding Log statements, helps during debugging
Change-Id: Iaff8515aeff1fed46b4552029c3520ed79015826
Signed-off-by: kiranvasudeva <kirankumar.v@altencalsoftlabs.com>
Michael Vorburger [Tue, 4 Jul 2017 22:55:20 +0000 (00:55 +0200)]
Bug 8800: Fix AclServiceStatefulTest newInterfaceWithMultipleAcl()
The expected objects of that tests were broken, but due to Bug 8800, the
test did not fail because of that. (This needs to be reviewed and
merged before the actual root cause fix for Bug 8800 can go it.)
Change-Id: I6e9f0a9f42a36cf242c8b43988a76c2a747f2178
Signed-off-by: Michael Vorburger <vorburger@redhat.com>
Aswin Suryanarayanan [Wed, 12 Jul 2017 18:14:19 +0000 (23:44 +0530)]
Bug 8838 - aclservice NPE's
Fixes NPE in programConntrackRecircRules in Egress and Ingress Acl
serivce.
Interface is now deleted from the cache after notify is called.
Change-Id: I42d6981f36e1309dc70e721e76b91176cae01319
Signed-off-by: Aswin Suryanarayanan <asuryana@redhat.com>
Aswin Suryanarayanan [Thu, 13 Jul 2017 18:17:13 +0000 (23:47 +0530)]
Bug 8838 - aclservice NPE's
Ignoring notification for ports which are not present in AclInterface
cache.
Change-Id: I8492eb356482d0d84afe5843db74c9e0ac1496b7
Signed-off-by: Aswin Suryanarayanan <asuryana@redhat.com>
Michael Vorburger [Thu, 29 Jun 2017 13:41:07 +0000 (15:41 +0200)]
Fix NullPointerException in QosInterfaceStateChangeListener
2017-06-27 18:29:37,596 | ERROR | eChangeHandler-0 |
QosInterfaceStateChangeListener | 344 -
org.opendaylight.netvirt.qosservice-impl - 0.4.1.SNAPSHOT |
Qos:Exception caught in Interface Operational State Up event
java.lang.NullPointerException: the Function passed to
Optional.transform() must not return null.
at
com.google.common.base.Preconditions.checkNotNull(Preconditions.java:226)[38:com.google.guava:18.0.0]
at
com.google.common.base.Present.transform(Present.java:71)[38:com.google.guava:18.0.0]
at
org.opendaylight.netvirt.qosservice.QosInterfaceStateChangeListener.getNeutronPort(QosInterfaceStateChangeListener.java:117)[344:org.opendaylight.netvirt.qosservice-impl:0.4.1.SNAPSHOT]
Change-Id: Ic2a6750ac605dedeb61572689f62a97c811b8fa2
Signed-off-by: Michael Vorburger <vorburger@redhat.com>
(cherry picked from commit
a32ac7563dda41f0b581b4e93826bee7ff4bee20)
Michael Vorburger [Mon, 3 Jul 2017 17:06:08 +0000 (19:06 +0200)]
Use manual setters instead @Immutable in IdentifiedAceBuilder.xtend
The mix of Immutables.org with Xtend.org seemed cool when I originally
came up with this 1 year ago, but after having run into issues with
in-IDE compilation order in Eclipse EVERY TIME I want to touch something
in aclservice, this got enough on my nerves to finally simplify this.
The issue basically is that the Xtend IDE and Eclipse APT processor seem
to struggle about the correct order in which Xtend active annotations
and APT processors should run (although it did, always, work, reliably,
in the mvn CLI build); Eclipse IDE could never get this right, and there
was too often red ink errors for (supposedly) missing
ImmutableIdentifiedAceBuilder... so basically this mix of Xtend with
Immutables.org probably wasn't a good idea.
The Xtend active annotations @Accessor (as well as @Data) really do
something quite similar to Immutables.org (yet let us keep the "pretty"
object literal initialization syntax of Xtend); however they generate
"real" setters, which do not "return this;", so instead it's simplest to
just hand write those few setters, for now. Later if this pattern is
used more widely, we could have a new Xtend active annotation, such as
@Builder.
This also, and strictly speaking kind of unrelated, changes the double
indirection of Builder -> Immutable -> DataObject to a simpler to
understand single level, without the intermediate Immutable (which makes
these Builders used to compose objects more easily in tests not thread
safe, which is perfectly fine, actually).
dito for IdentifiedInterfaceWithAclBuilder
Change-Id: Ia70f5f62fa9d3d71da0521cd22808907c97a4619
Signed-off-by: Michael Vorburger <vorburger@redhat.com>
Philippe Guibert [Thu, 30 Mar 2017 09:51:16 +0000 (10:51 +0100)]
Spec to support dualstack VMs in L3VPN
This specification brings the ability for a BGPVPN configured VPN to
distribute IPv4 and IPv6 addresses within the same VPN, and by using two
distinct routers to respectively handle IPv4 and IPv6 routing issues.
Change-Id: Ia46965797d35a3493b3065c6725aea1501f3941f
Signed-off-by: Philippe Guibert <philippe.guibert@6wind.com>
jenkins-releng [Fri, 14 Jul 2017 12:49:20 +0000 (12:49 +0000)]
Bump versions by x.y.(z+1)
Change-Id: I4c23006c1ef00c8233f637ad97217aaf3b2ae72a
Signed-off-by: jenkins-releng <jenkins-releng@opendaylight.org>
Shashidhar Raja [Tue, 4 Jul 2017 05:02:41 +0000 (10:32 +0530)]
Bug 8793: Null Pointer Exceptions in Carbon CSIT
This commit fixes NPE observed in bindAcl() of AclService
Change-Id: I7397677f0e63a56e0e023eb4b2697ae1ef095714
Signed-off-by: Shashidhar Raja <shashidharr@altencalsoftlabs.com>
Aswin Suryanarayanan [Mon, 3 Jul 2017 18:51:09 +0000 (00:21 +0530)]
Bug 8793: Null Pointer Exceptions in Carbon CSIT
Fixes NPE in programConntrackRecircRules and syncSpecificAclFlow in
AclService
Change-Id: I3217efdb82d35acd1de34699edbb01ef8820a7fd
Signed-off-by: Aswin Suryanarayanan <asuryana@redhat.com>
Vivekanandan Narasimhan [Mon, 3 Jul 2017 14:39:13 +0000 (20:09 +0530)]
Bug 8793: Null Pointer Exceptions in Carbon CSIT
Many NPEs are seen in Carbon CSIT on ACLService
and NAT Service.
This review fixes some of them.
Change-Id: I8949da5752a99825f8a0754be6349f4b7139273d
Signed-off-by: Vivekanandan Narasimhan <n.vivekanandan@ericsson.com>
Tarun Thakur [Fri, 23 Jun 2017 09:44:25 +0000 (15:14 +0530)]
Adding logs in ACL.
Adding logs in ACL which can be helpful for debugging.
Change-Id: I29b9ed92cedd06e4c325117e4419b01e190dc217
Signed-off-by: Tarun Thakur <tarun.t@altencalsoftlabs.com>
Sam Hague [Sun, 2 Jul 2017 20:53:22 +0000 (16:53 -0400)]
Fix NPE in aclService test
Change-Id: I859f5a1a188c67cc83fa429e3b6abe7917ebedc1
Signed-off-by: Sam Hague <shague@redhat.com>
Michael Vorburger [Thu, 22 Jun 2017 17:37:41 +0000 (19:37 +0200)]
Fix clear bug in BgpConfigurationManager's interVpnLinkService
The assignment to the field is just self-assigning the current value of
that field (null), not the injected argument, due to a typo... :-(
This could never have worked - isn't this kind of coded tested?! ;)
Found chance related to a closer look in the context of
https://git.opendaylight.org/gerrit/#/c/59205/. We really should be
working on enabling FindBugs (which, probably, would have spotted this)
on more projects in netvirt... is anyone volunteering to take this on,
if I can provide some coaching and would do reviews?
Change-Id: I0f8cffbd384262a5e68de523725ee3efa1564722
Signed-off-by: Michael Vorburger <vorburger@redhat.com>
karthik.p [Fri, 30 Jun 2017 07:14:32 +0000 (12:44 +0530)]
Revert "Neutron Port allocation for DHCP Service"
This reverts commit
30cba431cf065d25897dffe01cd5c4febc63e14a.
Change-Id: I38bd7587287ef23eb71cabc551983b74a4a6d0c9
Signed-off-by: karthik.p <karthik.p@altencalsoftlabs.com>
Signed-off-by: Achuth Maniyedath <achuth.m@altencalsoftlabs.com>
epgoraj [Fri, 23 Jun 2017 10:17:42 +0000 (15:47 +0530)]
BUG 7824: Handling Optimistic lock exception for elanmodule
Change-Id: Ia99b70eb9dd3798cb5a82dce6e07f6ab6b980eef
Signed-off-by: epgoraj <p.govinda.rajulu@ericsson.com>
Stephen Kitt [Fri, 23 Jun 2017 13:43:27 +0000 (15:43 +0200)]
Remove sfc-translator MdsalUtils::put()
Replace MdsalUtils::put() with
SingleTransactionDataBroker::syncWrite(). MdsalUtils is no longer used
at all and can be removed.
Change-Id: I7237b60fe63534591991bf4c2e559ab796f30aac
Signed-off-by: Stephen Kitt <skitt@redhat.com>
Signed-off-by: Sam Hague <shague@redhat.com>
Stephen Kitt [Fri, 23 Jun 2017 13:37:18 +0000 (15:37 +0200)]
Remove sfc-translator MdsalUtils::merge()
Replace MdsalUtils::merge() with
SingleTransactionDataBroker::syncUpdate().
Change-Id: Ic3fab14043525bb105a59515319a85e2d3a1a146
Signed-off-by: Stephen Kitt <skitt@redhat.com>
Signed-off-by: Sam Hague <shague@redhat.com>
Stephen Kitt [Fri, 23 Jun 2017 13:22:24 +0000 (15:22 +0200)]
Remove sfc-translator MdsalUtils::delete()
Replace MdsalUtils::delete() with
SingleTransactionDataBroker::syncDelete().
Change-Id: I502129e932c263b6210aa3c699250582b7ef8d84
Signed-off-by: Stephen Kitt <skitt@redhat.com>
Signed-off-by: Sam Hague <shague@redhat.com>
Stephen Kitt [Fri, 23 Jun 2017 13:17:00 +0000 (15:17 +0200)]
Remove sfc-translator MdsalUtils::read()
Replace MdsalUtils::read() with
SingleTransactionDataBroker::syncReadOptionalAndTreatReadFailedExceptionAsAbsentOptional()
(which is itself deprecated, but I didn’t want to redo the
error-handling in this patch).
Change-Id: I8233ce003f416a6d9575884b5592d0d5728e3f05
Signed-off-by: Stephen Kitt <skitt@redhat.com>
Signed-off-by: Sam Hague <shague@redhat.com>
Stephen Kitt [Fri, 23 Jun 2017 09:16:52 +0000 (11:16 +0200)]
Remove NatUtil::read()
As a step in the MD-SAL utilities clean-up, remove NatUtil::read(),
simply replacing it with
SingleTransactionDataBroker::syncReadOptionalAndTreatReadFailedExceptionAsAbsentOptional()
(which is itself deprecated, but I didn’t want to redo the
error-handling in this patch).
Change-Id: I6798e460d7fb61653a5e3f754d5c083fe1084c80
Signed-off-by: Stephen Kitt <skitt@redhat.com>
Signed-off-by: Sam Hague <shague@redhat.com>
Stephen Kitt [Thu, 22 Jun 2017 15:36:29 +0000 (17:36 +0200)]
Use documenting constants for put()
WriteTransaction provides constants for put()’s createMissingParents
parameter, this patches uses them throughout.
A few transaction-related collection constructions are replaced with
singletonList().
Change-Id: I5e77870eb2b898eb4a4e364ba0013496c884aa1e
Signed-off-by: Stephen Kitt <skitt@redhat.com>
Signed-off-by: Sam Hague <shague@redhat.com>
Stephen Kitt [Thu, 22 Jun 2017 12:37:12 +0000 (14:37 +0200)]
Remove old NetVirt
This is no longer used, let’s just remove it.
Change-Id: Idc36a582a39279caace8981e235a3ebb426a6af4
Signed-off-by: Stephen Kitt <skitt@redhat.com>
(cherry picked from commit
6ce3e7cc4ad11e10c33d62497e32ea5303724972)
Sridhar Gaddam [Mon, 19 Jun 2017 06:33:26 +0000 (12:03 +0530)]
Fix NPE triggered after disabling SG on a port
When port-security is disabled on a Neutron port, removeAclInterfaceMap
is invoked, but the entry in aclInterfaceMap is not deleted as there is
a difference (i.e., securityGroups element) in the port info present in
aclInterfaceMap and the one passed as argument. Because of this, the stale
entries remain in the aclInterfaceMap and are causing NPE issues for
subsequent operations. This patch fixes this issue by matching on the
interfaceId and deleting the corresponding entry.
Change-Id: I2c98f82beab8043052c9444f40f02eb26e8c3945
Signed-off-by: Sridhar Gaddam <sgaddam@redhat.com>
Shashidhar Raja [Wed, 7 Jun 2017 13:16:17 +0000 (18:46 +0530)]
Bug 8398 - communication between two tenant network failed while using
default SG (stateful mode)
This commit fixes issues observed in
https://git.opendaylight.org/gerrit/#/c/57705/
Depends-On:
0c01eacb8100ee5a224d52e44362fce98bd875f5
Change-Id: I827e99835c36509bad5b48f03034103d377102f8
Signed-off-by: Shashidhar Raja <shashidharr@altencalsoftlabs.com>
eupakir [Tue, 23 May 2017 14:50:01 +0000 (20:20 +0530)]
Bug 8398 - communication between two tenant network failed while
using default SG (stateful mode)
This fixes the remote ACL indirection flows for the case where a
router/bgpvpn is involved in communication. It enables the dispatcher
tables to write vpnId to the metadata so that Remote ACL flows can
correctly match and send the packet to the intended destination.
The fix makes use of the vpn footprint service to program ACL flows for
every host addition/deletion.
Depends-On:
b3e21d27e8f593b4ca326dd93239db38e1ead4cd
Change-Id: I1cbee44f951a75d4bbb382fed4e882162016c946
Signed-off-by: eupakir <kiran.n.upadhyaya@ericsson.com>
Signed-off-by: Vivekanandan Narasimhan <n.vivekanandan@ericsson.com>
Gowri R [Wed, 28 Jun 2017 05:13:11 +0000 (10:43 +0530)]
Bug 8645: Alarm not cleared while deleting the bgp neighbor
Clearing the alarm during the deletion of bgp neighbor.
Change-Id: I0692b18c3850536ffac73decc0d6b4d23523e963
Signed-off-by: Gowri R <gowri.r@ericsson.com>
Akash [Wed, 10 May 2017 09:09:31 +0000 (14:39 +0530)]
bug 7599 hwvtep ucast mac add performance improv
Batching changes for Dmac flows.
Making sure all macs installed at lower case to resolve duplicate mac
entries issue.
Deleting entries related to logicalSwitch in case of
l2GwConnectionDeletion.(More explanation in inline comment)
Fixing minor bug in HA (Only parent nodes needs to be updated in cache not
child node hence returning)
ElanInstanceListner has been updated to cluster listner as in case of
cluster reboot / or other situation cache should be cleared form all
nodes.
Change-Id: I4cc767050ee7036b12b2c07c5f139b42e2970134
Signed-off-by: Akash <a.k.sahu@ericsson.com>
ehvkand [Wed, 21 Jun 2017 13:02:24 +0000 (18:32 +0530)]
VrfEntryListener refactoring using chain of responsibility pattern
This is a secod review where RouteInterface handling is separated into a
new handler class RouterInterfaceVrfEntryHandler.
Change-Id: If4bc6ded02e23dc12f6217a7c412ac5cc1188f50
Signed-off-by: ehvkand <hanamantagoud.v.kandagal@ericsson.com>
Achuth Maniyedath [Thu, 11 May 2017 18:32:50 +0000 (00:02 +0530)]
Neutron Port allocation for DHCP Service
The feature enables DHCP service to make use of the
newly created Neutron ports provisioned by Openstack
networking-odl driver at the time of Subnet creation.
Additionaly an ARPResponder flow is also introduced to
service the ARP requests for these ports during VM
DHCP renewal.
All ArpResponder handling for Router interface gateway,
floating IP and DHCP ports will be moved to ElanService,
introducing a pipeline change for ARP handling.
LPort Dispatcher Table(17) -> ARP Check Table(43) -> ARPResponder Group
-> ELAN Base Table(48), ARP Responder Table(81), Controller
-> Egress Dispatcher Tabel (220)
Change-Id: I8ebbe05c37f5e7e4c5ae1dd1d17faa12965d5e65
Depends-On: I2dce5e253d3c94180dd18e51125c5c18170b5c69
Signed-off-by: Achuth Maniyedath <achuth.m@altencalsoftlabs.com>
Signed-off-by: Vijayalakshmi Chickkamenahalli Nagaraju <vijayalakshmi.c@altencalsoftlabs.com>
Signed-off-by: karthik.p <karthik.p@altencalsoftlabs.com>
Signed-off-by: Achuth Maniyedath <achuth.m@altencalsoftlabs.com>
eupakir [Wed, 28 Jun 2017 11:23:28 +0000 (16:53 +0530)]
Bug 7939 - CSIT Sporadic failures - Flow(s) missing in VPNService suite on
compute node
Fix ensures that table-19 is completely cleaned up on swap, before add the
vpn and router interfaces to the bgpvpn/routervpn instance
Change-Id: I9b106492234caa80ba9b2d961bb05a728cf5af27
Signed-off-by: eupakir <kiran.n.upadhyaya@ericsson.com>
Signed-off-by: Vivekanandan Narasimhan <n.vivekanandan@ericsson.com>
arthi.b [Tue, 20 Jun 2017 23:01:33 +0000 (19:01 -0400)]
BUG 8553: Rules are not deleted/added to the SG.
Problem: Event not processed when we delete last rule
and also addition of first rule to the SG.
Fix: Skip only if both before and after ACL list are empty
Change-Id: I7f8ccb10adbdb1b8c1cab0ca7af8a2dcc2105aea
Signed-off-by: arthi.b <arthi.b@hcl.com>
Vivekanandan Narasimhan [Sat, 24 Jun 2017 09:37:11 +0000 (15:07 +0530)]
Bug 8743 : Fix NAT Service to safely use vpn-to-dpn list.
Description : Its been observed the vpn-to-dpn list been not updated
properly resulting in missing dpn-state with active state. As a result,
installation of default route rule been missing resulting in SNAT/DNAT
failure.
Changes are done to make use of vpn footprint service to update the
vpn-to-dpn list whenever floating-ip and external fixed ips are added
and deleted.
Change-Id: Ic55fc5364ef45fbbe7ce63a9feef51bf39332541
Signed-off-by: Vivekanandan Narasimhan <n.vivekanandan@ericsson.com>
Signed-off-by: cgowdru <chetan.arakere@altencalsoftlabs.com>
Michael Vorburger [Mon, 19 Jun 2017 11:21:52 +0000 (13:21 +0200)]
Move fibmanager-shell blueprint.xml from OSGI-INF/blueprint to org/opendaylight/blueprint
This may (?) fix the SFT failing to find the DataBroker, as discussed on follow-up to
https://lists.opendaylight.org/pipermail/netvirt-dev/2017-June/004770.html; and if
this does fix that SFT problem, then this was an indirect impact of
https://git.opendaylight.org/gerrit/#/c/54885/.
This also reverts both https://git.opendaylight.org/gerrit/#/c/59105/ private field,
and the availability="optional" from https://git.opendaylight.org/gerrit/#/c/59051/
Change-Id: I856c0f4311a02a5380512496619aee74cf70ee2e
Signed-off-by: Michael Vorburger <vorburger@redhat.com>
(cherry picked from commit
b0a5086f16acf9d9bd5a38eea1a0715625744aa0)
Abhinav Gupta [Tue, 20 Jun 2017 08:44:51 +0000 (14:14 +0530)]
Netvirt wide blueprint xml path change: /OSGI-INF/ -> /org/opendaylight/
Also have removed fake "availability=optional"
Change-Id: Iae7caa155eeffa7c831822ac600bbc6283a7e2fc
Signed-off-by: Abhinav Gupta <abhinav.gupta@ericsson.com>
Gobinath [Tue, 20 Jun 2017 14:56:25 +0000 (20:26 +0530)]
Bug 8179 - Data validation failed for path bgp/networks
Issue:
Data validation failed exception during withdrawing prefix from BGP.
Analysis:
The issue occured while trying to delete an element from the
bgp/networks which was not in the desired format. The DPNid is
used as rd in case of extra routes with internal VPN. The prefixes in
the internal VPN are not be advertised to the BGP. The condition for
this already exists(rd is the vpnName here). For extra routes, DPNIDs
are used as rds for ECMP and they should also be advertised. This was
already handled but there was no handling to avoid the withdrawal from
the BGP in case of the removal.
Fix:
Condition has been added to avoid calling withdrawPrefix() for the extra
routes removal in case of the internal VPN.
Note: Also some other errors in logging have been fixed.
Change-Id: Ic5dbcf951cd36b0a0fb9624f4899184bc4e305f0
Signed-off-by: Gobinath <gobinath@ericsson.com>
Signed-off-by: Vivekanandan Narasimhan <n.vivekanandan@ericsson.com>
epgoraj [Tue, 20 Jun 2017 09:07:35 +0000 (14:37 +0530)]
Bug 7824: Handling CSIT Exception under Elan module
Change-Id: I2f34d558dc6887324cf5edef6551e1ee6451f816
Signed-off-by: epgoraj <p.govinda.rajulu@ericsson.com>
SivaKumar Perumalla [Tue, 13 Jun 2017 15:12:55 +0000 (15:12 +0000)]
Bug 8710: Bgp afi related bug fixes
Change-Id: Ie62474af854b2b4db5f53c57fa7e79499e163b21
Signed-off-by: SivaKumar Perumalla <sivakumar.perumalla@ericsson.com>
ehvkand [Wed, 14 Jun 2017 13:37:55 +0000 (19:07 +0530)]
VrfEntryListener refactoring using chain of responsibility pattern
This is a first review where BGP Route handling is separated into a new
handler class BgpRouteVrfEntryHandler.
All common methods are put into a new superclass i.e VrfEntryHandler. Each
of the subclasses like BgPRouteVrfEntryHandler will extend the superclass.
This is done to avoid circular depency on VrfEntryListener.
As part of chain of responsibility pattern , VrfEntryListener will pass on
the Vrf object to the first handler in a chain based on the RouteOrigin
type. Once the first handler completes processing the Vrf obj , it will
pass on the same to next handler.
In the current review BgpRouterVrfEntryHandler will process Routes of type
"RouteOrigin.BGP". BGP routes need not be processed by any other handler ,
hence next handler is not defined in the chain.
In the subsequent patches , when new handlers like
LocalRouteVrfEntryHandler, SubnetRouteVrfEntryHandler etc will have to
pass on the Vrf obj to next handler which will be
ImportExportVrfEntryHandler.
Change-Id: I538085bfcda7733b5f3ba5ffe49d541619ac40ec
Signed-off-by: ehvkand <hanamantagoud.v.kandagal@ericsson.com>