Why I did it
SONiC service determine-reboot-cause might run before driver creating reset cause files. In that case, the reset cause will be "Unknown". This PR introduces a wait mechanism to wait for reset cause sysfs files ready.
How I did it
/run/hw-management/config/reset_attr_ready is the file to indicate all reset cause files are ready. In chassis.get_reboot_cause function, it waits /run/hw-management/config/reset_attr_ready for up to 45 seconds.
How to verify it
Manual test on master/202211/202205
Microsoft ADO (25266920)
sonic-mgmt xoff test was failing for [100g,120km]. Needed to update total headroom pool size when 100G line card is used as T2 uplink.
This size was calculated assuming 100g is used for downlink so cable length was 2km whereas it can also be used for uplink (cable length - 120km). so we need to do calculation based on 120km not 2km. Although it will be some wastage for 2km scenario but it should cover both cases.
This fixesNokia-ION/ndk#22
Note that this PR must be coupled with NDK version >= 22.9.13
Why I did it
To provide proper support for CMIS compliant transceiver module CDB operations (including FW related operations).
How I did it
Enhanced the transport subsystem so as to provide for up to 2k bytes of data to be passed to/from modules (as contrasted with the prior max of 128 bytes).
How to verify it
Ensure that new FW (firmware) can be programmed to CMIS compliant module(s) using the 'sfputil firmware ...' commands.
What I did:
Fixes: #16468
Why I did:
On Some chassis there is no dedicated eth1-midplane interface on supervisor for supervisor and LC communication but instead Linux bridge br1 is used for that. Because of this changes that were done to white-list traffic over eth1-midplane would not work.
How I did:
To fix this we are using altname property of ip link command to set eth1-midplane as altname of br interface. This is done to keep design generic across chassis and between supervisor and LC also. IP-table rules are updated to get parent/base interface name of eth1-midplane.
Signed-off-by: Abhishek Dosi <abdosi@microsoft.com>
Why I did it
Release Notes for 8102-64H
• Fix NHG drop when route is added before members are added (MIGSMSFT-333 / SR 696141124)
• Added a new system device property "acl_set_dscp_encap_outer_only"
• IN_DISCARD counters report back per-port counters only instead of all counters that are per-port and also that are shared.
How I did it
Update platform version to 202205.2.2.12
#### Why I did it
To enable qos config for a certain backend deployment mode, for resource-type "Compute-AI".
This deployment has the following requirement:
- Config below enabled if DEVICE_TYPE as one of backend_device_types
- Config below enabled if ResourceType is 'Compute-AI'
- 2 lossless TCs' (2, 3)
- 2 lossy TCs' (0,1)
- DSCP to TC map uses 4 DSCP code points and maps to the TCs' as follows:
"DSCP_TO_TC_MAP": {
"AZURE": {
"48" : "0",
"46" : "1",
"3" : "3",
"4" : "4"
}
}
- WRED profile has green {min/max/mark%} as {2M/10M/5%}
This required template change <as in the PR> in addition to the vendor qos.json.j2 file (not included here).
### How I did it
#### How to verify it
- with the above change and the vendor config change, generated the qos.json file and verified that the objective stated in "Why I did it" was met
- verified no error
### Description for the changelog
Update qos_config.j2 for Comptue-AI deployment on one of backend device type roles
Why I did it
sonic-mgmt test failure is seen for update_firmware component API
Microsoft ADO: 25208748
How I did it
Edited API 2.0 to fix this issue.
How to verify it
Run sonic-mgmt test after the fix and verify it passes.
* [swss] Chassis db clean up optimization and bug fixes
This commit includes the following changes:
- Fix for regression failure due to error in finding CHASSIS_APP_DB in
pizzabox (#PR 16451)
- After attempting to delete the system neighbor entries from
chassis db, before starting clearing the system interface entries,
wait for sometime only if some system neighbors were deleted.
If there are no system neighbors entries deleted for the asic coming up,
no need to wait.
- Similar changes for system lag delete. Before deleting the
system lag, wait for some time only if some system lag memebers were
deleted. If there are no system lag members deleted no need to wait.
- Flush the SYSTEM_NEIGH_TABLE from the local STATE_DB. While asic
is coming up, when system neigh entries are deleted from chassis ap
db (as part of chassis db clean up), there is no orchs/process running to
process the delete messages from chassis redis. Because of this, stale system
neigh are entries present in the local STATE_DB. The stale entries result in
creation of orphan (no corresponding data path/asic db entry) kernel neigh
entries during STATE_DB:SYSTEM_NEIGH_TABLE entries processing by nbrmgr (after
the swss serive came up). This is avoided by flushing the SYSTEM_NEIGH_TABLE from
the local STATE_DB when sevice comes up.
Signed-off-by: vedganes <veda.ganesan@nokia.com>
* [swss] Chassis db clean up bug fixes review comment fix - 1
Debug logs added for deletion of other tables (SYSTEM_INTERFACE and SYSTEM_LAG_TABLE)
Signed-off-by: vedganes <veda.ganesan@nokia.com>
---------
Signed-off-by: vedganes <veda.ganesan@nokia.com>
(cherry picked from commit b13b41fc22)
Signed-off-by: anamehra anamehra@cisco.com
Added a check for DEVICE_METADATA before accessing the data. This prevents the j2 failure when var is not available.
Why I did it
Zebra core sometimes seen during config reload. Series of route-map deletions and then re-adds, and this triggers the hash table to realloc to grow to a larger size, then subsuquent route-map operations will be against a corrupted hash table.
Issue is seen when we have BFD Enable on Static Route table we see Static route-map being created/deleted based on bfd session state. However issue itself is very generic from FRR perspective.
Thie issue has detailed core info sonic-net/sonic-frr#37 . This PR fixes this issue.
Fixes#sonic-net/sonic-frr#37
Work item tracking
Microsoft ADO (17952227):
How I did it
This fix is already in Master frr/8.2.5. Porting this fix to 202205 branch to address this Zebra core.
sonic-net/sonic-frr@5f503e5
Solution:
The whole purpose of the delay of deletion and the storage of the route-map is to allow the using protocol the ability to process the route-map at a later time while still retaining the route-map name( for more efficient reprocessing ). The problem exists because we are keeping multiple copies of deletion events that are indistinguishable from each other causing hash havoc.
How to verify it
Verified running sonic-mgmt test, doing multiple config reloads.
chassis-packet: Update arp_update script for FAILED and STALE check (#16311)
1. Fixing an issue with FAILED entry resolution retry.
Neighbor entries in arp table may sometimes enter a FAILED state when the far end is down and reports the state as follows:
2603:10e2:400:3::1 dev PortChannel19 router FAILED
While the arp_update script handles the entries for FAILED in the following format, the above was not handled due to the token location (extra router keyword at index 4):
2603:10e2:400:3::1 dev PortChannel19 FAILED
The former format may appear if an arp resolution is tried on a link that is known but the far end goes down, e.g., pinging a STALE entry while the far end is down.
2. Refreshing STALE entries to make sure the far end is reachable.
STALE entries for some backend ports may appear in chassis-packet when no traffic is received for a while on the port. When the far end goes down, it is expected for BFD to stop sending packets on the session for which the far end is not reachable. But as the entry is known as stale, on the Cisco chassis, BFD keeps sending packets. Refreshing the stale entry will keep active links as reachable in the neighbor table while the entries for the far end down will enter a failed state. FAILED state entries will be retired and entered reachable when far end comes back up.
* [minigraph] remove number of lanes check for changing speed from 400G to 100G and set speed setting before lane reconfiguration (#15721)
8111 800G interface, split to 2x400G (each has 4 lanes) fails to change interface speed from 400G to 100G during deploy mg. In minigraph.xml, the interface speed configuration is good, but fails to generate the right value to config_db.json.
In order to support this SKU the speed transitioning should support both 4 lanes and 8 lanes in the port_config.ini.
Why I did it
before this change for a 400G to 100G transition, in all cases except when lanes are 8, we would continue and the line
ports.setdefault(port_name, {})['speed'] = port_speed_png[port_name]
would not be executed, hence the default speed will never be set for a case and config_db will not be updated,
where speed is transitioning from 400G to 100G or 40G, but lanes are not equal to 8.
In order for those cases to pass where lanes are not specifically 8, we need the change
Work item tracking
24242657
Signed-off-by: vaibhav-dahiya <vdahiya@microsoft.com>
* fix UT
Signed-off-by: vaibhav-dahiya <vdahiya@microsoft.com>
---------
Signed-off-by: vaibhav-dahiya <vdahiya@microsoft.com>
Bmc is a valid neighbor type in minigraph, however it was missing from the YANG model definition. Usually, the Bmc type device can be neighbor of BmcMgmtToRRouter. This PR is to introduce this type.
Co-authored-by: Yaqiang Zhu <zyq1512099831@gmail.com>
How I did it
Update Yang definition of ACL_TABLE_TYPE.
Update existing testcase.
Add new testcase to cover lowercase key scenario.
How to verify it
Verified by building sonic_yang_models-1.0-py3-none-any.whl. While building the target package, unit tests were run and passed.
Co-authored-by: Zhijian Li <zhijianli@microsoft.com>
Why I did it
According to ACL-Table-Type-HLD, the value type of MATCHES, ACTIONS and BIND_POINTS should be list instead of string. Opening this PR to update the definition of BMCDATA and BMCDATAV6.
How I did it
Update the definition of BMCDATA and BMCDATAV6 in minigraph-parser.
How to verify it
Verified by UT and build SONiC image.
Co-authored-by: Zhijian Li <zhijianli@microsoft.com>
- Why I did it
Revise lable name and fix typo in sensor.conf of 4600C
- How I did it
Revise lable name and fix typo in sensor.conf of 4600C
- How to verify it
Manual test
sonic-mgmt test_sensors.py
Co-authored-by: Junchao-Mellanox <57339448+Junchao-Mellanox@users.noreply.github.com>
On S6100 we are seeing almost 100K interrupts per second on intels i801 SMBUS controller which affects systems performance.
We now disable the i801 driver interrupt and instead enable polling
Microsoft ADO (number only): 24910530
How I did it
Disable the interrupt by passing the interrupt disable feature argument to i2c-i801 driver
How to verify it
This fix is NOT applicable for ARM based platforms. Applicable only for intel based platforms:-
- On SN2700 its already disabled in Mellanox hw-mgmt
- Celestica DX010 and E1031
- Dell S6100 verified the interrupts are no longer incrementing.
- Arista 7260CX3
Signed-off-by: Prince George <prgeor@microsoft.com>
Co-authored-by: Prince George <45705344+prgeor@users.noreply.github.com>
- Why I did it
Commit sonic-net/sonic-platform-daemons@153ea47 changed SfpStateUpdateTask from Process to Thread. In this commit, it raises an exception in SfpStateUpdateTask to make shutdown flow fast. But it does not work on Nvidia platform as Nvidia platform is passing timeout parameter of get_change_event to select. Linux select function can not be interrupted by a Python exception. There is no such issue on Nvidia platform before that commit. However, in order to comply with the commit and make shutdown flow fast, we decided to change Nvidia platform API implementation.
To fix issue #13591.
- How I did it
The select call in get_change_event should use no more than 1 second as timeout parameter.
Outside the select call, add a while loop to make sure timeout parameter of get_change_event work as expected
- How to verify it
Manual test
Co-authored-by: Junchao-Mellanox <57339448+Junchao-Mellanox@users.noreply.github.com>
* Fix the Loopback0 IPv6 address of LC's in chassis not reachable from peer device's
* Assign the metric vaule for Ipv6 default route learnt via RA message to higher value so that BGP learnt default route is higher priority.
Signed-off-by: Abhishek Dosi <abdosi@microsoft.com>
Co-authored-by: abdosi <58047199+abdosi@users.noreply.github.com>
How I did it
Fix the regex for L4 port range in openconfig_acl.py.
How to verify it
Build image and install on Arista-720DT DUT, then try the repro steps in #16189 and confirmed the ACL rule be setup correctly:
Co-authored-by: Zhijian Li <zhijianli@microsoft.com>
Why I did it
Common Release Notes for 8102-64H, T0/DualTor, and 8101-32FH
Fix for an issue where drop counters were incrementing twice for packets with invalid tag
Fix for the ECC errors reported in SR 695600099
Fix for fwutil show updates failure
How I did it
Update platform version to 202205.2.2.11
There is a redundant line in init_cfg.json.j2. It would cause pmon service always has "has_timer=False". However, we know that PMON has a timer now. So, I try to fix it here.
* [chassis]Chassis DB cleanup when asic comes up
Cleanup the entries from the following tables in chassis app db in
redis_chassis server in the supervisor
(1) SYSTEM_NEIGH
(2) SYSTEM_INTERFACE
(3) SYSTEM_LAG_MEMBER_TABLE
(4) SYSTEM_LAG_TABLE
As part of the clean up only those entries created by the asic that
is coming up are deleted. The LAG IDs used by the asics are also
de-allocated from SYSTEM_LAG_ID_TABLE and SYSTEM_LAG_ID_SET
- Added check to run the chassis db clean up only for voq switches.
Signed-off-by: vedganes <veda.ganesan@nokia.com>
Co-authored-by: vganesan-nokia <67648637+vganesan-nokia@users.noreply.github.com>
Why I did it
Update the platform_reboot of Nokia Platform IXR-7250E-36x400G to displays the correct reboot-cause history when reboot from supervisor card.
Work item tracking
Microsoft ADO (number only):
How I did it
Modify the platform_reboot script to copy the correct reboo-cause.txt file from NDK to the /host/reboot-cause directory at the down cycle when the reboot is issued from Supervisor (for both reboot right after install a new image and normal reboot)
Signed-off-by: mlok <marty.lok@nokia.com>
Co-authored-by: Marty Y. Lok <76118573+mlok-nokia@users.noreply.github.com>
This is a fix for PR #6051
The original PR will disable intel idle driver but it cannot limit the max c-state to 1 due to system will fall back to acpi idle driver.
Currently intel_idle.max_cstate=0 is already present, which will disable intel idle driver. With the added option, common idle driver will be disabled as well, so there will not be idle management. This is to prevent a bug that can be triggered by idle instruction on intel platform.
Work item tracking
Microsoft ADO (number only): 24867921
How I did it
Add the option to installer file beside intel_idle.max_cstate=0
Signed-off-by: Xichen Lin <lukelin0907@gmail.com>
