MCLAG feature for SONIC #2514

shine4chen · 2019-02-01T07:50:23Z

- What I did
submit mclag to sonic
Note: mclag feature need sonic-swss and sonic-utils submodules to do some enhancement. We will submit another PR for these enhancements

- How I did it
accomplish it according MCLAG HLD document

- How to verify it
Complie and test it on nephos lab

- Description for the changelog

MCLAG feature support

This PR must work with the following PRs.

sonic-swss:

[teammgrd]during warm-reboot teamd need to recover system-id from saved lacp-pdu sonic-swss#1003: [teammgrd]during warm-reboot teamd need to recover system-id from saved lacp-pdu
[fdborch] support mac update sonic-swss#877: [fdborch] support mac update
[intfsorch]:add support to change rif mac address sonic-swss#814: [intfsorch]:add support to change rif mac address
[mclag]:add mclagsyncd sonic-swss#811: [mclag]:add mclagsyncd
[aclorch]: add support for acl rule to match out port sonic-swss#810: [aclorch]: add support for acl rule to match out port
[portsorch]: add support to set mac-address learning attribute on bridge-port sonic-swss#809: [portsorch]: add support to set mac-address learning attribute on bridge-port

sonic-swss-common

Add schema for ACL: APP_ACL_TABLE and APP_ACL_RULE_TABLE (#325) sonic-swss-common#275: Add schema for ACL: APP_ACL_TABLE and APP_ACL_RULE_TABLE (Support centec platform #325)

sonic-utilities

add support for MCLAG sonic-utilities#453: add support for MCLAG

- A picture of a cute animal (not mandatory but encouraged)

lguohan · 2019-02-01T08:57:47Z

files/build_templates/docker_image_ctl.j2

@@ -137,6 +137,9 @@ start() {
        -v /usr/share/sonic/device/$PLATFORM:/usr/share/sonic/platform:ro \
 {%- if docker_container_name != "database" %}
        -v /usr/share/sonic/device/$PLATFORM/$HWSKU:/usr/share/sonic/hwsku:ro \
+{%- endif %}
+{%- if docker_container_name == "iccpd" %}
+        -v /lib/modules:/lib/modules:ro \


why do we need lib modules?

MCLAG uses lib module ebtable_filter and ebtables

can you document how ebtable_filter and ebtables are used in the design document, it needs to be review.

besides, even mclag is uses ebtable, it does not mean we need to mount /lib/modules into the docker.

@jianjundong, @shine4chen , can you please address this comment? Please raise a separate PR for this. Thanks.

We found that lib module ebtable_filter and ebtables are already installed in the docker default by someone, thus no need to do this here. These codes are removed by lguohan in PR '[docker-iccp]: do not mount kernel module into iccp container (#4372)'.

dockers/docker-iccpd/Dockerfile

lguohan · 2019-02-05T18:36:57Z

we are moving all containers to use stretch base, new containers should be stretch only. Please check #2524 for example.

jianjundong · 2019-02-13T08:33:44Z

we are moving all containers to use stretch base, new containers should be stretch only. Please check #2524 for example.

We will move ICCPD containers to use stretch base early as possible. Thanks!

update docker-iccpd to stretch-based container Signed-off-by: shine.chen <shine.chen@nephosinc.com>

Signed-off-by: shine.chen <shine.chen@nephosinc.com>

Fix some bugs per nephos test result Signed-off-by: shine.chen <shine.chen@nephosinc.com>

Signed-off-by: shine.chen <shine.chen@nephosinc.com>

Praveen-Brcm · 2019-07-15T22:32:24Z

src/iccpd/src/iccp_cmd.c

+    }
+    else if(strncmp(cp,"mclag_interface:",16) == 0)
+    {
+        cp += 10;


The length used here is incorrect, this needs to be updated to 16. Without this change LACP wont come up. In general can we #define values instead of hard code.

The length used here is incorrect, this needs to be updated to 16. Without this change LACP wont come up. In general can we #define values instead of hard code.

Thanks very much! We have fixed it several days ago, and will push to github ASAP.

Signed-off-by: shine.chen <shine.chen@nephosinc.com>

lisahnguyen · 2019-07-24T23:46:08Z

src/iccpd/src/iccp_netlink.c

+            }
+        }
+    }
+    else /*peerlink portchannel */


Both port learn_mode setting and src_port field in port-isolation mode setting are done using port-channel name instead of its members' names, would you please explain why we need to retrieve the peer-link port-channel members from the kernel. In other words, should we trigger learn_mode and port-isolation settings directly from set_peer_link() using just port-channel?

Both port learn_mode setting and src_port field in port-isolation mode setting are done using port-channel name instead of its members' names, would you please explain why we need to retrieve the peer-link port-channel members from the kernel. In other words, should we trigger learn_mode and port-isolation settings directly from set_peer_link() using just port-channel?

@lisahnguyen, Please see the codes in function update_peerlink_isolate_from_all_csm_lif(), the src port is peer-link port or portchannel, the dst ports are all the members of portchannel.

@jianjundong: since this section of the code handles peer-link which is always the source port (not dest ports) in port-isolation setting, we don't need the list of all member ports of a port-channel peer link. In the update_peerlink_isolate_from_all_csm_lif(), only dest ports are the list of port-channel's member ports

@lisahnguyen, learn-mode is set on peer-link portchannel, if one member add to or remove from portchannel, we must recover the learn-mode of this member, so we need the list of all member ports of a port-channel peer link. You are correct, we don't need to call update_peerlink_isolate_from_all_csm_lif() when the member list of peer-link changed.

@jianjungdong: There are two scenarios:

Scenario 1: an existing member is removed from a port-channel: in this case, we need to put back the learning mode on the actual port-channel member. Hence, set_peerlink_mlag_port_learn() should be called. I agree with this change.

Scenario 2: a new member is added to an existing port channel: since set_peerlink_mlag_port_learn(), sets learn mode on the port-channel (not the port-channel's members ports), is it correct that we should not call set_peerlink_mlag_port_learn() in this case?

@lisahnguyen, We are not sure if a new member is added to an existing port channel, the learn mode of the new member follows the port channel in all vendor's implementation. So we call set_peerlink_mlag_port_learn() in this case, the parameter is port channel if, not member if. Thanks.

lisahnguyen · 2019-07-25T00:21:07Z

src/iccpd/src/mlacp_fsm.c

+    /*Update port isolate after 3 seconds*/
+    if (csm->isolate_update_time != 0 && (time(NULL) - csm->isolate_update_time) >= 3)
+    {
+        /*update_peerlink_isolate_from_all_csm_lif(csm);*/


Based on this logic, we no longer apply a 3-sec delay for port-isolation setting. Should we remove all settings for isolatate_update_time to avoid making unneeded time() system calls?

Based on this logic, we no longer apply a 3-sec delay for port-isolation setting. Should we remove all settings for isolatate_update_time to avoid making unneeded time() system calls?

@lisahnguyen, OK, we will remove these. Thanks.

lisahnguyen · 2019-07-25T01:01:05Z

src/iccpd/src/mlacp_sync_prepare.c

+    if (vlan_id != NULL)
+        num_of_vlan_id++;
+
+    tlv_len = sizeof(struct mLACPPortChannelInfoTLV) + sizeof(struct VLAN_ID) * num_of_vlan_id;


tlv_len calculation is not correct. Only the size of the VLAN ID field (i.e. vid) should be counted instead of the entire struct VLAN_ID since only VLAN ID is sent to peer.

tlv_len calculation is not correct. Only the size of the VLAN ID field (i.e. vid) should be counted instead of the entire struct VLAN_ID since only VLAN ID is sent to peer.

@lisahnguyen, you are right, sizeof(struct VLAN_ID) must be sizeof(struct mLACPVLANData).

lisahnguyen · 2019-07-25T01:02:16Z

src/iccpd/src/mlacp_sync_prepare.c

+    tlv->icc_parameter.u_bit = 0;
+    tlv->icc_parameter.f_bit = 0;
+    tlv->icc_parameter.type = htons(TLV_T_MLACP_PORT_CHANNEL_INFO);
+    tlv->icc_parameter.len = htons(sizeof(struct mLACPPortChannelInfoTLV) - sizeof(ICCParameter) + sizeof(struct VLAN_ID) * num_of_vlan_id);


Incorrect length calculation (similar issue as tlv_len calculation)

Incorrect length calculation (similar issue as tlv_len calculation)

Thanks!

boralt

https://github.com/Azure/sonic-buildimage/blob/896aa416da40f188a3167f92f0a71b0a5956e866/src/iccpd/src/mlacp_sync_update.c#L187-L194

Question. Why for orphan interface name from peer should match any local interface name?

Praveen-Brcm · 2019-07-25T20:38:51Z

src/iccpd/src/mlacp_sync_update.c

+            /*If the interface is not exist, return*/
+            return -255;
+        }
+    }


does the Peer node interface is part of local If .? I assume not. Then this check would fail for orphan port mac learn if the incoming interface is not present here.
I see someone posted the same comment , please clarify.?

@Praveen-Brcm, you are right. For orphan port, name check is not needed.

Praveen-Brcm · 2019-07-25T20:42:11Z

src/iccpd/src/mlacp_sync_update.c

+
+                        /*Send mac add message to mclagsyncd*/
+                        add_mac_to_chip(mac_msg, MAC_TYPE_DYNAMIC);
+                    }


This update might overwrite the local static MAC. There is a check just right above this code for static mac , can we please consider moving this code inside that block or update mac only if existing mac is not static.

@Praveen-Brcm, Before ICCP submit to community, static MAC can not be configured, we can refine the MAC type later. Thanks.

lisahnguyen · 2019-07-25T22:16:57Z

src/iccpd/include/mlacp_tlv.h

+{
+    ICCParameter    icc_parameter;
+    uint8_t         type;/*add or del*/
+    char     mac_str[32];


For each MAC, we should send only 6-byte MAC address instead of the 32-byte MAC string to MLAG peer.
For L2 MCLAG, large number of MACs will be synched between MLAG peers. Hence, we plan to make the change as part of the L2 MCLAG enhancements and will contribute back to the community.

Your proposal is reasonable. This can be refine to 18-byte string. If define to 6-bytes, the receiver must transfer it to 18-byte MAC string. This is CPU consuming.

lisahnguyen · 2019-07-25T22:21:40Z

src/iccpd/src/mlacp_sync_prepare.c

+    mlacp_fill_icc_header(csm, icc_hdr, msg_len);
+
+    /* Prepare for ARP information TLV */
+    tlv = (struct mLACPMACInfoTLV*)malloc(tlv_len);


Why do we need to allocate memory for tlv structure? Stack variable can be used to avoid unneeded dynamic memory allocation to minimize memory fragmentation. We plan to address this as part of L2 MCLAG enhancements and contribute back to the community.

OK, we will refine this ASAP.

jianjundong · 2019-07-26T01:22:11Z

https://github.com/Azure/sonic-buildimage/blob/896aa416da40f188a3167f92f0a71b0a5956e866/src/iccpd/src/mlacp_sync_update.c#L187-L194

Question. Why for orphan interface name from peer should match any local interface name?

@boralt, you are right. For orphan port, name check is not needed.

boralt · 2019-07-26T19:09:19Z

https://github.com/Azure/sonic-buildimage/blob/896aa416da40f188a3167f92f0a71b0a5956e866/src/iccpd/src/mlacp_link_handler.c#L1501-L1506

Could you clarify where 0x7f000006 (127.0.0.6) and port 2626 are coming from. Also, there is warning message in this code that mentions unix socket, but it is network socket currently in-use here.

Praveen-Brcm · 2019-07-26T22:53:12Z

src/iccpd/src/mlacp_sync_update.c

+                mac_msg->age_flag &= ~MAC_AGE_PEER;
+                ICCPD_LOG_DEBUG(__FUNCTION__, "Recv ACK, Remove peer age flag:%d ifname  %s, add %s vlan-id %d, op_type %d",
+                                mac_msg->age_flag, mac_msg->ifname, mac_msg->mac_str, mac_msg->vid, mac_msg->op_type);
+            }


It looks the code to send the MAC_SYNC_ACK is being used, this unused code can be commented.
Can someone please clarify what is the original intention of having the MAC_SYNC_ACK, is this to handle MAC out of sync scenarios between McLAG peer nodes .? Also is there any plan to make the changes active .?

OK, we will delete these unused codes. Yes, the original intention of MAC_SYNC_ACK is to prevent MAC out of sync. If anyone find any scenario that MAC out of sync in current implementation, we will refine it.

Signed-off-by: shine.chen <shine.chen@mediatek.com>

shine4chen · 2019-10-25T06:32:41Z

retest this please

disaster123 · 2019-10-25T13:30:03Z

Is there a build available already containing MLAG? I need layer2 mlag for a test setup.

shine4chen · 2019-10-31T06:44:18Z

retest this please

Signed-off-by: shine.chen <shine.chen@mediatek.com>

rlhui · 2020-03-23T16:37:25Z

retest this please

rules/docker-iccpd.mk

Signed-off-by: shine.chen <shine.chen@mediatek.com>

abdosi · 2020-04-03T05:33:07Z

Approved based on my comments being addressed. Need to be approved by other reviewers too

rlhui · 2020-04-03T05:34:17Z

retest this please

jpsenior · 2020-04-15T01:55:58Z

dockers/docker-iccpd/iccpd.j2

+    mclag_interface:{{MC_LAG[mclag_id]['mclag_interface']}}
+{% endfor %}
+system_mac:{{DEVICE_METADATA['localhost']['mac']}}
+


Hello there - I am testing iccpd, and observet his jinja does not render in any correct data format, whitespace control might need some looking at.

root@leaf-1-1:/etc/iccpd# cat iccpd.conf mclag_id:1 local_ip: peer_ip:172.16.0.5 peer_link:PortChannel3 mclag_interface: system_mac:cc:37:ab:17:6a:00

@jpsenior, We checked the format of iccpd.j2 carefully, and can not replicate this issue.

* add mclag support(sonic-net#2514) Signed-off-by: shine.chen <shine.chen@nephosinc.com> * Fix show acl table can't display mclag acl rules Signed-off-by: shine.chen <shine.chen@nephosinc.com> * refine mclagsyncd code Signed-off-by: shine.chen <shine.chen@nephosinc.com> * change APP_ACL_TABLE_NAME to APP_ACL_TABLE_TABLE_NAME Signed-off-by: shine.chen <shine.chen@nephosinc.com> * optimize acl rule operation Signed-off-by: shine.chen <shine.chen@nephosinc.com> * refine function name using camelcased Signed-off-by: shine.chen <shine.chen@mediatek.com> * refine readData function per selectable base-class change Signed-off-by: shine.chen <shine.chen@mediatek.com> * refine sys log Signed-off-by: shine.chen <shine.chen@mediatek.com> * remove binary file Signed-off-by: shine.chen <shine.chen@mediatek.com> * set traffic isolation using ACL_TABLE_MCLAG Signed-off-by: shine.chen <shine.chen@mediatek.com> Co-authored-by: shine.chen <shine.chen@nephosinc.com> Co-authored-by: shine.chen <shine.chen@mediatek.com>

… latest HEAD sonic-swss: ``` * 13ef25bf - (HEAD -> master, origin/master, origin/HEAD) [teamd]: Clean teamd process if LAG creation fails (sonic-net#2888) (5 days ago) [Lawrence Lee] * ae010bfa - Support type7 encoded CAK key for macsec in config_db (sonic-net#2892) (13 days ago) [judyjoseph] * e6f134fb - [orchagent]: admin-disable port before setPortSerdesAttribute() (sonic-net#2831) (4 weeks ago) [Aman Singhal] * a67d4a77 - Change default branch to build_branch (sonic-net#2885) (4 weeks ago) [Ze Gan] * d44761cc - Make sure new binaries replace existing binaries in docker-sonic-vs (sonic-net#2870) (4 weeks ago) [Saikrishna Arcot] * 7102220a - [Fixbug]: Fix vnet attribute miss if route action is vnet_direct and vnet test cases (sonic-net#2873) (sonic-net#2877) (4 weeks ago) [Ze Gan] * 873455b7 - Merge pull request sonic-net#2878 from Pterosaur/enable_dash_vstest (4 weeks ago) [Guohan Lu] |\ | * 44457c86 - Simplify test task and remove duplicated artifacts (4 weeks ago) [Ze Gan] | * 8bca4ed2 - Trigger Azp (4 weeks ago) [Ze Gan] | * 6de56ee1 - Merge branch 'master' into enable_dash_vstest (4 weeks ago) [Ze Gan] | |\ | * | aad88a36 - Disable test_dash_crm (5 weeks ago) [Ze Gan] | * | 61126eb0 - Enable Dash test in Pipeline (5 weeks ago) [Ze Gan] * | | ecd88108 - update portStatIds for cisco (sonic-net#2876) (4 weeks ago) [Zhixin Zhu] * | | f1294999 - [ppi] Relax port attributes validation (sonic-net#2872) (4 weeks ago) [Nazarii Hnydyn] | |/ |/| * | b4fcfc9f - Remove fabric queue counters. (sonic-net#2862) (5 weeks ago) [jfeng-arista] |/ * bb99f418 - Merge pull request sonic-net#2856 from theasianpianist/master-dash-merge (5 weeks ago) [Guohan Lu] |\ | * 84b32af2 - Merge branch 'master' into master-dash-merge (5 weeks ago) [Lawrence Lee] | |\ | |/ |/| * | ca728200 - [FEC] Adding support of override based on attribute query of SAI_PORT_ATTR_AUTO_NEG_FEC_MODE_OVERRIDE (sonic-net#2874) (5 weeks ago) [Sudharsan Dhamal Gopalarathnam] / * 3bb71809 - Merge branch 'master' into master-dash-merge (5 weeks ago) [Lawrence Lee] * 574940dd - Merge branch 'master' into master-dash-merge (6 weeks ago) [Lawrence Lee] * 660e5e4c - Merge branch 'master' into master-dash-merge (6 weeks ago) [Lawrence Lee] * 6d941746 - Merge branch 'master' into master-dash-merge (6 weeks ago) [Lawrence Lee] * c87c86e6 - [dash]: Refactor DASH orch by protobuf format (sonic-net#2722) (8 weeks ago) [Ze Gan] * c999ea32 - [tests]: Change DVS ENV HWKSU to DPU-2P for DASH vstest (sonic-net#2847) (8 weeks ago) [prabhataravind] * b2c25dcd - (conflict)[dash] Improve dash orchagent ZMQ code. (sonic-net#2836) (8 weeks ago) [Hua Liu] * 409b3833 - [tests]: Set HWSKU to NPU-2P for dash vstests (sonic-net#2833) (8 weeks ago) [prabhataravind] * f2365af9 - (conflict)Enable/disable Zmq by parameter (sonic-net#2828) (8 weeks ago) [Hua Liu] * 3ade5fc3 - (conflict)[dash] Change dash orchagent from Redis consumer state table to ZMQ consumer state table. (sonic-net#2779) (8 weeks ago) [Hua Liu] * eaf1bb85 - [crm]: Remove NOT_IMPLEMENTED checks (8 weeks ago) [Lawrence Lee] * 916d2f10 - [azp]: Don't run DASH tests for regular test runs (8 weeks ago) [Lawrence Lee] * 208e80bd - update azp to use public pipeline artifacts (8 weeks ago) [Lawrence Lee] * 2168554d - [dash][ci] fix build pipeline (8 weeks ago) [Yakiv Huryk] * b6036635 - [dash][ci] update build pipeline to build with bullseye (8 weeks ago) [Yakiv Huryk] * 37a61ddb - (conflict)[azp] Add DASH to PR trigger for non-DASH VS tests (sonic-net#2813) (8 weeks ago) [Lawrence Lee] * 37d27b01 - Fix Dash orchagent build issue. (sonic-net#2788) (8 weeks ago) [Hua Liu] * 07cce313 - (conflict)[CRM][DASH] Extend CrmOrch to support DASH resources. (sonic-net#2739) (8 weeks ago) [Oleksandr Ivantsiv] * 7c435d1e - [DASH]: Miscellaneous bug fixes and adding vstests (sonic-net#2745) (8 weeks ago) [prabhataravind] * 6613dd4f - [dash]: Check if overlay IP is specified (sonic-net#2741) (8 weeks ago) [Lawrence Lee] * c863d48b - [dash] Do not use an action drop with the inbound routing table. (sonic-net#2710) (8 weeks ago) [Oleksandr Ivantsiv] * 89ce4e0c - [dash]: Don't attempt to bind empty ACL groups (sonic-net#2613) (8 weeks ago) [Lawrence Lee] * 8ec36a6f - (conflict)[dash]: ACL orchagent (sonic-net#2470) (8 weeks ago) [Ze Gan] * 29c23b12 - [DASH] Fix compilation issue caused by merge from the master branch. (sonic-net#2594) (8 weeks ago) [Oleksandr Ivantsiv] * fd3539e5 - [DASH] Add retry logic for VNET mapping table (sonic-net#2583) (8 weeks ago) [Lawrence Lee] * 9b179c07 - [dash] add USE_DST_VNET_VNI attribute to CA-to-PA entry (sonic-net#2533) (8 weeks ago) [Yakiv Huryk] * f7fe55fa - Add SAI_ENI_ATTR_VM_UNDERLAY_DIP and SAI_ENI_ATTR_VM_VNI attributes to ENI entry (sonic-net#2514) (8 weeks ago) [prabhataravind] * aa2a02c5 - (conflict)[Azp]: Add Azp for DASH (sonic-net#2501) (8 weeks ago) [Ze Gan] * 2d1972f2 - (conflict)orchagent: DASH changes (sonic-net#2459) (8 weeks ago) [prabhataravind] * acf0fe42 - [DPU] Fix unit tests compilation after merge from master branch. (sonic-net#2478) (8 weeks ago) [Oleksandr Ivantsiv] * 22c62f63 - (conflict)[DPU] Simplify SWSS initialization to meet DPU requirements. (sonic-net#2440) (8 weeks ago) [Oleksandr Ivantsiv] ``` sonic-sairedis ``` * cfa8da4 - (HEAD -> master, origin/master, origin/HEAD) Add extra parameter to pass vendor LDFLAGS for libsai.so (sonic-net#1291) (3 days ago) [Kamil Cudnik] * 8046908 - [CRM][DASH] Add the possibility of querying availability for OIDs. (sonic-net#1245) (5 days ago) [Oleksandr Ivantsiv] * 9547060 - Install nlohmann-json3-dev package for codeql (sonic-net#1290) (10 days ago) [Saikrishna Arcot] * f3b4dd5 - Use json.hpp from nlohmann-json-dev instead of swss-common (sonic-net#1289) (11 days ago) [Saikrishna Arcot] * 40c9d13 - [azp] Update az pipeline for swss docker to add syslog (sonic-net#1287) (2 weeks ago) [Kamil Cudnik] * 4c2527f - port counter support on sonic-vs (sonic-net#1275) (3 weeks ago) [Vishnu Shetty] * 92c58cf - [Azp]: Change default branch to build_branch (sonic-net#1279) (4 weeks ago) [Ze Gan] * 7178fb6 - [submodule] Update SAI to latest v1.12 branch (sonic-net#1284) (4 weeks ago) [Oleksandr Ivantsiv] * 52247b9 - [syncd] Fix missing comma (sonic-net#1278) (4 weeks ago) [Kamil Cudnik] * 44cd8c4 - [azp] Attempt to fix swss missing libs (sonic-net#1277) (4 weeks ago) [Ze Gan] * ee308bb - [submodule] Update SAI to latest v1.12 branch (sonic-net#1272) (5 weeks ago) [Kamil Cudnik] ``` Signed-off-by: Ze Gan <ganze718@gmail.com>

… latest HEAD (#16599) sonic-swss: ``` * 13ef25bf - (HEAD -> master, origin/master, origin/HEAD) [teamd]: Clean teamd process if LAG creation fails (#2888) (5 days ago) [Lawrence Lee] * ae010bfa - Support type7 encoded CAK key for macsec in config_db (#2892) (13 days ago) [judyjoseph] * e6f134fb - [orchagent]: admin-disable port before setPortSerdesAttribute() (#2831) (4 weeks ago) [Aman Singhal] * a67d4a77 - Change default branch to build_branch (#2885) (4 weeks ago) [Ze Gan] * d44761cc - Make sure new binaries replace existing binaries in docker-sonic-vs (#2870) (4 weeks ago) [Saikrishna Arcot] * 7102220a - [Fixbug]: Fix vnet attribute miss if route action is vnet_direct and vnet test cases (#2873) (#2877) (4 weeks ago) [Ze Gan] * 873455b7 - Merge pull request #2878 from Pterosaur/enable_dash_vstest (4 weeks ago) [Guohan Lu] |\ | * 44457c86 - Simplify test task and remove duplicated artifacts (4 weeks ago) [Ze Gan] | * 8bca4ed2 - Trigger Azp (4 weeks ago) [Ze Gan] | * 6de56ee1 - Merge branch 'master' into enable_dash_vstest (4 weeks ago) [Ze Gan] | |\ | * | aad88a36 - Disable test_dash_crm (5 weeks ago) [Ze Gan] | * | 61126eb0 - Enable Dash test in Pipeline (5 weeks ago) [Ze Gan] * | | ecd88108 - update portStatIds for cisco (#2876) (4 weeks ago) [Zhixin Zhu] * | | f1294999 - [ppi] Relax port attributes validation (#2872) (4 weeks ago) [Nazarii Hnydyn] | |/ |/| * | b4fcfc9f - Remove fabric queue counters. (#2862) (5 weeks ago) [jfeng-arista] |/ * bb99f418 - Merge pull request #2856 from theasianpianist/master-dash-merge (5 weeks ago) [Guohan Lu] |\ | * 84b32af2 - Merge branch 'master' into master-dash-merge (5 weeks ago) [Lawrence Lee] | |\ | |/ |/| * | ca728200 - [FEC] Adding support of override based on attribute query of SAI_PORT_ATTR_AUTO_NEG_FEC_MODE_OVERRIDE (#2874) (5 weeks ago) [Sudharsan Dhamal Gopalarathnam] / * 3bb71809 - Merge branch 'master' into master-dash-merge (5 weeks ago) [Lawrence Lee] * 574940dd - Merge branch 'master' into master-dash-merge (6 weeks ago) [Lawrence Lee] * 660e5e4c - Merge branch 'master' into master-dash-merge (6 weeks ago) [Lawrence Lee] * 6d941746 - Merge branch 'master' into master-dash-merge (6 weeks ago) [Lawrence Lee] * c87c86e6 - [dash]: Refactor DASH orch by protobuf format (#2722) (8 weeks ago) [Ze Gan] * c999ea32 - [tests]: Change DVS ENV HWKSU to DPU-2P for DASH vstest (#2847) (8 weeks ago) [prabhataravind] * b2c25dcd - (conflict)[dash] Improve dash orchagent ZMQ code. (#2836) (8 weeks ago) [Hua Liu] * 409b3833 - [tests]: Set HWSKU to NPU-2P for dash vstests (#2833) (8 weeks ago) [prabhataravind] * f2365af9 - (conflict)Enable/disable Zmq by parameter (#2828) (8 weeks ago) [Hua Liu] * 3ade5fc3 - (conflict)[dash] Change dash orchagent from Redis consumer state table to ZMQ consumer state table. (#2779) (8 weeks ago) [Hua Liu] * eaf1bb85 - [crm]: Remove NOT_IMPLEMENTED checks (8 weeks ago) [Lawrence Lee] * 916d2f10 - [azp]: Don't run DASH tests for regular test runs (8 weeks ago) [Lawrence Lee] * 208e80bd - update azp to use public pipeline artifacts (8 weeks ago) [Lawrence Lee] * 2168554d - [dash][ci] fix build pipeline (8 weeks ago) [Yakiv Huryk] * b6036635 - [dash][ci] update build pipeline to build with bullseye (8 weeks ago) [Yakiv Huryk] * 37a61ddb - (conflict)[azp] Add DASH to PR trigger for non-DASH VS tests (#2813) (8 weeks ago) [Lawrence Lee] * 37d27b01 - Fix Dash orchagent build issue. (#2788) (8 weeks ago) [Hua Liu] * 07cce313 - (conflict)[CRM][DASH] Extend CrmOrch to support DASH resources. (#2739) (8 weeks ago) [Oleksandr Ivantsiv] * 7c435d1e - [DASH]: Miscellaneous bug fixes and adding vstests (#2745) (8 weeks ago) [prabhataravind] * 6613dd4f - [dash]: Check if overlay IP is specified (#2741) (8 weeks ago) [Lawrence Lee] * c863d48b - [dash] Do not use an action drop with the inbound routing table. (#2710) (8 weeks ago) [Oleksandr Ivantsiv] * 89ce4e0c - [dash]: Don't attempt to bind empty ACL groups (#2613) (8 weeks ago) [Lawrence Lee] * 8ec36a6f - (conflict)[dash]: ACL orchagent (#2470) (8 weeks ago) [Ze Gan] * 29c23b12 - [DASH] Fix compilation issue caused by merge from the master branch. (#2594) (8 weeks ago) [Oleksandr Ivantsiv] * fd3539e5 - [DASH] Add retry logic for VNET mapping table (#2583) (8 weeks ago) [Lawrence Lee] * 9b179c07 - [dash] add USE_DST_VNET_VNI attribute to CA-to-PA entry (#2533) (8 weeks ago) [Yakiv Huryk] * f7fe55fa - Add SAI_ENI_ATTR_VM_UNDERLAY_DIP and SAI_ENI_ATTR_VM_VNI attributes to ENI entry (#2514) (8 weeks ago) [prabhataravind] * aa2a02c5 - (conflict)[Azp]: Add Azp for DASH (#2501) (8 weeks ago) [Ze Gan] * 2d1972f2 - (conflict)orchagent: DASH changes (#2459) (8 weeks ago) [prabhataravind] * acf0fe42 - [DPU] Fix unit tests compilation after merge from master branch. (#2478) (8 weeks ago) [Oleksandr Ivantsiv] * 22c62f63 - (conflict)[DPU] Simplify SWSS initialization to meet DPU requirements. (#2440) (8 weeks ago) [Oleksandr Ivantsiv] ``` sonic-sairedis ``` * cfa8da4 - (HEAD -> master, origin/master, origin/HEAD) Add extra parameter to pass vendor LDFLAGS for libsai.so (#1291) (3 days ago) [Kamil Cudnik] * 8046908 - [CRM][DASH] Add the possibility of querying availability for OIDs. (#1245) (5 days ago) [Oleksandr Ivantsiv] * 9547060 - Install nlohmann-json3-dev package for codeql (#1290) (10 days ago) [Saikrishna Arcot] * f3b4dd5 - Use json.hpp from nlohmann-json-dev instead of swss-common (#1289) (11 days ago) [Saikrishna Arcot] * 40c9d13 - [azp] Update az pipeline for swss docker to add syslog (#1287) (2 weeks ago) [Kamil Cudnik] * 4c2527f - port counter support on sonic-vs (#1275) (3 weeks ago) [Vishnu Shetty] * 92c58cf - [Azp]: Change default branch to build_branch (#1279) (4 weeks ago) [Ze Gan] * 7178fb6 - [submodule] Update SAI to latest v1.12 branch (#1284) (4 weeks ago) [Oleksandr Ivantsiv] * 52247b9 - [syncd] Fix missing comma (#1278) (4 weeks ago) [Kamil Cudnik] * 44cd8c4 - [azp] Attempt to fix swss missing libs (#1277) (4 weeks ago) [Ze Gan] * ee308bb - [submodule] Update SAI to latest v1.12 branch (#1272) (5 weeks ago) [Kamil Cudnik] ```

shine.chen added 4 commits January 31, 2019 23:02

MCLAG feature for sonic

97d51af

MCLAG feature for sonic

d99e9d6

Merge branch 'master' of https://github.com/shine4chen/sonic-buildimage

0dff4db

remove binary file

db361fd

lguohan reviewed Feb 1, 2019

View reviewed changes

dockers/docker-iccpd/Dockerfile Outdated Show resolved Hide resolved

remove unused dockerfile

fcc3421

update docker-iccpd to stretch-based container Signed-off-by: shine.chen <shine.chen@nephosinc.com>

jleveque added the Enhancement ➕ label Feb 15, 2019

shine.chen added 2 commits February 14, 2019 22:59

minor fix for isolation port setting

7b99945

Signed-off-by: shine.chen <shine.chen@nephosinc.com>

iccpd docker would start on demand

68621fd

Signed-off-by: shine.chen <shine.chen@nephosinc.com>

shine4chen force-pushed the master branch from e65628d to 68621fd Compare March 7, 2019 12:02

shine.chen added 3 commits July 2, 2019 02:51

merge iccpd to sonic master branch

4ebb4d0

Fix some bugs per nephos test result Signed-off-by: shine.chen <shine.chen@nephosinc.com>

Add x attribute on mclagdctl file

027a8df

Signed-off-by: shine.chen <shine.chen@nephosinc.com>

add warm-reboot support for MCLAG

1a9fb62

Signed-off-by: shine.chen <shine.chen@nephosinc.com>

Praveen-Brcm reviewed Jul 15, 2019

View reviewed changes

merge to master branch and reformat iccpd file

896aa41

Signed-off-by: shine.chen <shine.chen@nephosinc.com>

lisahnguyen reviewed Jul 24, 2019

View reviewed changes

lisahnguyen reviewed Jul 25, 2019

View reviewed changes

boralt reviewed Jul 25, 2019

View reviewed changes

Praveen-Brcm reviewed Jul 25, 2019

View reviewed changes

lisahnguyen reviewed Jul 25, 2019

View reviewed changes

Praveen-Brcm reviewed Jul 26, 2019

View reviewed changes

remove iccpd dependency with teamd

bc194ea

Signed-off-by: shine.chen <shine.chen@mediatek.com>

print log level when dump mclag status

0b08967

Signed-off-by: shine.chen <shine.chen@mediatek.com>

lisahnguyen previously approved these changes Nov 23, 2019

View reviewed changes

This was referenced Nov 26, 2019

[tests/mclag] Submit mclag testcases according to mclagtest plan sonic-net/sonic-mgmt#1237

Closed

[tests/mclag] Submit mclag testcases according to mclag test plan sonic-net/sonic-mgmt#1239

Open

William-zx mentioned this pull request Dec 2, 2019

[ansible/MCLAG]: add t0-mclag and t0-mclag-16 typologies support sonic-net/sonic-mgmt#1251

Open

1 task

rlhui self-assigned this Feb 24, 2020

rlhui assigned abdosi Mar 22, 2020

abdosi reviewed Mar 29, 2020

View reviewed changes

rules/docker-iccpd.mk Outdated Show resolved Hide resolved

abdosi reviewed Mar 29, 2020

View reviewed changes

rules/docker-iccpd.mk Show resolved Hide resolved

revise per community review

c4b6059

Signed-off-by: shine.chen <shine.chen@mediatek.com>

shine4chen dismissed stale reviews from lisahnguyen, laodulaodu, Praveen-Brcm, and gitsabari via c4b6059 March 30, 2020 07:02

Merge branch 'master' into master

f49200a

abdosi self-requested a review April 3, 2020 05:31

abdosi approved these changes Apr 3, 2020

View reviewed changes

rlhui merged commit 524cf9e into sonic-net:master Apr 4, 2020

jpsenior reviewed Apr 15, 2020

View reviewed changes

Praveen-Brcm mentioned this pull request Apr 5, 2021

MCLAG enhacements ICCPd initial code commit #4819

Merged

MCLAG feature for SONIC #2514

MCLAG feature for SONIC #2514

Conversation

shine4chen commented Feb 1, 2019 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

lguohan commented Feb 5, 2019

jianjundong commented Feb 13, 2019 • edited Loading

Choose a reason for hiding this comment

jianjundong Jul 16, 2019 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

jianjundong Jul 25, 2019 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

boralt left a comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

Choose a reason for hiding this comment

lisahnguyen Jul 25, 2019 • edited Loading

Choose a reason for hiding this comment

Choose a reason for hiding this comment

jianjundong commented Jul 26, 2019

boralt commented Jul 26, 2019

Choose a reason for hiding this comment

Choose a reason for hiding this comment

shine4chen commented Oct 25, 2019

disaster123 commented Oct 25, 2019

shine4chen commented Oct 31, 2019

rlhui commented Mar 23, 2020

abdosi commented Apr 3, 2020

rlhui commented Apr 3, 2020

Choose a reason for hiding this comment

Choose a reason for hiding this comment

shine4chen commented Feb 1, 2019 •

edited

Loading

jianjundong commented Feb 13, 2019 •

edited

Loading

jianjundong Jul 16, 2019 •

edited

Loading

jianjundong Jul 25, 2019 •

edited

Loading

lisahnguyen Jul 25, 2019 •

edited

Loading