* Include platform info in name.
Get SONiC Version as parameter and use
Make additional tag as optional.
Avoid repetitions by using function.
* Per review comments, make SONIC_VERSION optional and added some comments.
* 1) Added additional params are optional
2) Handle DOCKER_IMAGE_TAG only if given
3) Use BUILD_NUMBER only if SONIC_VERSION not given
4) Tag with SONIC_VERSION if given.
Current behavior is not changed, unless SONIC_VERSION is given.
* Update per review comments
1) Added new args with options
2) Handle PORT possible being empty
3) Exhibit new behavior only if both version & platform are given.
* Drop redundant quotes
- This commit fixes the issue #4021 (Sensord not able to start up on
both master and 201911 branches)
- Add the new patch file for "lm-sensors" to perform dh_installinit to
include sensord.init in the packed deb. So that the missing init
script file for 'sensord' will be included in the created 'sensord'
deb.
Signed-off-by: Charlie Chen <charlie_chen@edge-core.com>
sai_xml contains info about port splits, previously it simply linked to the MSN3800 sai xml, which does not have splits. New version describes splits and speeds according to Mellanox-SN3800-D112-C8 SKU.
Practically it can cause port recreation on SAI init.
Signed-off-by: Mykola Faryma <mykolaf@mellanox.com>
* Fix the CMD for the PROCESSSTATS entries so that
there is a space between the command name and the
arguments.
Signed-off-by: Garrick He <garrick_he@dell.com>
fix#4315
download and install cmake 3.13.2 from azure storage
latest cmake 3.16.3 from stretch-backports breaks the libyang 1.0.73 build
Signed-off-by: Guohan Lu <lguohan@gmail.com>
[fwutil]: Use overlay driver when mounting next image filesystem (#825)
Fix for adding L3 interface to Vlan group (#826)Fix for adding L3 interface to Vlan group (#826)
[db_migrator]Do DB migration for buffer pool size change on Mellanox platform (#833)
explicitly specify command with underscores (#846)
[intfutil] set speed to 0 when interface speed is not available (#839)
maintains backward compatibility to search original x509 location
when telemetry table does not have certs
Signed-off-by: Guohan Lu <lguohan@gmail.com>
* add MSN4700 device files
* update ACS-MSN4700 sai profile
* update buffer pool size, headroom, sensor conf, port config and reboot scripts
* fix ident
* update sensor conf and buffer pool
* [sn4700] add sku 4700 to chassis.py
* [Mellanox-4700] Add 4700 info to psu and thermal platform API
* update buffer config file template to the latest.
update SAI profile to use 100G X 4lanes for now
update port_config.ini according to the SAI profile
* [Mellanox]Update the buffer configurations for 4700
* fix alignment in pg_profile_lookup.ini
* add platform components file for new sku
* Update device/mellanox/x86_64-mlnx_msn4700-r0/ACS-MSN4700/pg_profile_lookup.ini
Co-Authored-By: Nazarii Hnydyn <nazariig@mellanox.com>
* remove redundant line
* [Mellanox]Correct type, buffer size
Co-authored-by: Nazarii Hnydyn <nazariig@mellanox.com>
Co-authored-by: junchao <junchao@mellanox.com>
Co-authored-by: Stephen Sun <stephens@mellanox.com>
Instead of updating hostname manualy on Config DB hostname change,
simply share containers UTS namespace with host OS.
Ideally, instead of setting `--uts=host` for every container in SONiC,
this setting can be set per container if feature requires.
One behaviour change is introduced in this commit, when `--privileged`
or `--cap-add=CAP_SYS_ADMIN` and `--uts=host` are combined, container
has privilege to change host OS and every other container hostname.
Such privilege should be fixed by limiting containers capabilities.
Signed-off-by: Stepan Blyschak <stepanb@mellanox.com>
- What I did
Add configuration to avoid ntpd from panic and exit if the drift between new time and current system time is large.
- How I did it
Added "tinker panic 0" in ntp.conf file.
- How to verify it
[this assumes that there is a valid NTP server IP in config_db/ntp.conf]
Change the current system time to a bad time with a large drift from time in ntp server; drift should be greater than 1000s.
Reboot the device.
Before the fix:
3. upon reboot, ntp-config service comes up fine, ntp service goes to active(exited) state without any error message. This is because the offset between new time (from ntp server) and the current system time is very large, ntpd goes to panic mode and exits. The system continues to show the bad time.
After the fix:
3. Upon reboot, ntp-config comes up fine, ntp services comes up from and stays in active (running) state. The system clock gets synced with the ntp server time.
We believe that the supervisord issue in face of clock rolling backwards
has been addressed. Therefore reverting change 2598 to allow ntp sync
to right clock at the start up time.
Signed-off-by: Ying Xie <ying.xie@microsoft.com>
Brief:
The community code for has been updated, which mismatched with the existing sonic-mgmt-framework. Hence it broke the build.
Submodule-PR: #43
Details:
The SONiC mgmt Framework, pulls down the community modules in two steps.
Pull the existing; This does a transparent compile too.
Then checkout the specific commit-id for each and do the final compile.
The community code change broke in the compilation phase of step-1.
Fix: For the two affected modules, do download only.
Proper fix:
As part of the next phase of changes, the mgmt-framework is moving go-submodule mode, which will ensure the download happens for the assigned commit-ids only.
e.g. Refer go.mod in sonic-telemetry
Master and 201911 use the same version of hw-mgmt 2f65914 , but the hw-mgmt patch was cherry-picked to 201911 badly, the file contents have some diff, and it does not apply on 201911
Signed-off-by: Mykola Faryma <mykolaf@mellanox.com>
Removing explicit libyang plugin path settings. Libyang debian package installs the extensions in standard location.
Libyang debian package installs the extensions in standard location as per build time configuration (install prefix is set in patch/libyang.patch). If plugin directory is not set, libyang explicitly searches the plugins in LYEXT_PLUGINS_DIR which is populated during build. The plugins are installed in LYEXT_PLUGINS_DIR itself through debian package installation.
$ grep -nr LYEXT_PLUGINS_DIR src/
src/plugin_config.h.in:18:#define LYEXT_PLUGINS_DIR "@EXTENSIONS_PLUGINS_DIR_MACRO@" /**< directory with YANG extension plugins */
src/plugins.c:420: pluginsdir = LYEXT_PLUGINS_DIR;
Multi-Db changes for NAT feature. (#1202)
[cfgmgr] clear loopback and vrf in kernel if not warmstart (#1141)
upon cold reboot, skip remove mgmt vrf table from the kernel (#1214)
Multi-Db changes for NAT feature. (#1202)
[cfgmgr] clear loopback and vrf in kernel if not warmstart (#1141)
upon cold reboot, skip remove mgmt vrf table from the kernel (#1214)
* [Mellanox]Integrate hw-mgmt 7.0000.3012
* [sonic-linux-kernel]Advance the submodule head
Advance the sonic-linux-kernel
[sFlow]: Patch to fix skb_over_panic in psample driver (#120)
Added support in the kernel for fullcone 3-tuple unique nat. (#100)
Adding support to compile ARM architecture (#102)
[ixgbe] Support bcm54616s external phy in ixgbe (#122)
Fix i2c ISMT DMA buffer alignment issue (#123)
[mellanox]: Add SN4700 patches. (#126)
This patch upgrade the kernel from version
4.9.0-9-2 (4.9.168-1+deb9u3) to 4.9.0-11-2 (4.9.189-3+deb9u2)
Co-authored-by: rajendra-dendukuri <47423477+rajendra-dendukuri@users.noreply.github.com>
Take advantage of an SDK environment variable to customize the location where sdk_socket exists.
In the latest SDK sdk_socket has been moved from /tmp to /var/run which is a better place to contain this kind of file.
However, this prevents the subdirs under /var/run from being mapped to different volumes. To resolve this, we take advantage of an SDK variable to designate the location of sdk_socket.
This requires every process that requires to access sdk_socket have this environment variable defined. However, to define environment variable for each process is less scalable. We take advantage of the docker scope environment variable to avoid that.
It depends on PR 4227
