At scale, it appears that we sometimes hit the 1000ms timeout to create
the PID file when a container is created or executed.
Increasing the value to 60s should help when running a lot of containers
in heavy-loaded environment.
Related #1495Fixes#1816
Signed-off-by: Emilien Macchi <emilien@redhat.com>
Engineers get testing status via their PR's but another round of testing
happens post-merge, without any direct feedback. Fix this in a small
way, by adding a dynamic status badge on the front-page. If this
turns red, it means Cirrus-CI testing of the master branch failed
for some reason.
Nearly always it's something harmless, but once and a while, this
catches really nasty problems caused by merge-sequence issues.
Having that feedback on the front page ensures the right people will
eventually get called into action.
Signed-off-by: Chris Evich <cevich@redhat.com>
Previously it was required to call the verify, unit, and integration
scripts in order to build/install dependencies, and libpod. This
wastes time during the (optional) system-testing, since the
actual unit/integration testing is also happening in parallel.
Consolidate only the distribution-specific build steps into the
system-testing script. This way, only the required steps are performed
in their respective (parallel) tasks.
Signed-off-by: Chris Evich <cevich@redhat.com>
Previously, several magic strings were in place to affect cirrus-ci
operations. Two were buried within scripts. One to optionally
execute system-tests within a PR. Another to avoid re-building
cache-images upon every merge.
Move these magic strings out into the open, buy locating their
logic up-front in the ``.cirrus.yml`` file. This improves
readability and reduces surprise/astonishment at runtime.
Signed-off-by: Chris Evich <cevich@redhat.com>
Having a standardized image allows uniform application of format and
lint checking across multiple host platforms. This ensures all
contributors and disparate CI systems to play by a common set of basic
rules. It also makes it easier to maintain the common rules over-time.
Signed-off-by: Chris Evich <cevich@redhat.com>
/etc/resolv.conf and /etc/hosts should not be created and mounted when the
network is disabled.
We should not be calling the network setup and cleanup functions when it is
disabled either.
In doing this patch, I found that all of the bind mounts were particular to
Linux along with the generate functions, so I moved them to
container_internal_linux.go
Since we are checking if we are using a network namespace, we need to check
after the network namespaces has been created in the spec.
Signed-off-by: Daniel J Walsh <dwalsh@redhat.com>
Currently we modify the git checkout which ends up showing that the checkout is dirty. This patch sets the PYTHON_VERSION so that python code will handle
it correctly without having to modify the actual code.
Signed-off-by: Daniel J Walsh <dwalsh@redhat.com>
We had a breaking change to rootless containers' temporary files
directories in Podman 0.11.1. Document a workaround to make the
change no longer working.
Signed-off-by: Matthew Heon <matthew.heon@gmail.com>
For the shell examples related to building and installing podman, remove the
shell prompt indicator character '$'.
This makes copying and pasting lines much easier.
Retain the prompt indicator for the post-install shell examples, since they
(often) mix shell commands and shell command output, so it serves a useful
purpose to distinguish them.
Signed-off-by: Jonathan Dowland <jdowland@redhat.com>
While we have these string literals on repeated on multiple places in the
library, I cannot see real tangible benefit extracting these to constants
considering following facts:
(1) while 'unknown' or 'host' are repeated, they are often times used in
different context and thus perhaps worth extra const per each use.
(2) while these string literals repeat, the library is full of string literals
with special meaning that should be made constants too
(3) readability would suffer
Signed-off-by: Šimon Lukašík <slukasik@redhat.com>
Redefining err by := operator within block makes this err variable block local.
Addressing lint:
libpod/oci.go:368:3⚠️ ineffectual assignment to err (ineffassign)
Signed-off-by: Šimon Lukašík <slukasik@redhat.com>
Not just nil.
Addressing:
pkg/varlinkapi/images.go:273:15⚠️ ineffectual assignment to err (ineffassign)
Signed-off-by: Šimon Lukašík <slukasik@redhat.com>
We fixated on old metalinter dependency in past based on experience of
metalinter being oftentimes broke and hence broking our build. See
762f508d9ca97cdbaee6053b663e98aee9cae081 in cri-o for more details.
Now, dated metalinter is messing up with my environment (like it is
panicing on containters/storage) so let's see if we can move to more
current version of metalinter.
Signed-off-by: Šimon Lukašík <slukasik@redhat.com>
Need to return an error pointing user in right direction if rootless podman
fails, because of no /etc/subuid or /etc/subgid files.
Also fix up man pages to better describe rootless podman.
Signed-off-by: Daniel J Walsh <dwalsh@redhat.com>
Add a rootless field to the info data (e.g., `podman info`) to indicate
if the executing user is root or not. In most cases, this can be
guessed but now it is clear and may aid in debugging, reporting and
understanding certain issues.
Signed-off-by: Valentin Rothberg <vrothberg@suse.com>
Matthew Heon
TomSweeneyRedHat
OpenShift Merge Robot
Emilien Macchi
Chris Evich
Giuseppe Scrivano
Daniel J Walsh
Qi Wang
Jonathan Dowland
Xavier Krantz
Šimon Lukašík
Valentin Rothberg