NOTE: This is cherry-pick from 1911/2012 to master.
- Why I did it
To fix LAG IP configuration race
- How I did it
Extended timeout for teammgrd
- How to verify it
Add >80 router LAGs. Do config reload
Signed-off-by: Nazarii Hnydyn <nazariig@nvidia.com>
#### Why I did it
The process of config generation (sonic-cfggen) fails, but the services continue to run with invalid config
#### How I did it
* add exit with error on errors in start.sh script (because supervisord relies on start.sh return code).
* fix jinja template. Jinja use common python expressions under the hood and `has_key` method was removed from dict in py3, so use check by `in` operator as it is supported by both py2 and py3.
#### How to verify it
* compile sonic with enabled iccp.
* add mclag config to CONFIG_DB.
```
'MC_LAG|1' => {
"local_ip": "10.0.0.2",
"peer_ip": "10.0.0.3",
"peer_link": "Ethernet8",
"mclag_interface": "Ethernet12"
}
* unmaks, enable and start swss and iccpd services in sonic.
* log in into the iccpd container and check the config file `/etc/iccpd/iccpd.conf`
* expected config:
```
mclag_id:1
local_ip:10.0.0.2
peer_ip:10.0.0.3
peer_link:Ethernet8
mclag_interface:Ethernet12
system_mac:YOUR_SYSTEM_MAC
#### Description for the changelog
Fixed initial iccpd startup configuration.
#### Why I did it
ethtool can be used to query and change settings such as speed, auto- negotiation and checksum offload on many network devices, especially Ethernet devices.
#### How I did it
add package extension to docker-platform-monitor/Dockerfile.j2
#### Why I did it
The libpci library provides portable access to configuration registers of devices connected to the PCI bus.
#### How I did it
update dockers/docker-platform-monitor/Dockerfile.j2
A recent version of contextlib2 (https://pypi.org/project/contextlib2/21.6.0/#history) has broken Python2 compatibility, so the version picked up by netaddr when using Python2 must be specified, or else builds fail
Co-authored-by: Tom Zhu <tom.zhu@metaswitch.com>
Signed-off-by: Arvindsrinivasan Lakshmi Narasimhan <arlakshm@microsoft.com>
In the multi asic platforms all the ASIC are advertising the same IPv6 /64 network from Loopback4096.
Therefore, the IPv6 loopback address of backend asic is not learnt on the frontend asic.
Change the bgpd.conf.main.conf.j2 template file to advertise the Loopback4096 ipv6 address as /128
Why I did it
Quagga is no longer being used. Remove quagga-related code (e.g., docker-fpm-quagga, sonic-quagga, etc.).
How I did it
Remove quagga-related code.
Why I did it
Quagga is no longer being used. Remove quagga-related code (e.g., docker-fpm-quagga, sonic-quagga, etc.).
How I did it
Remove quagga-related code.
Signed-off-by: Yong Zhao yozhao@microsoft.com
Why I did it
Currently we leveraged the Supervisor to monitor the running status of critical processes in each container and it is more reliable and flexible than doing the monitoring by Monit. So we removed the functionality of monitoring the critical processes by Monit.
How I did it
I removed the script process_checker and corresponding Monit configuration entries of critical processes.
How to verify it
I verified this on the device str-7260cx3-acs-1.
- Why I did it
migrate to python3 support
add dependent packages for Klish
allow login as non-root user
- How I did it
update sonic-cli script to start Klish with user name, system name and timeout
update the Dockerfile.j2 to resolve dependent packages
add python3-dev for Klish use
- How to verify it
Incremental buster build with Azure/sonic-mgmt-framework#76 and verify the sonic-cli
- Description for the changelog
Migrate to python3.7 support, update sonic-cli script and resolve package dependencies
#### Why I did it
To avoid the following error
```
Traceback (most recent call last):
File "/usr/local/bin/flush_unused_database", line 10, in <module>
if 'PONG' in output:
TypeError: a bytes-like object is required, not 'str'
```
`communicate` method returns the strings if streams were opened in text mode; otherwise, bytes.
In our case text arg in Popen is not true and that means that `communicate` return the bytes
#### How I did it
Set `text=True` to get strings instead of bytes
#### How to verify it
run `/usr/local/bin/flush_unused_database` inside database container
I added `chassis_db_init` to the startup tasks for the `docker-platform-monitor` docker so that the script is run on startup of the switch and the chassis info is correctly provisioned to STATE_DB.
Depends on https://github.com/Azure/sonic-platform-daemons/pull/183
Signed-off-by: Yong Zhao yozhao@microsoft.com
Why I did it
This PR aims to monitor the memory usage of streaming telemetry container and restart streaming telemetry container if memory usage is larger than the pre-defined threshold.
How I did it
I borrowed the system tool Monit to run a script memory_checker which will periodically check the memory usage of streaming telemetry container. If the memory usage of telemetry container is larger than the pre-defined threshold for 10 times during 20 cycles, then an alerting message will be written into syslog and at the same time Monit will run the script restart_service to restart the streaming telemetry container.
How to verify it
I verified this implementation on device str-7260cx3-acs-1.
- Why I did it
To give SONiC Application Extension developers an environment to run and develop their apps.
- How I did it
Created sonic-sdk and sonic-sdk-buildenv dockers and their dbg versions.
- How to verify it
Build:
$ make -f slave target/sonic-sdk.gz target/sonic-sdk-buildenv.gz
#### Why I did it
To avoid the following logs
```
Mar 15 15:52:04.599302 igk-dut-04 INFO database#/supervisord: flushdb /bin/bash: /usr/local/bin/flush_unused_database: /usr/bin/python: bad interpreter: No such file or directory
Mar 15 15:52:04.599947 igk-dut-04 INFO database#supervisord 2021-03-15 15:52:04,599 INFO exited: flushdb (exit status 126; not expected)
```
#### How I did it
Fix shebang
#### How to verify it
Check the logs
fuser support is required since new cisco hardware watchdog plugin uses them to check anyone else use's /dev/watchdogX resource. The actual validation happens in the platform code, but the package is required for pmon container. Currently the /dev/watchdogX is being used by cisco platform-monitor service. Cisco chassis level watchdog plugin uses "fuser" to claim the watchdog release from platform-monitor service.
#### Why I did it
MSN4700 A1/A0 used different sensor chip but keep the existing platform name *x86_64-mlnx_msn4700-r0*, this is a workaround to replace the sensor conf on MSN4700 A1/A0
#### How I did it
Use a shell script to get the sensor conf path and copy that files to /etc/sensors.d/sensors.conf
Why I did it
ndppd by default reads /proc/net/ipv6_route ever 30 seconds. Since T1s advertise so many routes to ToRs, this file is extremely large, and reading it causes ndppd's CPU usage to spike every 30 seconds
How I did it
Increase the delay for reading this file to the maximum possible value (max integer value), which will result in CPU spikes every ~24 days instead of every 30 seconds
How to verify it
Start ndppd with the new config file, confirm that no CPU spikes are seen except at startup
Signed-off-by: Lawrence Lee <lawlee@microsoft.com>
The aiohttp package is required by azure.kusto.data which is used by sonic-mgmt/test_reporting.
This change is to ensure that the dependent package is installed in the sonic-mgmt docker.
Signed-off-by: Xin Wang <xiwang5@microsoft.com>
Why I did it
Support readonly version of the command vtysh
How I did it
Check if the command starting with "show", and verify only contains single command in script.
* Why I did it
Upgrade to the latest ixnetwork-restpy and ixnetwork-open-traffic-generator pypi packages
* How I did it
Updated the pip install entries for the packages in the Dockerfile.j2
* How to verify it
pip show ixnetwork-restpy
pip show ixnetwork-open-traffic-generator
Co-authored-by: Neetha John <nejo@microsoft.com>
For the split mode, the config files, like bgpd.conf, zebra.conf and so on, were provided by outside. But the docker_init.sh will overwrite the outside config files if restart bgp service.
How I did it
Add a split mode checking in docker_init.sh, if docker_routing_config_mode is split, don't overwrite the existing routing config files.
How to verify it
Set split mode in config db
{
"DEVICE_METADATA": {
"localhost": {
"hwsku": "Force10-S6000",
"platform": "x86_64-kvm_x86_64-r0",
"docker_routing_config_mode": "split"
...
}
}
}
Replace your bgpd.conf to /etc/sonic/frr/bgpd.conf
Restart bgp service by sudo service bgp restart
The /etc/sonic/frr/bgpd.conf your provided shouldn't be overwritten
Signed-off-by: Ze Gan <ganze718@gmail.com>
- Support compile sonic arm image on arm server. If arm image compiling is executed on arm server instead of using qemu mode on x86 server, compile time can be saved significantly.
- Add kernel argument systemd.unified_cgroup_hierarchy=0 for upgrade systemd to version 247, according to #7228
- rename multiarch docker to sonic-slave-${distro}-march-${arch}
Co-authored-by: Xianghong Gu <xgu@centecnetworks.com>
Co-authored-by: Shi Lei <shil@centecnetworks.com>
This commit has following changes:
* Add templates and code to support VoQ chassis iBGP peers
* Add support to convert a new VoQChassisInternal element in the
BGPSession element of the minigraph to a new BGP_VOQ_CHASSIS_NEIGHBOR
table in CONFIG_DB.
* Add a new set of "voq_chassis" templates to docker-fpm-frr
* Add a new BGP peer manager to bgpcfgd to add neighbors from the
BGP_VOQ_CHASSIS_NEIGHBOR table using the voq_chassis templates.
* Add a test case for minigraph.py, making sure the VoQChassisInternal
element creates a BGP_VOQ_CHASSIS_NEIGHBOR entry, but not if its
value is "false".
* Add a set of test cases for the new voq_chassis templates in
sonic-bgpcfgd tests.
Note that the templates expect the new
"bgp bestpath peer-type multipath-relax" bgpd configuration to be
available.
Signed-off-by: Joanne Mikkelson <jmmikkel@arista.com>
With the latest 201911 image, the following error was seen on staging devices with TSB command ( for both single asic, multi asic ). Though this err message doesn't affect the TSB functionality, it is good to fix.
admin@STG01-0101-0102-01T1:~$ TSB
BGP0 : % Could not find route-map entry TO_TIER0_V4 20
line 1: Failure to communicate[13] to zebra, line: no route-map TO_TIER0_V4 permit 20
% Could not find route-map entry TO_TIER0_V4 30
line 2: Failure to communicate[13] to zebra, line: no route-map TO_TIER0_V4 deny 30
In addition, in this PR I am fixing the message displayed to user when there are no BGP neighbors configured on that BGP instance. In multi-asic device there could be case where there are no BGP neighbors configured on a particular ASIC.
Avoid the following error messages while dynamic buffer calculation is enabled
```
ERR monit[491]: 'swss|buffermgrd' status failed (1) -- '/usr/bin/buffermgrd -l' is not running in host
```
Change /usr/bin/buffermgrd -l to /usr/bin/buffermgrd. The buffermgrd is started by -l for traditional model or -a for dynamic model. So we need to use the common section of both.
Signed-off-by: Stephen Sun <stephens@nvidia.com>
The default bgp connect retry timer is 120 seconds. A reconnection will happen 120 seconds if the initial connection fails. This PR aims to allow a more frequent retry.
Why I did it
It was observed that on a multi-asic DUT bootup, the BGP internal sessions between ASIC's was taking more time to get ESTABLISHED than external BGP sessions. The internal sessions was coming up almost exactly 120 secs later.
In multi-asic platform the bgp dockers ( which is per ASIC ) on switch start are bring brought up around the same time and they try to make the bgp sessions with neighbors (in peer ASIC's) which may be not be completely up. This results in BGP connect fail and the retry happens after 120sec which is the default Connect Retry Timer
How I did it
Add the command to set the bgp neighboring session retry timer to 10sec for internal bgp neighbors.
