~smoser/curtin/trunk.pre-17.1 : changes

~smoser/curtin/trunk.pre-17.1 » Changes from revision 551

From Revision 551 to 532

Rev	Summary	Authors	Tags	Date
551	merge from trunk merge from trunk	Scott Moser		6 years ago
550	Bump version to 17.0 in anticipation of release of 17.1. Bump version to 17.0 in anticipation of release of 17.1. The version is recorded in two places: curtin/__init__.py debian/changelog.trunk The changelog is just a tool for building debs for developers. The curtin.__version__ however will be exposed in curtin.version So we'll this at 17.0 just to avoid any 17.1 showing up and confusing things until the actual release.	Scott Moser	17.0	6 years ago
549	tools/build-deb: recognize if building a release tag. tools/build-deb: recognize if building a release tag. When a release tag is being built, do not include ~bzrREVNO. This isn't all that important, though as tools/build-deb is only an upstream tool. All official debs (including daily debs) do not use this tool.	Scott Moser		6 years ago
548	vmtests: exercise rootfs over an lvm logical volume vmtests: exercise rootfs over an lvm logical volume - modify mkfs to handle key=value parameter replacement for mkfs.xfs to control the uuid value of the filesystem - Allow templating of the lvmroot yamls to specify different rootfs format types. This allow for testing with multiple fstypes without duplicating input yaml - Fix a lurking CURTIN_REUSE_TOPDIR=1 bug under uefi mode where we attempt to rewrite grub.conf always - Add a specific uefi_lvmroot configuration to match user issue from Bug #1731490 - Update vgcreate, lvcreate invocations to zero and wipe target devices and volumes where possible (> trusty) which eliminates the scary warning about detecting filesystem signatures on lv creation.	Ryan Harper		6 years ago
547	Switch network dep filter to test for ifupdown/nplan instead... Switch network dep filter to test for ifupdown/nplan instead of release name Curtin will install some additional packages in-target depending on the configuration. For example, if the curtin config includes bridges, then it will ensure that bridge-utils is installed. These deps are valid for releases using ifupdown. In Artful and newer (Bionic) netplan is used instead and these tools are not needed. This patch changes a release based check to a package install check. If a system does not have ifupdown and also has netplan (nplan package) then it will NOT install any ifupdown related deps (bridge-utils, vlan, ifenslave).	Ryan Harper		6 years ago
546	Allow control of curtin install unmounting Allow control of curtin install unmounting A Curtin install always unmounts the target. In the subiquity use-case they would like to support a) if something goes wrong give a shell to the user to poke around in the target b) skip unmounting automatically to allow writing of additional files in the target without re-invoking curtin c) control where curtin mounts up storage devices. This branch adds two fields to the curtin 'install' config key: The 'target' value is a path that will be used instead of a temporary directory. The 'unmount' key if found to be present and has the value of 'disabled' then curtin install will skip the unmount operation. Additionally to aid in unmounting a target outside of the install subcommand this branch introduces a curtin 'unmount' subcommand which takes two params: the 'target' path to the mount point, the default 'target' value is to use the TARGET_MOUNT_POINT environment variable. Curtin will by default do unmounts in a recursive manner meaning any directory mounted underneath the target will also be unmounted. A user may supply '--disable-recursive-mounts' as a flag which will disable the recursion resulting in a single unmount operation.	Ryan Harper		6 years ago
545	vmtests: Add Bionic release to tests and update classes. vmtests: Add Bionic release to tests and update classes.	Scott Moser		6 years ago
544	storage: add 'options' key mount type to specify mount param... storage: add 'options' key mount type to specify mount parameters for filesystems Add support, documentation, unittests and vmtests to deliver a new parameter to the mount configuration dictionary. This parameter if set will modify the the mount options field in the rendered /etc/fstab. LP: #1709284	Ryan Harper		6 years ago
543	Re-add curthooks.write_files method for backwards compat Re-add curthooks.write_files method for backwards compat In revno 512, curthooks.write_files() was re-implemented under curtin.futil.write_files and supports the original dictionary format as well as other modes. This broke legacy images which used hooks that invoked curthooks.write_files. This branch restores the function and behavior while utilizing the change in implementation. - Add warning when legacy path is called pointing to new module location - Add unittests to exercise the code path. LP: #1731709	Ryan Harper		6 years ago
542	vmtest: Remove ArtfulTestBridging skip_by_date check, bug fi... vmtest: Remove ArtfulTestBridging skip_by_date check, bug fixed Bug 1721157 is now fixed, so do not need to skip by date anymore.	Scott Moser		6 years ago
541	Drop Precise from vmtest Drop Precise from vmtest It's dead Jim (Precise is EOL)	Ryan Harper		6 years ago
540	clear_holders: bcache log IO/OS exceptions but do not raise clear_holders: bcache log IO/OS exceptions but do not raise While stopping bcache devices, curtin races with the kernel w.r.t how quickly it removes the bcache device (and the sysfs tree related). The result is that as curtin attempts to write "1" into the sysfs tree of bcache device to stop it, the range of errors changes depending on when we race with the kernel asynchronously removing the sysfs path. Therefore we log the exception errno we got but do not re-raise. The the calling process is watching whether the same sysfs path is being removed; if the bcache device fails to go away then curtin will have a log of the exceptions to debug.	Ryan Harper		6 years ago
539	vmtest: Support newer qemu and multipath. vmtest: Support newer qemu and multipath. If you pass the same backing device for a disk to qemu, newer versions (2.10+) will fail. In order to allow this, it requires you to pass 'file.locking=off'. This change makes the multipath tests add a 'file.locking=off' argument to the disks that are actually intended to be added more than once. xkvm is modified to remove that flag if the qemu being used does not support it. In order to accomplish this: a.) make launch use xkvm '--disk' for all its disks rather than passing -drive and -device. This is good in that we know have a single way of passing disks around. b.) fix a logic bug in xkvm when --disk= specified format= on it. vmtest and launch now only use '--disk=' args, which are interpreted by xkvm rather than using a mix of '--disk=' and '-drive'.	Scott Moser		6 years ago
538	block: enable control over exclusive_open use when wiping vo... block: enable control over exclusive_open use when wiping volumes The exclusive_open check on a device is very useful to determine if curtin is removing all metadata from a device w.r.t storage configuraiton. However, in some cases during creation of new storage configations when curtin attempts to add additional partitions to a device that has partitions already added to an raid device, the exclusive open is blocked while we attempt to zero some data on the underlying device. The Mirrorboot-UEFI vmtest exemplifies this situation. This patch exposes a flag to the zero_file_at_offset method to allow callers (block_meta in this case) to disable the exclusive_open. This lets curtin wipe data at specific offsets on a volume without requiring exclusive ownership of the device.	Ryan Harper		6 years ago
537	block: handle wiping bcache parts block: handle wiping bcache parts In some cases block devices may have bcache parts which are owned by the block layer but are not assembled into a bcache device. The result is that curtin fails to get exclusive opens when attempting to wipe the device. Resolve this by testing if a volume has a 'bcache' directory under sysfs and issuing the appropriate stop commands before wiping. It's also possible for old "buried" metadata for bcache or lvm or raid at particular partition offsets. Once curtin starts partitioning the device the bcache kernel layer "finds" the parts and claims device ownership and prevents wiping of the partition. We resolve this issue by having curtin wipe 1M at the location of the partition before we actually create it. Fix 'curtin clear-holders' subcommand. LP: #1718699	Ryan Harper		6 years ago
536	vmtests: Defer ArtfulNetworkMtu SkipbyDate to 2018 vmtests: Defer ArtfulNetworkMtu SkipbyDate to 2018 Unlikely to expect an fix for LP: #1671951 before then.	Ryan Harper		6 years ago
535	bcache: accept sysfs write failure in shutdown handler if pa... bcache: accept sysfs write failure in shutdown handler if path missing Bcache device shutdown is asynchronus. The kernel may remove the device after we've tested to see if the device needs to be stopped but before we issue a 'stop' command to the bcache sysfs file. When that happens, curtin throws an exception; this patch catches those and tests if the target path is missing; and if so, ignores the error as it's expected due to device removal. All other write failures where the path is still present are still valid and reported. LP: #1700564	Ryan Harper		6 years ago
534	vmtest: Rephrase a message about no disks to be less scary vmtest: Rephrase a message about no disks to be less scary In the case where a test has passed and there are no disks to collect we logged an 'error' however in that case it wasn't really an error. This branch changes that message to a debug level bit of information.	Scott Moser		6 years ago
533	block_meta: use block.wipe_volume(mode=superblock) to clear ... block_meta: use block.wipe_volume(mode=superblock) to clear MBR/GPT tables In the case curtin is not directed to wipe the partition table via the wipe: configuration, a disk may contain an MBR and sgdisk --clear does not successfully wipe MBR tables. Replace the call to sgdisk with curtin block.wipe_volume(mode=superblock) to clear both MBR and GPT LP: 1722322	Ryan Harper		6 years ago
532	vmtest: fix artful networking vmtest: fix artful networking - Remove skip_by_date calls in vmtests that are no longer valid - ifenslave brings in ifupdown, so filter out ifenslave package if target release is artful (LP: #1706744) - vlan and bridge-utils are not needed in artful, filter those packages from being installed in artful as well. - Adjust artful bonding test to skip checking for ifslave and instead check that ifenslave is not installed - Drop mtu settings in network-static-routes configuration, the test case only needs to test static routes - Move static routes under an interface subnet to be compatible with netplan format which requires routes under an interface - Refactor ip_route_show parsing, ignoring default route and fetching variable settings like 'proto'. - Skiptest bridging in artful, need a cloud-init fix for stp in netplan (LP: #1721157) - Skiptest for mtu in artful, need networkd to support mtu6 support (LP: #1671951) Fixes: LP:#1714028 LP:#1718216, LP:#1706744	Ryan Harper		6 years ago

Older »