This patch is changing the _wait_for_disks() method behavior to wait to
a specific disk if any device hints is specified. There are cases where
the deployment might fail or succeed randomly depending on the order and
time that the disks shows up.
If no root device hints is specified, the method will just wait for any
suitable disk to show up, like before.
The _wait_for_disks call was made into a proper hardware manager method.
It is now also called each time the cached node is updated, not only
on start up. This is to ensure that we wait for the device, matching
root device hints (which are part of the node).
The loop was corrected to avoid redundant sleeps and warnings.
Finally, this patch adds more logging around detecting the root device.
Co-Authored-By: Dmitry Tantsur <dtantsur@redhat.com>
Change-Id: I10ca70d6a390ed802505c0d10d440dfb52beb56c
Closes-Bug: #1670916
When inspecting block devices on a node, discover and
report the /dev/disk/by-path/XXX name along with the
/dev/XXX block device name.
The second name does not change between Linux system
reboots and has greater chances to be the same across
similarly configured nodes.
Note: this patch depends on
https://review.openstack.org/#/c/500524/
library patch, but this dependency can't be expressed
with Depends-On clause. Therefore once this patch
requires a followup patch to enable one currently disabled
test in this patch.
Change-Id: I09874f19890500d352521f89573e2aaf50a29022
Closes-Bug: #1679726
It seems to be incorrect at least for some iLO machines.
Also harden the code against invalid output from ipmitool.
Change-Id: I733785e9c7d86eadca963f0776910504bf91bcfe
Closes-Bug: #1714944
It may happen that BMC is configured to use non-zero channel.
In this case we should iterate across all of them as long as we
get a correct IP address (in this case different than "0.0.0.0"
which is a placeholder for not configured console).
Change-Id: I3c351af1882b24c8f56e4363249b19b5c3a4a446
Closes-Bug: #1702514
Oslo.config deprecated parameter enforce_type and change its
default value to True in Ifa552de0a994e40388cbc9f7dbaa55700ca276b0.
Remove the usage of it to avoid DeprecationWarning: "Using the
'enforce_type' argument is deprecated in version '4.0' and will be
removed in version '5.0': The argument enforce_type has changed its
default value to True and then will be removed completely."
Change-Id: I0f0fb540c43edde64e489915c5199da40a0da9c1
Related--Bug: #1517839
Adds an extra field ``biosdevname`` to network interface inventory
collected by ``default`` inspection collector (which collects the whole
inventory returned by hardware manager) of ironic-python-agent.
This feature requires biosdevname utility to collect the bios given NIC
names. The tooling module for tinyIPA is created for the same purpose.
For CoreOS IPA pxe images, biosdevname tooling module is limited,
because Docker repository is created and embedded into CoreOS pxe
images. The Docker repository uses debian to download the packages.
Debian does not have biosdevname package.
Adds an export variable TINYIPA_REQUIRE_BIOSDEVNAME. Set this
variable to ``true`` in your shell before building tinyIPA.
Closes-Bug: #1635351
Change-Id: Ia96af59e2a74868cac59e5a88cfbb3be60d85687
This change introduces a new base test class that mocks out
utils.execute and forces an exception if it gets called.
This has rooted out many tests that were doing this as a side effect of
calling other functions, doing things like modprobe and running iscsi
on the host's actual machine.
The tests are all now appropriately patched in places where this was
happening, and the new base class permanently prevents this from
accidentally happening again.
If you really want to call utils.execute() then you need to re-mock it
in your unit test.
Change-Id: Idf87d09a9c01a6bfe2767f8becabe65c02983518
Add missing 'autospec' keyword argument to mock.patch and
mock.patch.object calls. Use 'autospec=True' except for a few cases
where it fails because the mocked function is a @classmethod and it
doesn't work. In that case explicity set it to 'autospec=False'
Change-Id: I620dce91abaa4440e1803aeefb3e93c0b65d1419
Global requirements was recently updated to force psutil=>3.0.1. This
patch removes support for older versions of psutil as well as changing
to opportunistically attempt to work if a version >5 is released but
doesn't change the interface we use.
Change-Id: I1f7fab33fd275fb8b5cd7704dc13375402756d06
Related-bug: #1659137
An upper-constraints update to psutil caused IPA to start using psutil
5.0.1. We had a hard-coded assumption that psutil would be major version
1 or 2. This allows us to use the updated psutil and attempts to simply
fail gracefully if an unrecognized psutil version is used.
Change-Id: Ibe072440159561b34a29b478d955876e5fb7f103
Closes-Bug: 1659137
This patch adds a new hardware manager, which will disable the embedded
LLDP agent on Intel CNA network cards in order to allow the gathering of
LLDP data during the inspection process.
Change-Id: I572756ac6a7bf67a7f446738ba9d145e1c1bdc48
Closes-Bug: #1623659
This patch is adding a "hctl" attribute to the BlockDevices. HCTL stands
for: Host, Channel, Target and Lun, which is basically the SCSI address.
The idea behind this patch is to allow root device hints to find the
disk for deployment based on the SCSI address.
Partial-Bug: #1648036
Change-Id: If8897c68609e0df0378ee919b803ca5e497def02
This patch dispatches out the network_interface_info
to allow vendor hardware managers to plug the spacific
implementation. It also move neworking releated methods
form hardware to netutils
Related-Bug: #1532534
Change-Id: Idcd25c4753c009b5ba70bea97ee4eb83391a77a9
This patch is updating IPA to use the match_root_device_hints() method
provided by ironic-lib version 2.2.0.
Partial-Bug: #1561137
Depends-On: I1d9dc7a57ea391a3419710c289242b39a4201463
Change-Id: Id93dd0360137df600f5a656348279e56c6b84bf9
Following OpenStack Style Guidelines[1]:
http://docs.openstack.org/developer/hacking/#unit-tests-and-assertraises
[H203] Unit test assertions tend to give better messages for more specific assertions.
As a result, assertIsNone(...) is preferred over assertEqual(None, ...)
Change-Id: I53c6d6d3e53d9fac38c50faf065c131d3aa55224
This patch is adding a new cleaning step called "erase_devices_metadata"
to the GenericHardwareManager. This step is responsible for erasing the
metadata of the disks present in the node (partition tables, signatures,
filesystem identifiers etc...).
It's important to note that the "erase_devices" cleaning step will also
remove all these metadatas (because it will zero/shred the whole disk)
but, it takes a lot of time to run and for some usages of Ironic only
cleaning the device metadata and leaving the data from previous tenants
on the disk after the machine is recycled is fine. That's the use case
for systems using Ironic just to install the same base image onto many
nodes which will run another cloud on top afterwards (TripleO).
The new cleaning step has a default priority of 99, so it should run
before the "erase_devices" cleaning step so that we can guarantee that
the metadata was removed even in case of a failure when cleaning the
disks.
The version of the GenericHardwareManager was bumped to "1.1" with the
addition of this new clean step.
This patch make use of the "destroy_disk_metadata" method from ironic-lib
to get rid of the metadata.
Closes-Bug: #1603411
Change-Id: I3d7b39d5ee3e03ce63185e4168b1ac954a896c93
When matching the root device hints passed to Ironic and the devices
present in the disk, the logic to the "size" attribute was different was
outside the main loop. This patch refactors that uses the same loop for
matching the size attribute, improving a little the readability.
This patch also makes sure that we convert the value of "size" to
integer before attempting to match it.
Change-Id: I619153232b9b58696369185ad5be90ccfd38a4ed
It is required to issue iscsistart -b on those hardware in order for
/dev/sda /dev/sdb etc to be exposed to the OS.
On this type of hw, without the fix
introspection fails with - missing: ['local_gb']' err.
Change-Id: If6667b123e9b890d97c03612596fb78ee2b0ae85
Closes-Bug: 1590606
This patch is extending the root device hints to also consider if a disk
is a rotational device or not. This hint makes it easier to separate
SSDs and HDDs when selecting the device to deploy the image onto.
Partial-Bug: #1599517
Change-Id: I270fe57df825929bdef7911b3a6757cf7163a5f1
This replaces dict.get(key) with dict[key] in the unit tests,
since we should be explicit about whether we expect the key to
exist or not. (These are tests, after all.)
Change-Id: Ide98ceba325214e55cab2063d9ace150d63f9617
Closes-Bug: #1590808
To support multi-tenant networking in Ironic we need to be able to
discover not just the NICs a baremetal machine has but also the physical
connectivity to switches in the network.
This patch collects LLDP (Link Layer Discovery Protocol) data as part of
the list interfaces stage of the generic hardware manager. This
information can then be processed by the ironic inspector to populate
the local link information on each ironic port.
The processing done on this data in ironic python agent is limited, this
is to allow for server side processing hooks to process as much or as
little of the data as they want. This is to allow for multi-vendor
environments that might use different parts of the LLDP packet to use a
generic ramdisk and configure the processing server side using inspector
plugins.
Reserved fields switch_port_descr and switch_chassis_descr have been
deprecated for removal in Ocata in favor of passing the whole packet.
Change-Id: Idae9b1ede1797029da1bd521501b121957ca1f1a
Partial-Bug: #1526403
This patch replace assertRaisesRegexp with
assertRaisesRegex which is deprecated in python3
https://docs.python.org/3.2/library/unittest.html
Also it update the base tests to be oslotest
BaseTestCase for python2.7 and python3 compatibility
Change-Id: I02571946f0643247e208d98dc91ea78cd9d351ee
https://review.openstack.org/#/c/320295/ introduced two internal
variables: _DISK_WAIT_ATTEMPTS and _DISK_WAIT_DELAY. These values are
hardcoded. This patch adds configuration options for these so
that an operator can change them based on their own needs/fleet of
hardware.
Change-Id: I2ba97669ec710fb4a435307466cd8add9c2293ba
Closes-Bug: #1585663
These flags will be processed in a new ironic-inspector plugin
to support setting capabilities like cpu_vt (virtualization enabled).
Change-Id: I5fe9310c316841eabdd2d5e2ef2ae30afa03d29a
Partial-Bug: #1571580
Adds a new BootInfo object with 2 fields:
* current_boot_mode - bios or uefi, detected from presence of /sys/firmware/efi
as per the following answer: http://askubuntu.com/a/162896
This field will be used for setting the boot_mode capability in ironic-inspector
* pxe_interface - PXE booting interface, if it can be detected.
This fields is already used by ironic-inspector, added here for consistency.
Change-Id: Ib36b592ffaba3bfa055d65c9526607867d302584
Partial-Bug: #1571580
In order to support a more complex syntax for root device hints (e.g
operators: greater than, less than, in, etc...) we need to stop relying
on the kernel command line for passing the root device hints. This patch
changes this approach by getting the root device hints from a cached
node object that was set in the hardware module.
Two new functions: "cache_node" and "get_cached_node" were added to the
hardware module. The idea is to facilitate the access to a node object
representation from the hardware extension methods without changing
method signatures, which would break compatibility with out-of-tree
hardware managers.
Note that the new "get_cached_node" is just a guard function to
facilitate the tests for the code.
The function parse_root_device_hints() and its tests were removed since
it's not used/needed anymore.
Partial-Bug: #1561137
Change-Id: I830fe7da1a59b46e348213b6f451c2ee55f6008c
Some kernel modules take substantial time to initialize. For example,
with mpt2sas RAID driver inspection and deployment randomly fail
due to IPA starting before the driver finishes initialization.
This problem is probably impossible to solve in a generic case, as
modern Linux environment do not have a notion of "hardware is fully
initialized" moment. All hardware is essentially hotplug.
To solve it at least for the simplest case, this patch adds a wait loop
on start up waiting for at least one suitable disk to appear in inventory.
Note that root device hints are not considered, as the node might not
be known at that moment yet.
Change-Id: Id163ca28f7c140c302ea04947ded3f3c58b284de
Partial-Bug: #1582797
I would've voted -1 on the patch in question had I reviewed it, and per
standard OpenStack/Ironic procedure, I'm reverting it for re-review and
discussion.
In this case; I don't think the new method in the HWM interface is
needed, and that evaluate_hardware_support() is intended to handle the
cases handled.
This reverts commit 0962cae1da69a1a2981d5950ad741d91115dac06.
Change-Id: Ic08e44bdf116403444b257ee9f4e5b906f5eac53
Some kernel modules take substantial time to initialize. For example,
with mpt2sas RAID driver inspection and deployment randomly fail
due to IPA starting before the driver finishes initialization.
Add a new hardware manager method initialize_hardware, which gets
run on start up before other hardware manager method invocations.
The generic implementation is to call udev settle and wait for
at least one suitable disk device to appear with the hardcoded
timeout of 15 seconds. Also preload the IPMI modules instead of
calling modprobe every time the inventory is requested.
Change-Id: If7758bb6e3faac7d05451baa3a26adb8ab9953d5
Partial-Bug: #1582797
Introduce a new parameter in driver_internal_info called
agent_erase_devices_zeroize to control the behavior of shred. This
parameter controls the --zero argument used when invoking shred.
Configuring this to false disabled the last pass of zeroes, leaving the
device with random data.
Change-Id: I7053034f5b5bc6737b535ee601e6fb71284d4a83
Partial-bug: #1568811
Depends-On: Ia7ea8d909df9ae86a6dbd68ba94746b171535eb8
Create a helper function: create_hdparm_info() and use it in the unit
tests.
Also fix mispelling of 'indicative'
Change-Id: Ifa15fde72bc0ca6d925408c1dcafce85c192abb7
Presently should the ATA erasure operation fails, IPA halts the
cleaning process and the node goes to CLEANFAIL state as a result.
This failure could be the result of a previous cleaning failure
that left drive security enabled, for which code has been added
in an attempt to address this case by attempting to unlock the
the drive.
In the event that an operator wishes to automatically fallback to
disk scrubbing operations, the capability has been added through
a driver_internal_info field "agent_continue_if_ata_erase_failed"
that can be set to True, however defaults to False keeping the
same behavior that IPA presently exhibits in the event of ATA
erase operations failing.
Partial-Bug: #1536695
Change-Id: I88edd9477f4f05aa55b2fe8efa4bbff1c5573bb1
In the DIB build the DHCP code (provided by the dhcp-all-interfaces element)
races with the service starting IPA. It does not matter for deployment itself,
as we're waiting for the route to the Ironic API to appear. However, for
inspection it may result in reporting back all NIC's without IP addresses.
Inspection fails in this case.
This change makes inspection wait for *all* NIC's to get their IP addresses up
to a small timeout. The timeout is 60 seconds by default and can be changed
via the new ipa-inspection-dhcp-wait-timeout kernel option (0 to not wait).
After the wait inspection proceedes in any case, so the worst downside
is making inspection 60 seconds longer.
To avoid waiting for NIC's that are not even connected, this change extends the
NetworkInterface class with 'has_carrier' field.
Closes-Bug: #1564954
Change-Id: I5bf14de4c1c622f4bf6e3eadbe20c44759da5d66
This patch is making the list_all_block_devices() method to wait for
udev to settle it's event queue prior to listing the devices.
Sometimes the ironic-python-agent service may start before all devices
were detected and end up erroring out because it couldn't find a
suitable disk for deployment.
Closes-Bug: #1551300
Change-Id: I1ae2062a711115a1ea14b79ae9ace7ddd2fff9d5
Changed implementation to strip tokens up until the first 'Size: '
string. This will allow for less parsing errors in the first
six lines of the following output:
"dmidecode --type 17 | grep Size" returns:
Maximum Memory Module Size: 4096 MB
Maximum Total Memory Size: 8192 MB
Size: 2048 MB
Size: 2048 MB
Added a condition in the exception handling to address the
issue of the bug on other outputs like:
Installed Size: Not Installed
Enabled Size: Not Installed
Size: No Module Installed
Size: 1024 MB
Common strings like "No Module Installed" and "Not Installed" are
normal. These two strings are hard coded in the before mentioned
comparison and when found are logged as warnings instead of errors.
Change-Id: If3475afcebfc7af7e9256b99924919557c4d909c
Closes-Bug: #1521202
This patch set add hardware vendor information to data.
By using this data, we can get hints to detect driver.
Change-Id: I39385fd5d616edfad719c255f22642f215bfb532
This patch is extending the root device hints to also look at the device
name. This patch also refactors the tests for root device hints making
it easier to test a different hint per test.
Change-Id: I48d6456c75bbe6ddf16ac6561e5461ca51eb9c37
Partial-Bug: #1526732
Fix params order to correspond to real signature:
assertEqual(expected, actual)
Change-Id: I129aad5d8bb7ca0541c07913cb5edd8eef9e90ba
Closes-Bug: #1277104
Now pyudev raises DeviceNotFoundByFileError which does not inherit
from EnvironmentError, so our 'except' block in hardware.py no longer
catch the exception. It broke unit tests, but it can also potentially
break the deploy.
This patch updates hardware.py to catch both old a new exceptions.
Change-Id: Iaefd6089f6f766a241054d8e132b2f3098c8130d
Closes-Bug: #1522756
