aboutsummaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAge
* pkg/lookup: Return ID-only pointers on ErrNo*EntriesW. Trevor King2018-12-04
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | Callers that only care about the IDs should try to convert the identifier to an integer before calling the Get* functions, so they can save the cost of hitting the filesystem and maybe or maybe not finding the other fields (User.Name, etc.). But callers that *want* the other fields but only actually need the ID can, with this commit, just call the Get* function and ignore ErrNo*Entries responses: user, err := lookup.GetUser(mount, userIDorName) if err != nil && err != ErrNoPasswdEntries { return err } Previously, they'd have to perform their own integer-conversion attempt in Get* error handling, with logic like: user, err := lookup.GetUser(mount, userIDorName) if err == ErrNoPasswdEntries { uuid, err := strconv.ParseUint(userIDorName, 10, 32) if err == nil { user.Uid = int(uuid) } } else if err != nil { return err } Signed-off-by: W. Trevor King <wking@tremily.us>
* libpod/container_internal_linux: Allow gids that aren't in the group fileW. Trevor King2018-12-04
| | | | | | | | | | | | | | | | | | | | | | | | | | | When an image config sets config.User [1] to a numeric group (like 1000:1000), but those values do not exist in the container's /etc/group, libpod is currently breaking: $ podman run --rm registry.svc.ci.openshift.org/ci-op-zvml7cd6/pipeline:installer --help error creating temporary passwd file for container 228f6e9943d6f18b93c19644e9b619ec4d459a3e0eb31680e064eeedf6473678: unable to get gid 1000 from group file: no matching entries in group file However, the OCI spec requires converters to copy numeric uid and gid to the runtime config verbatim [2]. With this commit, I'm frontloading the "is groupspec an integer?" check and only bothering with lookup.GetGroup when it was not. I've also removed a few .Mounted checks, which are originally from 00d38cb3 (podman create/run need to load information from the image, 2017-12-18, #110). We don't need a mounted container filesystem to translate integers. And when the lookup code needs to fall back to the mounted root to translate names, it can handle erroring out internally (and looking it over, it seems to do that already). [1]: https://github.com/opencontainers/image-spec/blame/v1.0.1/config.md#L118-L123 [2]: https://github.com/opencontainers/image-spec/blame/v1.0.1/conversion.md#L70 Signed-off-by: W. Trevor King <wking@tremily.us>
* Merge pull request #1894 from jwhonce/bug/1876OpenShift Merge Robot2018-12-04
|\ | | | | Only include container SizeRootFs when requested
| * Only include container SizeRootFs when requestedJhon Honce2018-11-28
| | | | | | | | | | | | | | | | * API always returns value, so we remove it if not asked for Fixes #1876 Signed-off-by: Jhon Honce <jhonce@redhat.com>
* | Merge pull request #1907 from baude/createpodautomaticallyOpenShift Merge Robot2018-12-04
|\ \ | | | | | | create pod on the fly
| * | create pod on the flybaude2018-12-03
|/ / | | | | | | | | | | | | | | | | | | when a user specifies --pod to podman create|run, we should create that pod automatically. the port bindings from the container are then inherited by the infra container. this signicantly improves the workflow of running containers inside pods with podman. the user is still encouraged to use podman pod create to have more granular control of the pod create options. Signed-off-by: baude <bbaude@redhat.com>
* | Merge pull request #1913 from baude/podexistsOpenShift Merge Robot2018-12-03
|\ \ | | | | | | podman pod exists
| * | podman pod existsbaude2018-12-03
| | | | | | | | | | | | | | | | | | | | | | | | | | | like containers and images, users would benefit from being able to check if a pod exists in local storage. if the pod exists, the return code is 0. if the pod does not exists, the return code is 1. Any other return code indicates a real errors, such as permissions or runtime. Signed-off-by: baude <bbaude@redhat.com>
* | | Merge pull request #1923 from giuseppe/rootless-no-newuidmap-installedOpenShift Merge Robot2018-12-03
|\ \ \ | |/ / |/| | rootless: raise error if newuidmap/newgidmap are not installed
| * | rootless: raise error if newuidmap/newgidmap are not installedGiuseppe Scrivano2018-12-03
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | it was reported on IRC that Podman on Ubuntu failed as newuidmap/newgidmap were not installed by default. Raise an error if we are not allowing single mappings (used only by the tests suite) and any of the binaries is not present. Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>
* | | Merge pull request #1914 from baude/logslatestOpenShift Merge Robot2018-12-03
|\ \ \ | |/ / |/| | Add short-option handling to logs
| * | Add short-option handling to logsbaude2018-12-01
|/ / | | | | | | | | | | | | podman logs already supports the latest command line switch. users should be able to use the short-options combined (i.e. podman logs -lf). Signed-off-by: baude <bbaude@redhat.com>
* | Merge pull request #1903 from giuseppe/fix-rootless-testsOpenShift Merge Robot2018-12-01
|\ \ | | | | | | tests: fix rootless tests on Ubuntu
| * | tests: always install runc on UbuntuGiuseppe Scrivano2018-12-01
| | | | | | | | | | | | Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>
| * | cirrus: update ubuntu imageGiuseppe Scrivano2018-11-30
| | | | | | | | | | | | Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>
| * | cirrus: make apt noninteractiveGiuseppe Scrivano2018-11-30
| | | | | | | | | | | | Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>
| * | Dockerfile, .cirrus.yml: update runc commitGiuseppe Scrivano2018-11-30
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | we need to inherit this change from runc. commit 869add33186caff4a22e3e11a7472a2d48d77889: rootless: fix running with /proc/self/setgroups set to deny This is a regression from 06f789cf26774dd64cb2a9cc0b3c6a6ff832733b when the user namespace was configured without a privileged helper. To allow a single mapping in an user namespace, it is necessary to set /proc/self/setgroups to "deny". For a simple reproducer, the user namespace can be created with "unshare -r". Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>
| * | rootless: propagate XDG_RUNTIME_DIR to the OCI runtimeGiuseppe Scrivano2018-11-30
| | | | | | | | | | | | Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>
| * | Update ubuntu VM image w/ newer runcChris Evich2018-11-30
|/ / | | | | | | Signed-off-by: Chris Evich <cevich@redhat.com>
* | Merge pull request #1896 from ypu/mount-create-testOpenShift Merge Robot2018-11-30
|\ \ | | | | | | Add create test with --mount flag
| * | Add create test with --mount flagYiqiao Pu2018-11-29
| |/ | | | | | | Signed-off-by: Yiqiao Pu <ypu@redhat.com>
* | Merge pull request #1886 from edsantiago/pypod_run_argsOpenShift Merge Robot2018-11-30
|\ \ | | | | | | pypod run: ignore args intended for container command
| * | pypod create/run: ignore args for container commandEd Santiago2018-11-28
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Don't try to argparse command-line arguments on the right-hand side of the image; those are intended for the container command: pypodman create fedora ls -l pypodman run fedora find / -name foo pypodman run fedora bash -c 'echo hi' If/when `pypodman exec` gets implemented, it should use this too. Signed-off-by: Ed Santiago <santiago@redhat.com>
* | | Merge pull request #1317 from rhatdan/privilegedOpenShift Merge Robot2018-11-30
|\ \ \ | | | | | | | | Disable mount options when running --privileged
| * | | Disable mount options when running --privilegedDaniel J Walsh2018-11-28
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | We now default to setting storage options to "nodev", when running privileged containers, we need to turn this off so the processes can manipulate the image. Signed-off-by: Daniel J Walsh <dwalsh@redhat.com>
| * | | Vendor in latest containers/storageDaniel J Walsh2018-11-28
| | | | | | | | | | | | | | | | | | | | | | | | This allows us to modify the containers mount option on a per/container basis Signed-off-by: Daniel J Walsh <dwalsh@redhat.com>
* | | | Merge pull request #1777 from rhatdan/shmOpenShift Merge Robot2018-11-30
|\ \ \ \ | | | | | | | | | | /dev/shm should be mounted even in rootless mode.
| * | | | /dev/shm should be mounted even in rootless mode.Daniel J Walsh2018-11-28
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Currently we are mounting /dev/shm from disk, it should be from a tmpfs. User Namespace supports tmpfs mounts for nonroot users, so this section of code should work fine in bother root and rootless mode. Signed-off-by: Daniel J Walsh <dwalsh@redhat.com>
* | | | | Merge pull request #1906 from baude/addpspOpenShift Merge Robot2018-11-30
|\ \ \ \ \ | |_|_|_|/ |/| | | | add pod short option to ps
| * | | | add pod short option to psbaude2018-11-30
|/ / / / | | | | | | | | | | | | | | | | | | | | podman ps has a flag --pod; simply adding a short option of -p Signed-off-by: baude <bbaude@redhat.com>
* | | | Merge pull request #1890 from baude/disablecheckpointtestsOpenShift Merge Robot2018-11-28
|\ \ \ \ | | | | | | | | | | disable checkpoint tests on f29
| * | | | disable checkpoint tests on f29baude2018-11-28
| | |_|/ | |/| | | | | | | | | | | | | | | | | | temporarily disabling checkpoint tests on f29 as they don't currently pass. Signed-off-by: baude <bbaude@redhat.com>
* | | | Merge pull request #1714 from giuseppe/fix-rootless-testsOpenShift Merge Robot2018-11-28
|\ \ \ \ | | | | | | | | | | tests: fix rootless tests with an updated runc
| * | | | test, rootless: specify USER env variableGiuseppe Scrivano2018-11-28
| | | | | | | | | | | | | | | | | | | | Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>
| * | | | Revert "downgrade runc due a rootless bug"Giuseppe Scrivano2018-11-28
| | |/ / | |/| | | | | | | | | | | | | | | | | | This reverts commit 803efd4d380f091c85cd45e0b3c35cf6333d47b7. Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>
* | | | Merge pull request #1852 from rhatdan/completionsOpenShift Merge Robot2018-11-28
|\ \ \ \ | |/ / / |/| | | Cleaning up completions.
| * | | Fix completions to work with podman run commandDaniel J Walsh2018-11-28
| |/ / | | | | | | | | | | | | | | | | | | Also add missing --help and -h and add some consistency to the usage of boolean_options. Signed-off-by: Daniel J Walsh <dwalsh@redhat.com>
* | | Merge pull request #1888 from baude/hidekubeOpenShift Merge Robot2018-11-28
|\ \ \ | |/ / |/| | hide kube command for now
| * | hide kube command for nowbaude2018-11-28
|/ / | | | | | | | | | | | | until the kube commands are ironed out, we dont want it drawing attention in any release Signed-off-by: baude <bbaude@redhat.com>
* | Merge pull request #1851 from jwhonce/wip/cmdOpenShift Merge Robot2018-11-28
|\ \ | | | | | | Add support for --all in pypodman ps command
| * | Add support for --all in pypodman ps commandJhon Honce2018-11-28
| | | | | | | | | | | | | | | | | | | | | | | | * Updated field widths to match changes in go code Fixes #1654 Signed-off-by: Jhon Honce <jhonce@redhat.com>
* | | Merge pull request #1798 from giuseppe/fix-notify-socketOpenShift Merge Robot2018-11-28
|\ \ \ | | | | | | | | systemd: fix NOTIFY_SOCKET with patched runc
| * | | tests: fix NOTIFY_SOCKET testGiuseppe Scrivano2018-11-28
| | | | | | | | | | | | | | | | Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>
| * | | oci: propagate NOTIFY_SOCKET on runtime startGiuseppe Scrivano2018-11-28
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | with https://github.com/opencontainers/runc/pull/1807 we moved the systemd notify initialization from "create" to "start", so that the OCI runtime doesn't hang while waiting on reading from the notify socket. This means we also need to set the correct NOTIFY_SOCKET when start'ing the container. Closes: https://github.com/containers/libpod/issues/746 Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>
| * | | test: fix test for NOTIFY_SOCKETGiuseppe Scrivano2018-11-28
| | | | | | | | | | | | | | | | | | | | | | | | do not make any assumption on the path inside of the container. Signed-off-by: Giuseppe Scrivano <gscrivan@redhat.com>
* | | | Merge pull request #1885 from edsantiago/split_token_noneOpenShift Merge Robot2018-11-28
|\ \ \ \ | | | | | | | | | | _split_token(): handle None
| * | | | _split_token(): handle NoneEd Santiago2018-11-27
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The conditional + list comprehension in images.py:_split_token() wasn't quite working as intended; in particular, when fed None, it chokes with TypeError: 'NoneType' object is not iterable This is the correct behavior: comprehensions iterate first, then apply the conditional. Solution: special-case None, and remove the now-unnecessary conditional. Context: seen when trying 'pypodman run' against docker.io/stackbrew/centos:7, which has no .ContainerConfig.Eng Signed-off-by: Ed Santiago <santiago@redhat.com>
* | | | | Merge pull request #1881 from jwhonce/bug/1867OpenShift Merge Robot2018-11-28
|\ \ \ \ \ | | | | | | | | | | | | Fixes #1867
| * | | | | Fixes #1867Jhon Honce2018-11-28
| | |_|/ / | |/| | | | | | | | | | | | | | | | | | | | | | | * Some items included in the CLI and currently not supported by the API. Signed-off-by: Jhon Honce <jhonce@redhat.com>
* | | | | Merge pull request #1880 from baude/f29fixesOpenShift Merge Robot2018-11-28
|\ \ \ \ \ | | | | | | | | | | | | Fix golang formatting issues
generated by cgit v1.2.3-54-g00ecf (git 2.39.0) at 2024-11-15 00:26:19 +0000