413 may return on the upload, but sometimes not the status
query. Support the direct evaluation of the return code.
Change-Id: I4e8e8f4a03adc00d4c1de3a1fe4fbcb5d1b3d8f3
Fallbackdata was actually being preferred rather than a
last resort. Fix by changing to a more usual return
scheme.
Change-Id: I098dfc424c173c87c24a712efdebefb46f724473
It was observed that an adapter update could be inflicted
with an expiring token. Aggressively refresh to mitigate
this.
Change-Id: I7624e8c9abbfdc4e8fd3b323f868d3606cd0039e
Certain calls were exceeding the lifetime of the token, add some
refreshes to keep token up to date.
Change-Id: I539f6323715ddacc5b435a61b2a1d9af65f29c2b
If firmware offers up an unparseable expression, prcoeed as if the
expression were not there in the first place. The occurance observed
was not viable. This hampers the improved error message, but the
firmware will still enforce any rules.
Change-Id: I2be9db12b9fa8da9d40bf6829b3cdc02bbdc946d
Independent concurrent requests are going to be a way of life, add
a generic helper. Future modify to take a dict of urls to handlers
to launch handlers in a pool as well.
Names are usually more descriptive, but are occasionally not unique.
Fallback to Id for devices that conflict.
Store name map to urls for faster performance of
get_inventory_of_component
Change-Id: I83207d24c8bd2a23ac0d16ab5e1777f7df8d4369
This is a mechanically generated change to replace openstack.org
git:// URLs with https:// equivalents.
This is in aid of a planned future move of the git hosting
infrastructure to a self-hosted instance of gitea (https://gitea.io),
which does not support the git wire protocol at this stage.
This update should result in no functional change.
For more information see the thread at
http://lists.openstack.org/pipermail/openstack-discuss/2019-March/003825.html
Change-Id: Ib91d991abd3e2166797130573808810492a88089
This permits the code to be able to risk redundant fetches without high
performance penalty through caching.
It also introduces a way for calling code to provide a concurrency
pool, e.g. a GreenPool from eventlet.
Change-Id: I60c4aabb5064a5ee19b0804a520d4ace1195710e
This is compatible with the ipmi model for now, and raising exceptions
in multi-interface environments. Future will have helper to list and
use interface as parameter.
Change-Id: I648edc1546a12dbd7f3ca9548ce6db3dfadd4827
We can have a SensorReading object that at least approximates ipmi
object, though we can never have readings.
Change-Id: Id33346263521e39f7b70ada456d0d134b6264e5c
It is possible for fault LED to be lit without
any active events. Have this trigger a critical,
since we have no idea what would have caused the light.
Change-Id: Id4cbbc64b0dd44c2757ce1f30deaa28a7e668472
The inventory had diverged between firmware and hardware,
for now just give each a dedicated name.
Change-Id: I8f755a0f9ed21369a1c05e7e3139ab43c645a01f
If software attempts to upload a file
larger than the XCC supports. Report
a more understandable condition in
such a case.
Change-Id: I8b5a80c0be8fbca141bdce34d5da39095217b30b
Error string was incorrectly trying to use '+' to
add number into the message. Fix by using format
Change-Id: Iedc83383ffa4770b28538895369c9b3d58bb5bad
Measure the M.2 status of a disk more
directly, avoiding modifying the slotNo unless
the disk is actually M.2
Change-Id: Ic8bd578654f586791f04d8f880842cc11e678ed5
While a thorn in theory could be valid,
in practice the more likely intention
is yet another nonstandard string termination
attempt.
Change-Id: I71f837725b1e286b9f94786aa520deb12ce208e4
Lenovo tools use mriName to label configuration groups. Change to this
by default, while retaining legacy name
as alias for backwards compatibility.
Change-Id: I220ab18e21947745cf942d78536668d580ffa38b
The first verify phase may be delayed,
and in such a case could straddle bad
timing of the token expiry. Refresh
between the phases to mitigate risk
of expiring the token in the middle of
running.
Change-Id: Ia96fa8824a2e9e8eff8fd2d353b1b0bc7fa710a7
If the XCC/IMM bails out in the midst of an update,
it was inducing a vague result. Rework to be more
specific in the error.
Change-Id: Ie35735cc2581e7bcd7a0f4484650366d37ea613d
An ordered setting (e.g. bootorder) requires
different handling. Support it by the appropriate modifications.
Change-Id: I5b8931f3b034ee8f48285797bec4a4016a5fecc5
The builddeb script attempted to build in a directory
without .git, breaking the version deterimination logic.
Change-Id: Ia2be3a75bdcdf9335586bb40e05e635f4b497ca5
Debian wheezy has a python vintage that
errors on unpack from a bytearray.
Explicitly cast to bytes for unpack
to compensate.
Change-Id: Icaf3931a905102e6ee6ef03d91617c031a1afc86
Debian wheezy is just too old to easily work with,
for that platform, use the more amenable
cryptodomex library.
Change-Id: Ic21d9784158d9d3f529d8c17dbe60a49f244cc7c
Without pbr, the list has to be manually curated.
There is a setuptools directive that will work and
is used by pbr, but for now use the output with some
modules trimmed.
Change-Id: I772b7bef263639167b3b5c54bc7812e1e8acbaf6
There are some build environments we are being asked
to support that pbr does not. A quick fallback
to handle such environments.
Change-Id: Ie5dac2555c73e80e4cf589932a18f31a530bdf65
When the retry logic was moved to console, it exposed a problem with
the sol handler being called in the midst of transmitting without
clearing the pending payload, resulting in an infinite recursion
of sending the same break.
Resolve by using the common call to sendoutput to have it occur
only after the current head of pending output has been removed.
Change-Id: Iaae86a3bd870c1841530e200ef0f5780a5cc6993
SMM may not allow a new user to access web. Make the user_access
extensible and use that to have web privilege follow the ipmi
privilege.
Change-Id: Ib08e0064b3fb402de833a159c471e28579520a5d
