119e7928 | 18-Oct-2024 |
Abby <buchmana@google.com> |
metric-ipmi-blobs: adding tmpfs
Adding tmpfs metric coverage to pre-ttf machines
Tested: ``` df -h Filesystem Size Used Available Use% Mounted on dev 468.9M
metric-ipmi-blobs: adding tmpfs
Adding tmpfs metric coverage to pre-ttf machines
Tested: ``` df -h Filesystem Size Used Available Use% Mounted on dev 468.9M 0 468.9M 0% /dev tmpfs 469.9M 12.6M 457.3M 3% /run /dev/mtdblock4 20.6M 20.6M 0 100% /run/initramfs/ro /dev/mtdblock6 3.0M 436.0K 2.6M 14% /run/initramfs/rw cow 3.0M 436.0K 2.6M 14% / tmpfs 469.9M 8.0K 469.9M 0% /dev/shm tmpfs 469.9M 0 469.9M 0% /tmp tmpfs 469.9M 40.0K 469.8M 0% /var/volatile tmpfs 469.9M 12.6M 457.3M 3% /var/lib/systemd/coredump ```
Change-Id: I50c763707268252ed51a903d8b7e02543dab6166 Signed-off-by: Abby <buchmana@google.com>
show more ...
|
a06304b0 | 16-Sep-2024 |
Yuxiao Zhang <yuxiaozhang@google.com> |
bare-metal-gpio-monitor: don't start on non-bm mode
This adds a condition check for the bm flag. So that the service will not take effect on non-bm mode.
Change-Id: I0d15247a6e05aa5edf93468793cd74b
bare-metal-gpio-monitor: don't start on non-bm mode
This adds a condition check for the bm flag. So that the service will not take effect on non-bm mode.
Change-Id: I0d15247a6e05aa5edf93468793cd74bd7d9ea158 Signed-off-by: Yuxiao Zhang <yuxiaozhang@google.com>
show more ...
|
3a5c9f3a | 03-Sep-2024 |
Patrick Williams <patrick@stwcx.xyz> |
meson: switch code dependency to libsystemd
For a while now systemd has packaged a separate `libsystemd`. For meson dependency, `systemd` should be used when looking for systemd service file paths
meson: switch code dependency to libsystemd
For a while now systemd has packaged a separate `libsystemd`. For meson dependency, `systemd` should be used when looking for systemd service file paths and `libsystemd` should be used for `sd_*` functions.
Update the dependency accordingly.
Signed-off-by: Patrick Williams <patrick@stwcx.xyz> Change-Id: I9241adf8ce73df70fb7e50cc50f16015322bea5b
show more ...
|
c66ebc35 | 16-Aug-2024 |
Patrick Williams <patrick@stwcx.xyz> |
clang-format: re-format for clang-18
clang-format-18 isn't compatible with the clang-format-17 output, so we need to reformat the code with the latest version. The way clang-18 handles lambda forma
clang-format: re-format for clang-18
clang-format-18 isn't compatible with the clang-format-17 output, so we need to reformat the code with the latest version. The way clang-18 handles lambda formatting also changed, so we have made changes to the organization default style format to better handle lambda formatting.
See I5e08687e696dd240402a2780158664b7113def0e for updated style. See Iea0776aaa7edd483fa395e23de25ebf5a6288f71 for clang-18 enablement.
Change-Id: I362352dcb341658501899267c2ff3ad044ed5912 Signed-off-by: Patrick Williams <patrick@stwcx.xyz>
show more ...
|
1e76060a | 07-Mar-2024 |
Yuxiao Zhang <yuxiaozhang@google.com> |
meta-google: dhcp-done: Adding status report
Previously dhcp-done only sends status code, this one provides the capability to send status code + status message for better troubleshooting. Provide a
meta-google: dhcp-done: Adding status report
Previously dhcp-done only sends status code, this one provides the capability to send status code + status message for better troubleshooting. Provide a way to let other process upgrade the status.
Tested: Unit test passed.
Change-Id: I9c689f90502a32b586c41e3491ad47ebc78fcc38 Signed-off-by: Yuxiao Zhang <yuxiaozhang@google.com>
show more ...
|
177e7270 | 10-Jul-2024 |
Yuxiao Zhang <yuxiaozhang@google.com> |
ncsid: fix default wait time
The default wait time needs to shift with current timestamp, otherwise the script will keep looping with 0 intervals and 0 wait time in RA, if the initial loop does not
ncsid: fix default wait time
The default wait time needs to shift with current timestamp, otherwise the script will keep looping with 0 intervals and 0 wait time in RA, if the initial loop does not capture any RA data.
Tested: bring down the interface to simulate the nic not ready, start this script and verified that it loops every 10s. Then bring up the interface, and see the route is setup with no issue
Change-Id: I4189f39c1e9efc0dd9dbbdf1116c505b755a6982 Signed-off-by: Yuxiao Zhang <yuxiaozhang@google.com>
show more ...
|
08988f3d | 26-Jun-2024 |
William A. Kennington III <wak@google.com> |
ncsid: update_ra_gw: Fix old_rtr check
An iteration of a previous commit changed the default value of `old_rtr` from empty string to explicitly `invalid`. We need to use the correct comparison for e
ncsid: update_ra_gw: Fix old_rtr check
An iteration of a previous commit changed the default value of `old_rtr` from empty string to explicitly `invalid`. We need to use the correct comparison for early router setting.
Change-Id: I27c198a93cf06622c6fc6f01092f12d72e05f722 Signed-off-by: William A. Kennington III <wak@google.com>
show more ...
|
f62e303c | 18-Jun-2024 |
Ed Tanous <ed@tanous.net> |
Remove Ed from Google-misc maintainers
Change-Id: I92a9e7ece6a2ae170989d0022376d5b0a1e7f12b Signed-off-by: Ed Tanous <ed@tanous.net> |
a9bd43e7 | 20-Jun-2024 |
William A. Kennington III <wak@google.com> |
ncsid: Make fixup run for all RA
We want to ensure we get regular fixes to the router state without switching to different routers when our route accidentally disappears.
Change-Id: I14c4ea63b51474
ncsid: Make fixup run for all RA
We want to ensure we get regular fixes to the router state without switching to different routers when our route accidentally disappears.
Change-Id: I14c4ea63b514740d846063aaaac35e3c8998cb9e Signed-off-by: William A. Kennington III <wak@google.com>
show more ...
|
48282226 | 18-Jun-2024 |
William A. Kennington III <wak@google.com> |
ncsid: Reduce networkd reloads
This should help to avoid race conditions with phosphor-networkd. We still write the networkd files so that when reloads occur the information is updated correctly.
C
ncsid: Reduce networkd reloads
This should help to avoid race conditions with phosphor-networkd. We still write the networkd files so that when reloads occur the information is updated correctly.
Change-Id: I550ea04deb8de12db8c4490e9f26f0d3a04baef7 Signed-off-by: William A. Kennington III <wak@google.com>
show more ...
|
1e2289c8 | 17-Jun-2024 |
William A. Kennington III <wak@google.com> |
ncsid: Prefer fe80::1 gateway
Our labs and production environment normally use this as the gateway address. We often have noisy lab networks with many bad gateways advertising their presence. We kno
ncsid: Prefer fe80::1 gateway
Our labs and production environment normally use this as the gateway address. We often have noisy lab networks with many bad gateways advertising their presence. We know we generally prefer fe80::1 so we switch to it when available.
Tested: On a noisy lab network and verified that the script correctly locks onto fe80::1 and switches away from other gateways it receives.
Change-Id: I9ed6c5c2df7b1dec8015dcdfc0d320d471b5e7f3 Signed-off-by: William A. Kennington III <wak@google.com>
show more ...
|
a9b1822b | 17-Jun-2024 |
William A. Kennington III <wak@google.com> |
ncsid: Find IPv6 gateway stably
Some environments have multiple gateways for the node to use. We want to ensure that we aren't flip flopping between multiple gateways and lock on to a specific gatew
ncsid: Find IPv6 gateway stably
Some environments have multiple gateways for the node to use. We want to ensure that we aren't flip flopping between multiple gateways and lock on to a specific gateway until it goes away.
Tested: Ran in an environment with multiple gateways and verified that it did not switch after it acquired the first one.
Change-Id: I12bac9582d35bdca49320fab8f5063ebfb2f3986 Signed-off-by: William A. Kennington III <wak@google.com>
show more ...
|
5f0ae6fa | 19-Apr-2024 |
William A. Kennington III <wak@google.com> |
ncsid: Remove unused services
We no longer need IPv4 support or any of the neighbor finding logic, as we have all of the neighbor information updated via the primary RA from the router.
Tested: Ran
ncsid: Remove unused services
We no longer need IPv4 support or any of the neighbor finding logic, as we have all of the neighbor information updated via the primary RA from the router.
Tested: Ran on a lab and prod machine to verify the networking behavior is still consistent and gateway information is updated accurately.
Change-Id: I0e247291ae3a1f98064a3bc9a94b28d9a2223843 Signed-off-by: William A. Kennington III <wak@google.com>
show more ...
|
b47ca0df | 30-May-2024 |
Yuxiao Zhang <yuxiaozhang@google.com> |
ncsid: reconfigure gateway when default route is missing
This will ensure that default gateway can be reconfigured if any failure happens at the first time.
Tested: On a machine that has default ro
ncsid: reconfigure gateway when default route is missing
This will ensure that default gateway can be reconfigured if any failure happens at the first time.
Tested: On a machine that has default route configured, manually deleted the route, without this change it will stay broken. This change will bring the route back in 1mins or so.
Change-Id: I7f8f1fafcbcfb3f4b904af8999d1a5c17400e99e Signed-off-by: Yuxiao Zhang <yuxiaozhang@google.com>
show more ...
|
4dba220d | 08-Apr-2024 |
Willy Tu <wltu@google.com> |
metrics-ipmi-blobs: Support ECC Error Counts
Export the ECC Error Cournter from http://github.com/openbmc/phosphor-ecc to ipmi blob.
Tested: The blob output is updated if MemoryECC dbus object exis
metrics-ipmi-blobs: Support ECC Error Counts
Export the ECC Error Cournter from http://github.com/openbmc/phosphor-ecc to ipmi blob.
Tested: The blob output is updated if MemoryECC dbus object exists compared to the blob output if MemoryECC doesn't exists.
Change-Id: I2c63dbcd0970afc587f5c2ee01f8c261909f0b08 Signed-off-by: Willy Tu <wltu@google.com>
show more ...
|
0ddcb163 | 22-Mar-2024 |
William A. Kennington III <wak@google.com> |
ncsid: lib: Fix neighbor path
It no longer has its own hierarchy. Instead, it's directly below the ethernet object now.
Tested: Verified that neighbors are now correctly identified when they need t
ncsid: lib: Fix neighbor path
It no longer has its own hierarchy. Instead, it's directly below the ethernet object now.
Tested: Verified that neighbors are now correctly identified when they need to be deleted.
Change-Id: Ibd319dcc84fcff99d7c45a5e2fb2de58b6f84406 Signed-off-by: William A. Kennington III <wak@google.com>
show more ...
|
d36d9ef6 | 19-Mar-2024 |
William A. Kennington III <wak@google.com> |
ncsid: update_ra_gw: Improve gateway discovery speed
The rdisc6 binary when run in a subshell buffers all of the output instead of sending each line. This results in a long delay to figure out the g
ncsid: update_ra_gw: Improve gateway discovery speed
The rdisc6 binary when run in a subshell buffers all of the output instead of sending each line. This results in a long delay to figure out the gateway address.
We can fix this by running it through `script`, which removes the buffering and allows us to nearly instantly update our route.
Tested: Ran on our AMD Rome system and verified that the gateway was quickly being updated.
Change-Id: I22350d6bd95665ab853c3f1273c0d6e4ff60718e Signed-off-by: William A. Kennington III <wak@google.com>
show more ...
|
5034c566 | 19-Mar-2024 |
William A. Kennington III <wak@google.com> |
ncsid: update_ra_gw: Don't depend on phosphor-networkd
The daemon is currently racy for setting the GW settings, don't depend on it if we don't need to. We always prefer the RA discovered gateway se
ncsid: update_ra_gw: Don't depend on phosphor-networkd
The daemon is currently racy for setting the GW settings, don't depend on it if we don't need to. We always prefer the RA discovered gateway settings, so override them.
Tested: Ran on our AMD Rome system and verified that the gateway files get generated correctly and the system was using the right routes.
Change-Id: Ibeea9f02370ee8eefdee3eff29ea5c528890951d Signed-off-by: William A. Kennington III <wak@google.com>
show more ...
|
7f49370d | 08-Feb-2024 |
William A. Kennington III <wak@google.com> |
metrics-ipmi-blobs: Convert to nanopb
Tested: Decoded a metric proto from a host and verified that the fields all looked as expected and "normal" with decode successfully identifying everything.
Ch
metrics-ipmi-blobs: Convert to nanopb
Tested: Decoded a metric proto from a host and verified that the fields all looked as expected and "normal" with decode successfully identifying everything.
Change-Id: Iaa1b5500db6d1b93fae95d44a1045a858624cc51 Signed-off-by: William A. Kennington III <wak@google.com>
show more ...
|
70a33437 | 09-Feb-2024 |
William A. Kennington III <wak@google.com> |
subprojects: Fix wrapfile provides
Our packages depend on being able to map the dependencies so we should add them to the wrap.
Change-Id: Ie4500329c28d18bbde8d8c7e51015f5a3ad48466 Signed-off-by: W
subprojects: Fix wrapfile provides
Our packages depend on being able to map the dependencies so we should add them to the wrap.
Change-Id: Ie4500329c28d18bbde8d8c7e51015f5a3ad48466 Signed-off-by: William A. Kennington III <wak@google.com>
show more ...
|
8800984d | 04-Jan-2024 |
Yuxiao Zhang <yuxiaozhang@google.com> |
bare-metal-host-monitor: capture state and action
state/action needs to be captured by value instead of reference.
Change-Id: I68dbc5bd91979b9fc21e7911a4b6b1433a9852ba Signed-off-by: Yuxiao Zhang <
bare-metal-host-monitor: capture state and action
state/action needs to be captured by value instead of reference.
Change-Id: I68dbc5bd91979b9fc21e7911a4b6b1433a9852ba Signed-off-by: Yuxiao Zhang <yuxiaozhang@google.com>
show more ...
|
dd9478dd | 08-Dec-2023 |
John Wedig <johnwedig@google.com> |
Bare-metal: post complete GPIO monitor
This creates a service that monitors the post complete gpio, it reenabled ipmi when gpio is deasserted. And at the startup it checks the gpio and disable ipmi
Bare-metal: post complete GPIO monitor
This creates a service that monitors the post complete gpio, it reenabled ipmi when gpio is deasserted. And at the startup it checks the gpio and disable ipmi if it is asserted.
Tested: 1. reset host, ipmi is re-enabled 2. reboot bmc, ipmi is disabled after bmc booting up.
Change-Id: If52c72ce57e10f6efbb94fd3f0bdcb7655b48d61 Signed-off-by: John Wedig <johnwedig@google.com> Signed-off-by: Yuxiao Zhang <yuxiaozhang@google.com>
show more ...
|
dca92e47 | 17-Nov-2023 |
Willy Tu <wltu@google.com> |
update libcr51sign to support MAUV
Add feature support to manage MAUV with cr51 descriptor.
Change-Id: I3795259f81b1dbd3016d1c41c68895be00455061 Signed-off-by: Willy Tu <wltu@google.com> |
253e6463 | 28-Dec-2023 |
Willy Tu <wltu@google.com> |
build: Remove dependencies on fmtlib
With C++23, we can replace fmt::format with std::format and fmt::print with stdplus::print. This allows us to migrate to std::print once the support is available
build: Remove dependencies on fmtlib
With C++23, we can replace fmt::format with std::format and fmt::print with stdplus::print. This allows us to migrate to std::print once the support is available.
Change-Id: I790901fb3b96d815275a3919e1fb90973189fc02 Signed-off-by: Willy Tu <wltu@google.com>
show more ...
|
47a0ada2 | 29-Nov-2023 |
Patrick Williams <patrick@stwcx.xyz> |
build: use allowed over enabled
Meson feature options are typically in a tri-state of enabled, disabled, or auto. The enabled and disabled functions on an option (from `get_option`) no longer retur
build: use allowed over enabled
Meson feature options are typically in a tri-state of enabled, disabled, or auto. The enabled and disabled functions on an option (from `get_option`) no longer return true for auto features. Instead, the expectation is to use `allowed()` which is true for both enabled and auto.
Switch all uses of `enabled` to `allowed`.
Change-Id: I91fe34f9d6743a3afeef73c3e55847542cea143c Signed-off-by: Patrick Williams <patrick@stwcx.xyz>
show more ...
|