[Buffer config template] Modified buffer config template to include cable length between internal ASICs #4929
Conversation
This file contains 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
* Add the test signing certificates for secure boot * Remove unnecessary ca key file * Regenerate the certificates to not expose the ca key
Add changes for syslog support for containers running in namespaces on multi ASIC platforms. On Multi ASIC platforms Rsyslog service is only running on the host. There is no rsyslog service running in each namespace. On multi ASIC platforms the rsyslog service on the host will be listening on the docker0 ip address instead of loopback address. The rsyslog.conf on the containers is modified to have omfwd target ip to be docker0 ipaddress instead of loopback ip Signed-off-by: Arvindsrinivasan Lakshmi Narasimhan <arlakshm@microsoft.com>
* src/sonic-platform-common 82bbeab...42781ff (1): > [SfpBase] Fix key name typo in docstring (sonic-net#99) Signed-off-by: Volodymyr Boyko <volodymyrx.boiko@intel.com>
1. Upgrade SAI headers to v1.6.3 2. Fix traffic lost during FFB related to buffer config + optimize buffer config timing for FB 3. Add ACL fields BTH, IP flags 4. Add ACL infrastructure of different fields per ASIC type
…platform. (sonic-net#4779) * Support for connecting to DB in namespace via IP:port ( using docker bridge network ) for applications in multi-asic platform. * Added the default IP as 127.0.0.1 if the IPaddress derivation from interface fails. Moved the localhost loopback IP binding logic into the supervisor.j2 file.
* sonic-sairedis submodule update * Update BRCM SAI to 3.7.5.1
…et#4843) Updated the NAT iptables patch for 4.19 buster Depends on PR : sonic-net/sonic-linux-kernel#147 1 Known issue: With both NAT patch files for 4.19 buster kernel, seeing 1 display issue in iptables like explained below On Docker NAT, iptables supported version is 1.6.0 and on base OS it’s 1.8.2. So seeing an display issue of which fullcone option is not showing in version 1.8.2 iptables output and no issues in functionality. Display issue – For example of comparsion: NAT Docker: root@sonic:/home/admin# docker exec -it nat bash root@sonic:/# iptables -t nat -nvL Chain PREROUTING (policy ACCEPT 0 packets, 0 bytes) pkts bytes target prot opt in out source destination 0 0 DNAT all -- * * 0.0.0.0/0 0.0.0.0/0 to:1.1.1.1 fullcone Chain INPUT (policy ACCEPT 0 packets, 0 bytes) pkts bytes target prot opt in out source destination Chain OUTPUT (policy ACCEPT 30 packets, 2749 bytes) pkts bytes target prot opt in out source destination Chain POSTROUTING (policy ACCEPT 30 packets, 2749 bytes) pkts bytes target prot opt in out source destination root@sonic:/# Base OS: root@sonic:/home/admin# iptables-legacy -t nat -nvL Chain PREROUTING (policy ACCEPT 0 packets, 0 bytes) pkts bytes target prot opt in out source destination 1 36 DNAT all -- * * 0.0.0.0/0 0.0.0.0/0 to:1.1.1.1 Chain INPUT (policy ACCEPT 1 packets, 36 bytes) pkts bytes target prot opt in out source destination Chain OUTPUT (policy ACCEPT 41 packets, 3572 bytes) pkts bytes target prot opt in out source destination Chain POSTROUTING (policy ACCEPT 41 packets, 3572 bytes) pkts bytes target prot opt in out source destination root@sonic:/home/admin# To fix this issue, iptables need to update from 1.6.0 to 1.8.2 version and have to update the NAT docker from stretch to buster. Will raise a new PR with this. Signed-off-by: Akhilesh Samineni akhilesh.samineni@broadcom.com Signed-off-by: Akhilesh Samineni <akhilesh.samineni@broadcom.com>
… both host and docker namespace (sonic-net#4838) * [sonic-buildimage] Changes to make network specific sysctl common for both host and docker namespace (in multi-npu). This change is triggered with issue found in multi-npu platforms where in docker namespace net.ipv6.conf.all.forwarding was 0 (should be 1) because of which RS/RA message were triggered and link-local router were learnt. Beside this there were some other sysctl.net.ipv6* params whose value in docker namespace is not same as host namespace. So to make we are always in sync in host and docker namespace created common file that list all sysctl.net.* params and used both by host and docker namespace. Any change will get applied to both namespace. Signed-off-by: Abhishek Dosi <abdosi@microsoft.com> * Address Review Comments and made sure to invoke augtool only one and do string concatenation of all set commands * Address Review Comments.
…is no one (sonic-net#4860) If some table with a list of tuples (interface name, ip prefix) has ip prefixes without a mask length, it will cause issues in SONiC. For example quagga and frr will treat ipv4 address without a mask, so "10.20.30.40" address will be treated as "10.0.0.0/8", which is dangerous. The fix here is that when pfx_filter get a tuple (interface name, ip prefix), where the ip prefix doesn't have prefix mask length, add a mask by default: "/32 for ipv4 addresses, /128 for ipv6 addresses". Co-authored-by: Pavel Shirshov <pavel.contrib@gmail.com>
* Tests of FRR templates which rendered by sonic-cfggen
…t#4825) * Loopback IP changes for multi ASIC devices multi ASIC will have 2 Loopback Interfaces - Loopback0 has globally unique IP address, which is advertised by the multi ASIC device to its peers. This way all the external devices will see this device as a single device. - Loopback4096 is assigned an IP address which has a scope is within the device. Each ASIC has a different ip address for Loopback4096. This ip address will be used as Router-Id by the bgp instance on multi ASIC devices. This PR implements this change for multi ASIC devices Signed-off-by: Arvindsrinivasan Lakshmi Narasimhan <arlakshm@microsoft.com>
…onic-net#4825)" This reverts commit cae65a4.
…c-net#4880) make swss build depends only on libsairedis instead of syncd. This allows to build swss without depending on vendor sai library. Currently, libsairedis build also buils syncd which requires vendor SAI lib. This makes difficult to build swss docker in buster while still keeping syncd docker in stretch, as swss requires libsairedis which also build syncd and requires vendor to provide SAI for buster. As swss docker does not really contain syncd binary, so it is not necessary to build syncd for swss docker. * [submodule]: update sonic-sairedis * ccbb3bc 2020-06-28 | add option to build without syncd (HEAD, origin/master, origin/HEAD) [Guohan Lu] * 4247481 2020-06-28 | install saidiscovery into syncd package [Guohan Lu] * 61b8e8e 2020-06-26 | Revert "sonic-sairedis: Add support to sonic-sairedis for gearbox phys (sonic-net#624)" (sonic-net#630) [Danny Allen] * 85e543c 2020-06-26 | add a README to tests directory to describe how to run 'make check' (sonic-net#629) [Syd Logan] * 2772f15 2020-06-26 | sonic-sairedis: Add support to sonic-sairedis for gearbox phys (sonic-net#624) [Syd Logan] Signed-off-by: Guohan Lu <lguohan@gmail.com>
- Xilinx/pericom peripherals are not actively used in DellEMC S6100 switch. - These peripherals are throwing PCIE corrected messages in some of the units and filling syslog. - Since it is not usable disabling it at startup.
* src/sonic-platform-daemons abe115e...9b8bfa1 (1): > [xcvrd] Update key names in 'get_media_settings_value()' (sonic-net#63)
also update submodule * 01f810f 2020-07-02 | fix compiling issue for gcc8.3 (sonic-net#1339) [lguohan] * 9b13120 2020-07-03 | Fix in script to avoid orchagent crash when port down followed by fdb delete (sonic-net#1340) [rupesh-k] * 9b01844 2020-07-01 | [qosorch] Update QoS scheduler params for shaping features (sonic-net#1296) [Michael Li] * 86b5e99 2020-07-02 | [mirrororch] Port Mirroring implementation (sonic-net#1314) [rupesh-k] * c05601c 2020-06-24 | [portsyncd]: add debug message if a port cannot be found in port able (sonic-net#1328) [lguohan] * a0b6412 2020-06-23 | COPP_DEL_fix: DEL for one trap group from SONIC is resetting all the trap IDs (sonic-net#1273) [SinghMinu] Signed-off-by: Guohan Lu <lguohan@gmail.com>
…onic-net#4890) Signed-off-by: Volodymyr Boyko <volodymyrx.boiko@intel.com>
thermalctld throwing error messages because it is not yet fully configured, disabling it for now on arista platforms. Co-authored-by: Zhi Yuan Carl Zhao <zyzhao@arista.com>
Resubmitting the changes for (sonic-net#4825) with fixes for sonic-bgpcdgd test failures Signed-off-by: Arvindsrinivasan Lakshmi Narasimhan <arlakshm@microsoft.com>
upgrade kernel to latest maintenance version 4.9.118 Signed-off-by: Guohan Lu <lguohan@gmail.com>
…et#4610) This pull request was cherry picked from "sonic-net#1238" to resolve the conflicts. - Why I did it Add support to specify source address for TACACS+ - How I did it Add patches for libpam-tacplus and libnss-tacplus. The patches parse the new option 'src_ip' and store the converted addrinfo. Then the addrinfo is used for TACACS+ connection. Add a attribute 'src_ip' for table "TACPLUS|global" in configDB Add some code to adapt to the attribute 'src_ip'. - How to verify it Config command for source address PR in sonic-utilities config tacacs src_ip <ip_address> - Description for the changelog Add patches to specify source address for the TACACS+ outgoing packets. - A picture of a cute animal (not mandatory but encouraged) **UT logs: ** UT_tacacs_source_intf.txt
Signed-off-by: Ying Xie <ying.xie@microsoft.com>
…-net#4913)" (sonic-net#4915) This reverts commit f427d2e.
Add related files to the device folder: buffer config templates pg lookup profile port_config.ini sai profile sensor conf plugins Co-authored-by: Stephen Sun <stephens@mellanox.com>
sonic-cfggen call is slow and this is taking place in the SONiC boot up process. The change uses templates to assemble all required vars into single template file. With this change, telemetry now calls once into sonic-cfggen. signed-off-by: Tamer Ahmed <tamer.ahmed@microsoft.com>
arlakshm
reviewed
Jul 10, 2020
| {%- set roles2 = neighbor_role + '_' + switch_role %} | ||
| {%- set roles1 = roles1 | lower %} | ||
| {%- set roles2 = roles2 | lower %} | ||
| {%- if 'asic' == switch_role | lower or 'asic' == neighbor_role | lower %} |
There was a problem hiding this comment.
The switch_role will not asic, so I think we don't need this check if 'asic' == switch_role
| @@ -57,7 +57,9 @@ def sort_by_port_index(value): | |||
| if not value: | |||
| return | |||
| if isinstance(value, list): | |||
| value.sort(key = lambda k: int(k[8:])) | |||
| value.sort( | |||
There was a problem hiding this comment.
Why do we need this change ?
There was a problem hiding this comment.
It's due to internal interfaces names, they start with "Ethernet-BPxxx". Without this change sorting fails. Adding 1024 so that internal interfaces are sorted to the end, no particular reason.
|
closing this PR as merge issues due to 'history rewrite' for master. created #4959. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
- Why I did it
To support buffer config generation for multi ASIC platform. The cable length used to select buffer profile is determined by the role of local and peer switch. In case of multi ASIC platforms cable length needs to determined between ASICs.
- How I did it
Added a new role called "internal" with a cable length of "5m". The role becomes "internal" if one of the ASICs in the pair is defined as 'internal' in minigraph.
- How to verify it
Verified config generation on multi-ASIC and single ASIC platforms. Sent traffic with different priorities, verified the queue counters. Visual confirmation of buffer configuration values.
- Description for the changelog
Modified the 'buffers_config' template to generate cable length between internal ASICs.
Modified sort API to account for internal interfaces that are named as "Ethernet-BPx".
- A picture of a cute animal (not mandatory but encouraged)