Sub PRs:
sonic-net/sonic-host-services#84
#17191
Why I did it
According to the design, the database instances of DPU will be kept in the NPU host.
Microsoft ADO (number only): 25072889
How I did it
To follow the multiple ASIC design, I assume a new platform environment variable NUM_DPU will be defined in the /usr/share/sonic/device/$PLATFORM/platform_env.conf. Based on this number, NPU host will launch a corresponding number of instances for the DPU database.
Signed-off-by: Ze Gan <ganze718@gmail.com>
- Why I did it
Mellanox MSN2700 platforms have a non-functional error log: "ERR pmon#sensord: Error getting sensor data: dps460/#10: Can't read". This error is because of a firmware issue with some PSU, we are not able to upgrade the FW online. Since there is no functional impact, this error log can be ignored safely.
- How I did it
Add a new rsyslog rule to the rsyslog-container.conf.j2, if the docker name is pmon and the platform name matches, the new rule will be inserted into the docker rsyslogd.conf
- How to verify it
run regression on the MSN2700 platform to make the error log will not be printed to the syslog.
Signed-off-by: Kebo Liu <kebol@nvidia.com>
* [buffers] Add create_only_config_db_buffers.json for MLNX devices (not MSFT SKU), inject it at the start of the swss docker
Signed-off-by: vadymhlushko-mlnx <vadymh@nvidia.com>
* [buffers] Align the sonic-device_metadata.yang
Signed-off-by: vadymhlushko-mlnx <vadymh@nvidia.com>
---------
Signed-off-by: vadymhlushko-mlnx <vadymh@nvidia.com>
Why I did it
Fix: #16699
Fast reboot is failing from old OS versions (eg., 201911 image) to latest (eg., master branch) after PR #15685
The system wide flag for FAST_REBOOT is still required when the base OS version does not support the new fast-reboot reconciliation logic (no db dump)
Add alternate name eth1-midplane to Linux bridge br1 created on supervisor on some chassis platforms.
See description here: #16504
Signed-off-by: Abhishek Dosi <abdosi@microsoft.com>
Why I did it
Add dhcp_server ipv4 feature to SONiC.
HLD: sonic-net/SONiC#1282
How I did it
To be clarify: This container is disabled by INCLUDE_DHCP_SERVER = n for now, which would cause container not build.
Add INCLUDE_DHCP_SERVER to indicate whether to build dhcp_server container
Add docker file for dhcp_server, build and install kea-dhcp4 inside container
Add template file for dhcp_server container services.
Add entry for dhcp_server to FEATURE table in config_db.
How to verify it
Build image with INCLUDE_DHCP_SERVER = y to verify:
Image can be install successfully without crush.
By config feature state dhcp_server enabled to enable dhcp_server.
- Why I did it
The recent change #15685 (comment) removed the db migration for non first reboots.
This is problematic for many deployments which doesn't rely on ZTP and push a custom config_db.json
Port to older branches after #15685 is ported back
- How I did it
Re-introduce the logic to run the db_migrator on non-first boots
- How to verify it
Verified reboot and warm-reboot cases
Signed-off-by: Vivek Reddy Karri <vkarri@nvidia.com>
* Fix CONFIG_DB_INITIALIZED flag check logic and set/reset flag for warm-reboot
* Fix db-cli usage
* Handle same image warm-reboot and generalize handling of INIT flag
* Cover boot from ONIE case: set config init flag when minigraph, config_db are missing
* Handle case: first boot of SONiC
* Check for config init flag
* Simplify logic, and do not call db_migrator for same image reboot
Why I did it
Support FIPS DB configuration
Design Doc: sonic-net/SONiC#1372
Work item tracking
Microsoft ADO (number only): 24411148
How I did it
Add the FIPS Yang model to make FIPS configurable in ConfigDB.
How to verify it
See TestPlan: sonic-net/sonic-mgmt#9092
Build the image and run the tests: sonic-net/sonic-mgmt#9091
Why I did it
To reduce the container's dependency from host system
Work item tracking
Microsoft ADO (number only):
17713469
How I did it
Move the k8s container startup script to config engine container, other than mount it from host.
How to verify it
Check file path(/usr/share/sonic/scripts/container_startup.py) inside config engine container.
Signed-off-by: Yun Li <yunli1@microsoft.com>
Co-authored-by: Qi Luo <qiluo-msft@users.noreply.github.com>
Why I did it
Sharing the storage of syncd with other proprietary application extensions allows them to communicate with syncd in differnt ways.
If one container wants to pass some information to syncd then shared storage can be used. However, today the shared storage isn't cleaned on restarts making it possible for syncd to read out-of-date information generated in the past.
NOTE: No plans to use it for standard SONIC dockers and we are working on removing the SDK dependency from PMON docker
How I did it
Implemented new service to clean the shared storage.
How to verify it
Do reboot/fast-reboot/warm-reboot/config-reload/systemctl restart swss and verify /tmp/ is cleaned after each restart in syncd container.
Signed-off-by: Stepan Blyschak <stepanb@nvidia.com>
This reverts commit 02b17839c3.
Reverts #14933
The earlier commit caused a race condition that particularly broke cross branch warm upgrade.
Issue happens when db_migrator is still migrating the DB and finalizer is checking DB for list of components to reconcile.
If migration is not complete, finalizer get an empty list to wait for. Due to this, finalizer concludes warmboot (deletes system wide warmboot flag) and cause all the services to do cold restart.
ADO: 24274591
Why I did it
Fix the issue where db_migrator is called before DB is loaded w/ config. This leads to db_migrator:
Not finding anything, and resumes to incorrectly migrate every missing config
This is not expected. migration should happen after the old config is loaded and only new schema changes need migration.
Since DB does not have anything when migrator is called, db_migrator fails when some APIs return None.
The reason for incorrect call is that:
database service starts db_migrator as part of startup sequence.
config-setup service loads data from old-config/minigraph. However, since it has Requires=database.service.
Hence, config-setup starts only when database service is started. And database service is started when db_migrator is completed.
Fixed by:
Check if this is first time boot by checking pending_config_migration flag.
If pending_config_migration is enabled, then do not call db_migrator as part of database service startup.
Let database service start which triggers config-setup service to start.
Now call db_migrator after when config-setup service loads old-config/minigraph
- Why I did it
We suspect the issue #13791 is caused by redis server being temporarily unavailable during system initialization so we do not use -d in sonic-cfggen, for now, to avoid accessing redis server
- How I did it
Provide a string containing required json data when calling sonic-cfggen
- How to verify it
Manually test it
Signed-off-by: Stephen Sun <stephens@nvidia.com>
#### Why I did it
Enhance the error message output mechanism during swss docker creating
#### How I did it
Capture the output to stderr of `sonic-cfggen` and output it using `echo` to make sure the error message will be logged in syslog.
#### How to verify it
Manually test
Fixe #12047. After the c++ implementation of the sonic-db-cli, sonic-db-cli PING command tries to initialize the global database for all instances database starting. If all instance database-config.json are not ready yet. it will crash and generate core file. PR sonic-net/sonic-swss-common#701 only fix the crash and the process abortion.
Signed-off-by: mlok <marty.lok@nokia.com>
- Why I did it
In to-sonic and multi-asic KVM-test, pretest sometimes failed. Reason is rsyslogd process can not start in teamd container. Because rsyslog.conf is empty caused by sonic-cfggen execute failed
- How I did it
If sonic-cfggen -d execute failed, execute without -d because the template file has the default value.
- How to verify it
Build image and test it over 40 times, all passed pretest.
Signed-off-by: Chun'ang Li <chunangli@microsoft.com>
- Why I did it
Added platform specific script to be invoked during SAI failure dump. Added some generic changes to mount /var/log/sai_failure_dump as read write in the syncd docker
- How I did it
Added script in docker-syncd of mellanox and copied it to /usr/bin
- How to verify it
Manual UT and new sonic-mgmt tests
After upgrade to brcmsai 8.1, the sdk running environment (container) recommended with mininum memory size as below
TH4/TD4(ltsw) uses 512MB
TH3 used 300MB
Helix4/TD2/TD3/TH/TH 256 MB
Base on this requirement, adjust the default syncd share memory size and set the memory size for special ACISs in platform_env.conf file for different types of Broadcom ASICs.
How I did it
Add the platform_env.conf file if none of it for broadcom platform (base on platform_asic file)
Add the 'SYNCD_SHM_SIZE' and set the value
for ltsw(TD4/TH4) devices set to 512M at least (update the platform_env.conf)
for Td2/TH2/TH devices set to 256M
for TH3 set to 300M
verify
How to verify it
verify the image with code fix
Check with UT
Check on lab devices
On a problematic device which cannot start successfully
Run with the command
$ cat /proc/linux-kernel-bde
Broadcom Device Enumerator (linux-kernel-bde)
Module parameters:
maxpayload=128
usemsi=0
dmasize=32M
himem=(null)
himemaddr=(null)
DMA Memory (kernel): 33554432 bytes, 0 used, 33554432 free, local mmap
No devices found
$ docker rm -f syncd
syncd
$ sudo /usr/bin/syncd.sh start
Cannot get Broadcom Chip Id. Skip set SYNCD_SHM_SIZE.
Creating new syncd container with HWSKU Force10-S6000
a4862129a7fea04f00ed71a88715eac65a41cdae51c3158f9cdd7de3ccc3dd31
$ docker inspect syncd | grep -i shm
"ShmSize": 67108864,
"Tag": "fix_8.1_shm_issue.67873427-9f7ca60a0e",
On Normal device
$ docker inspect syncd | grep -i shm
"ShmSize": 268435456,
"Tag": "fix_8.1_shm_issue.67873427-9f7ca60a0e"
change the config syncd_shm.ini to b85=128m
$ docker rm -f syncd
syncd
$ sudo /usr/bin/syncd.sh start
Creating new syncd container with HWSKU Force10-S6000
3209ffc1e5a7224b99640eb9a286c4c7aa66a2e6a322be32fb7fe2113bb9524c
$ docker inspect syncd | grep -i shm
"ShmSize": 134217728,
"Tag": "fix_8.1_shm_issue.67873427-9f7ca60a0e",
change the config under
/usr/share/sonic/device/x86_64-dell_s6000_s1220-r0/Force10-S6000/platform_env.conf
and run command
$ cat /usr/share/sonic/device/x86_64-dell_s6000_s1220-r0/platform_env.conf
SYNCD_SHM_SIZE=300m
$ sudo /usr/bin/syncd.sh start
Creating new syncd container with HWSKU Force10-S6000
897f6fcde1f669ad2caab7da4326079abd7e811bf73f018c6dacc24cf24bfda5
$ docker inspect syncd | grep -i shm
"ShmSize": 314572800,
"Tag": "fix_8.1_shm_issue.67873427-9f7ca60a0e",
Signed-off-by: richardyu-ms <richard.yu@microsoft.com>
Multi-asic Docker instances are created behind Docker's default bridge
which doesn't allow talking to other Docker instances that are in the
host network (like database-chassis).
On linecards, we configure midplane interfaces to let per-asic docker
containers talk to CHASSIS_DB on the supervisor through internal chassis
network.
On the supervisor we don't need to use chassis internal network, but we
still need a similar setup in order to allow fabric containers to talk
to database-chassis
With this PR in, you flap BGP and use events_tool to see the published events.
With telemetry PR #111 in and corresponding submodule update done in buildimage, one could run gnmi_cli to capture BGP flap events.
*Preventing ebtables rules to be applied on KVM image. The ebtables rules in SONiC are added to prevent ARP as well as L2 forwarding to be blocked in linux kernel since the hardware will take care of the actual L2 forward. However this is not the case with KVM where linux needs to forward even L2 packets
What/Why I did:
Issue1: By setting up of ipvlan interface in interface-config.sh we are not tolerant to failures. Reason being interface-config.service is one-shot and do not have restart capability.
Scenario: For example if let's say database service goes in fail state then interface-services also gets failed because of dependency check but later database service gets restart but interface service will remain in stuck state and the ipvlan interface nevers get created.
Solution: Moved all the logic in database service from interface-config service which looks more align logically also since the namespace is created here and all the network setting (sysctl) are happening here.With this if database starts we recreate the interface.
Issue 2: Use of IPVLAN vs MACVLAN
Currently we are using ipvlan mode. However above failure scenario is not handle correctly by ipvlan mode. Once the ipvlan interface is created and ip address assign to it and if we restart interface-config or database (new PR) service Linux Kernel gives error "Error: Address already assigned to an ipvlan device." based on this:https://github.com/torvalds/linux/blob/master/drivers/net/ipvlan/ipvlan_main.c#L978Reason being if we do not do cleanup of ip address assignment (need to be unique for IPVLAN) it remains in Kernel Database and never goes to free pool even though namespace is deleted.
Solution: Considering this hard dependency of unique ip macvlan mode is better for us and since everything is managed by Linux Kernel and no dependency for on user configured IP address.
Issue3: Namespace database Service do not check reachability to Supervisor Redis Chassis Server.
Currently there is no explicit check as we never do Redis PING from namespace to Supervisor Redis Chassis Server. With this check it's possible we will start database and all other docker even though there is no connectivity and will hit the error/failure late in cycle
Solution: Added explicit PING from namespace that will check this reachability.
Issue 4:flushdb give exception when trying to accces Chassis Server DB over Unix Sokcet.
Solution: Handle gracefully via try..except and log the message.
Why I did it
To support address sanitizer for Mellanox syncd
How I did it
/var/log/asan is mapped for syncd container (the same as for swss)
container stop() has a timeout (60s) for syncd (the same as for swss)
This is so libasan has enough time to generate a report.
added ASAN's log path to Mellanox syncd supervisord.conf
added "asan: yes" to sonic_version.yml
How to verify it
Added artificial memory leaks
Compiled with ENABLE_ASAN=y
Installed the image on DUT
Rebooted the DUT
Verified that /var/log/asan/syncd-asan.log contains the leaks
Signed-off-by: Yakiv Huryk <yhuryk@nvidia.com>
Why I did it
In the bringup of tomahawk4/trident4, we realized that such chips need a larger size of /dev/shm in syncd container, so we added the option --shm-size to the docker create for syncd. The default value for shm-size is 64m; after this change, people can add SYNCD_SHM_SIZE=128m to platform_env.conf to change it to 128m.
How to verify it
We verified that after this change, 1) on existing platforms without platform_env.conf, the size of /dev/shm in syncd container (df -h | grep shm) is still the default 64M; 2) after we add SYNCD_SHM_SIZE=128m to platform_env.conf, /dev/shm in syncd becomes 128M.
Implement infrastructure that allows enabling address sanitizer
for docker containers. Enable address sanitizer for SWSS container.
- Why I did it
To add a possibility to compile SONiC applications with address sanitizer (ASAN).
ASAN is a memory error detector for C/C++. It finds:
1. Use after free (dangling pointer dereference)
2. Heap buffer overflow
3. Stack buffer overflow
4. Global buffer overflow
5. Use after return
6. Use after the scope
7. Initialization order bugs
8. Memory leaks
- How I did it
By adding new ENABLE_ASAN configuration option.
- How to verify it
By default ASAN is disabled and the SONiC image is not affected.
When ASAN is enabled it inspects all allocation, deallocation, and memory usage that the application does in run time. To verify whether the application has memory errors tests that trigger memory usage of the application should be run. Ideally, the whole regression tests should be run. Memory leaks reports will be placed in /var/log/asan/ directory of SONiC host OS.
Signed-off-by: Oleksandr Ivantsiv <oivantsiv@nvidia.com>
Why I did it
database.sh failed to create the database for namespace in multiasic platform.
The latest code Docker version 20.10.x, command "docker create" no longer takes optional "NET=" with empty value. Syntax error show with current docker create command in database.sh. Issue #9503
How I did it
Modify the docker_image_ctl.j2 to set default network setting NET="bridge" instead of empty for namespace database.
Why I did it
Fix for issue #8389
How I did it
The /etc/rsyslog.conf is empty file which cause the FATAL of the process rsyslogd in the global instance database container. The function updateSyslogConf() should only generate the rsyslog.conf for containers in the namespace. it should not do it for the containers in the global instance. Instead, default rsyslog.conf should be used. Especially for database container, updateSyslogConf() is called before the database container is created. The result cause the sonic-cfggen failed to generate the rsyslog.conf.Why I did it
Fix for issue #8389
How I did it
The /etc/rsyslog.conf is empty file which cause the FATAL of the process rsyslogd in the global instance database container. The function updateSyslogConf() should only generate the rsyslog.conf for containers in the namespace. it should not do it for the containers in the global instance. Instead, default rsyslog.conf should be used. Especially for database container, updateSyslogConf() is called before the database container is created. The result cause the sonic-cfggen failed to generate the rsyslog.conf.
Signed-off-by: mlok <marty.lok@nokia.com>
- Why I did it
This is to update the common sonic-buildimage infra for reclaiming buffer.
- How I did it
Render zero_profiles.j2 to zero_profiles.json for vendors that support reclaiming buffer
The zero profiles will be referenced in PR [Reclaim buffer] Reclaim unused buffers by applying zero buffer profiles #8768 on Mellanox platforms and there will be test cases to verify the behavior there.
Rendering is done here for passing azure pipeline.
Load zero_profiles.json when the dynamic buffer manager starts
Generate inactive port list to reclaim buffer
Signed-off-by: Stephen Sun <stephens@nvidia.com>
The common startup script used for SONiC containers is calling an inline
python command that uses Python 2 syntax, and thus errors out when run
with Python 3. Make this work with Python 3.
Signed-off-by: Saikrishna Arcot <sarcot@microsoft.com>
As a part of warmboot, redis database is dumped:
c97fe546e5/scripts/fast-reboot (L269)
However, this dump file is deleted, after it is loaded back into db post reboot.
The DB dump can be useful for debugging purpose, hence taking a backup of it can be useful.
Instead of deleting the dump, rename and keep the dump.
#### Why I did it
Use a predefined variable to get vendor information when the swss docker container is created
#### How I did it
Use `{{ sonic_asic_platform }}` instead of `$SONIC_CFGGEN -y /etc/sonic/sonic_version.yml -v asic_type`
#### How to verify it
Manually test.
This PR creates a directory firmware on the HOST with the path /usr/share/sonic/firmware, as well as this is
mounted on PMON container with the same path /usr/share/sonic/firmware. This is required for firmware
upgrade support for muxcable as currently by design all Y-Cable API's are called by xcvrd. As such if CLI has
to transfer a file to PMON we need to mount a directory from host to PMON just for getting the firmware files.
Hence we require this change.
Signed-off-by: vaibhav-dahiya <vdahiya@microsoft.com>
The voq system lag id boundary is set in redis-chassis. Changes include
setting this from database-chassis container. This fixes a timing issue
in finding datbase_config.json file from redis directory which is
created from database container. Since database container usually
starts after database-chassis container the existence of this file is
unreliable while running the command. Running the command under
database-chassis container makes sure that the database_config.json form
redis-chassis directory is guaranteed to be available and hence fixes the
timing issue.
Signed-off-by: vedganes <vedavinayagam.ganesan@nokia.com>
#### Why I did it
Following the discussion in another PR https://github.com/Azure/sonic-buildimage/pull/7708#discussion_r642933510 , since there will be multi subfolders under **/var/log/mellanox**, so we agreed to only mount this folder and the subfolders will be created afterward on demand.
#### How I did it
during the syncd docker creation, only mount folder **/var/log/mellanox**
#### How to verify it
build an Mellanox image and verify the related folder on the host and docker side.
Signed-off-by: vedganes <vedavinayagam.ganesan@nokia.com>
Changes for setting platfrom specific lag id boundary id in the chassis
app db. The platfrom specific lag id boundaries are supplied via
chassisdb.conf. The lag_id_start and lag_id_end boundary values sourced
from this file are set in chassis app db which will be used by lag id
allocator to allocate unique lag id in atomic fashion
- Why I did it
I made the docker_img_ctl.j2 applicable for more dockers (including application extensions dockers) by adding an option not to mount tmpfs on /tmp/ and /var/tmp/. In some applications /tmp/ is a different docker volume which can't be tmpfs.
Also, I added and ability to pass REPO[:TAG]|[@digest]/IMAGE_ID instead of just REPO name.
- How I did it
Modified docker_img_ctl.j2 and docker makefiles.
- How to verify it
Run it on the switch.
Following changes were done for ebtables:
- Support for Multi-asic platforms. Ebtable filters are installed in namespace for multi-asic and not host. On Single asic installed on host.
- For Multi-asic platforms we don't want to install on host otherwise Namespace-to-Namespace communication does not happens since ARP Request are not forwarded.
- Updated to use text file to restore ebtables rules then the binary format. Rules are restore as part of Database docker init instead of rc.local
- Removed the ebtable service files for buster as not needed as filters are restored/installed as part of database docker init.
All the binaries are pre-installed with ebtables* binary are same as ebatbles-legacy-*
Signed-off-by: Abhishek Dosi <abdosi@microsoft.com>
The Portchannels were not getting cleaned up as the cleanup activity was taking more than 10 secs which is default docker timeout after which a SIGKILL will be send.
Fixes#6199
To check if it works out for this issue in 201911 ? #6503
This issue is significantly seen in master branch compared to 201911 because the Portchannel cleanup takes more time in master. Test on a DUT with 8 Port Channels.
master
admin@str-s6000-acs-8:~$ time sudo systemctl stop teamd
real 0m15.599s
user 0m0.061s
sys 0m0.038s
Sonic 201911.v58
admin@str-s6000-acs-8:~$ time sudo systemctl stop teamd
real 0m5.541s
user 0m0.020s
sys 0m0.028s
* First cut image update for kubernetes support.
With this,
1) dockers dhcp_relay, lldp, pmon, radv, snmp, telemetry are enabled
for kube management
init_cfg.json configure set_owner as kube for these
2) Each docker's start.sh updated to call container_startup.py to register going up
As part of this call, it registers the current owner as local/kube and its version
The images are built with its version ingrained into image during build
3) Update all docker's bash script to call 'container start/stop/wait' instead of 'docker start/stop/wait'.
For all locally managed containers, it calls docker commands, hence no change for locally managed.
4) Introduced a new ctrmgrd service, that helps with transition between owners as kube & local and carry over any labels update from STATE-DB to API server
5) hostcfgd updated to handle owner change
6) Reboot scripts are updatd to tag kube running images as local, so upon reboot they run the same image.
7) Added kube_commands.py to handle all updates with Kubernetes API serrver -- dedicated for k8s interaction only.
