Ignore error when starting transient unit that already exists#1124
Merged
crosbymichael merged 1 commit intoopencontainers:masterfrom Oct 19, 2016
Merged
Conversation
Contributor
Author
Contributor
|
LGTM; applying this patch worked for me |
Contributor
|
lgtm |
crosbymichael
requested changes
Oct 19, 2016
|
|
||
| if _, err := theConn.StartTransientUnit(unitName, "replace", properties, nil); err != nil { | ||
| return err | ||
| if dbusError, ok := err.(dbus.Error); ok { |
Member
There was a problem hiding this comment.
The logic is a little confusing. It would be better to have this as a function and reduce a few ifs.
if _, err := theConn.StartTransientUnit(unitName, "replace", properties, nil); err != nil && !isUnitExist(err) {
return err
}
Contributor
Author
There was a problem hiding this comment.
@crosbymichael that works for me, will update.
Signed-off-by: Derek Carr <decarr@redhat.com>
derekwaynecarr
force-pushed
the
apply-idempotent
branch
from
a83acf7 to
d223e2a
Compare
Contributor
Author
|
@crosbymichael - updated per request. |
Contributor
|
LGTM |
This was referenced Apr 25, 2017
kolyshkin
added a commit
to kolyshkin/runc
that referenced
this pull request
Mar 23, 2023
Commit d223e2a ("Ignore error when starting transient unit that already exists" modified the code handling errors from startUnit to ignore UnitExists error. Apparently it was done so that kubelet can create the same pod slice over and over without hitting an error (see [1]). While it works for a pod slice to ensure it exists, it is a gross bug to ignore UnitExists when creating a container. In this case, the container init PID won't be added to the systemd unit (and to the required cgroup), and as a result the container will successfully run in a current user cgroup, without any cgroup limits applied. So, fix the code to only ignore UnitExists if we're not adding a process to the systemd unit. This way, kubelet will keep working as is, but runc will refuse to create containers which are not placed into a requested cgroup. [1] opencontainers#1124 Signed-off-by: Kir Kolyshkin <kolyshkin@gmail.com>
kolyshkin
added a commit
to kolyshkin/runc
that referenced
this pull request
Mar 31, 2023
Commit d223e2a ("Ignore error when starting transient unit that already exists" modified the code handling errors from startUnit to ignore UnitExists error. Apparently it was done so that kubelet can create the same pod slice over and over without hitting an error (see [1]). While it works for a pod slice to ensure it exists, it is a gross bug to ignore UnitExists when creating a container. In this case, the container init PID won't be added to the systemd unit (and to the required cgroup), and as a result the container will successfully run in a current user cgroup, without any cgroup limits applied. So, fix the code to only ignore UnitExists if we're not adding a process to the systemd unit. This way, kubelet will keep working as is, but runc will refuse to create containers which are not placed into a requested cgroup. [1] opencontainers#1124 Signed-off-by: Kir Kolyshkin <kolyshkin@gmail.com>
kolyshkin
added a commit
to kolyshkin/runc
that referenced
this pull request
Apr 3, 2023
Commit d223e2a ("Ignore error when starting transient unit that already exists" modified the code handling errors from startUnit to ignore UnitExists error. Apparently it was done so that kubelet can create the same pod slice over and over without hitting an error (see [1]). While it works for a pod slice to ensure it exists, it is a gross bug to ignore UnitExists when creating a container. In this case, the container init PID won't be added to the systemd unit (and to the required cgroup), and as a result the container will successfully run in a current user cgroup, without any cgroup limits applied. So, fix the code to only ignore UnitExists if we're not adding a process to the systemd unit. This way, kubelet will keep working as is, but runc will refuse to create containers which are not placed into a requested cgroup. [1] opencontainers#1124 Signed-off-by: Kir Kolyshkin <kolyshkin@gmail.com> (cherry picked from commit c253342) Signed-off-by: Kir Kolyshkin <kolyshkin@gmail.com>
jiusanzhou
pushed a commit
to jiusanzhou/runc
that referenced
this pull request
Apr 6, 2023
Commit d223e2a ("Ignore error when starting transient unit that already exists" modified the code handling errors from startUnit to ignore UnitExists error. Apparently it was done so that kubelet can create the same pod slice over and over without hitting an error (see [1]). While it works for a pod slice to ensure it exists, it is a gross bug to ignore UnitExists when creating a container. In this case, the container init PID won't be added to the systemd unit (and to the required cgroup), and as a result the container will successfully run in a current user cgroup, without any cgroup limits applied. So, fix the code to only ignore UnitExists if we're not adding a process to the systemd unit. This way, kubelet will keep working as is, but runc will refuse to create containers which are not placed into a requested cgroup. [1] opencontainers#1124 Signed-off-by: Kir Kolyshkin <kolyshkin@gmail.com>
dims
pushed a commit
to dims/libcontainer
that referenced
this pull request
Oct 19, 2024
Commit 94efc45 ("Ignore error when starting transient unit that already exists" modified the code handling errors from startUnit to ignore UnitExists error. Apparently it was done so that kubelet can create the same pod slice over and over without hitting an error (see [1]). While it works for a pod slice to ensure it exists, it is a gross bug to ignore UnitExists when creating a container. In this case, the container init PID won't be added to the systemd unit (and to the required cgroup), and as a result the container will successfully run in a current user cgroup, without any cgroup limits applied. So, fix the code to only ignore UnitExists if we're not adding a process to the systemd unit. This way, kubelet will keep working as is, but runc will refuse to create containers which are not placed into a requested cgroup. [1] opencontainers/runc#1124 Signed-off-by: Kir Kolyshkin <kolyshkin@gmail.com>
dims
pushed a commit
to dims/libcontainer
that referenced
this pull request
Oct 19, 2024
Commit 33d484a ("Ignore error when starting transient unit that already exists" modified the code handling errors from startUnit to ignore UnitExists error. Apparently it was done so that kubelet can create the same pod slice over and over without hitting an error (see [1]). While it works for a pod slice to ensure it exists, it is a gross bug to ignore UnitExists when creating a container. In this case, the container init PID won't be added to the systemd unit (and to the required cgroup), and as a result the container will successfully run in a current user cgroup, without any cgroup limits applied. So, fix the code to only ignore UnitExists if we're not adding a process to the systemd unit. This way, kubelet will keep working as is, but runc will refuse to create containers which are not placed into a requested cgroup. [1] opencontainers/runc#1124 Signed-off-by: Kir Kolyshkin <kolyshkin@gmail.com> (cherry picked from commit c253342) Signed-off-by: Kir Kolyshkin <kolyshkin@gmail.com>
dims
pushed a commit
to dims/libcontainer
that referenced
this pull request
Oct 19, 2024
Commit db41179 ("Ignore error when starting transient unit that already exists" modified the code handling errors from startUnit to ignore UnitExists error. Apparently it was done so that kubelet can create the same pod slice over and over without hitting an error (see [1]). While it works for a pod slice to ensure it exists, it is a gross bug to ignore UnitExists when creating a container. In this case, the container init PID won't be added to the systemd unit (and to the required cgroup), and as a result the container will successfully run in a current user cgroup, without any cgroup limits applied. So, fix the code to only ignore UnitExists if we're not adding a process to the systemd unit. This way, kubelet will keep working as is, but runc will refuse to create containers which are not placed into a requested cgroup. [1] opencontainers/runc#1124 Signed-off-by: Kir Kolyshkin <kolyshkin@gmail.com> (cherry picked from commit ada2c18260279870cd0b1709507e25ac8cd22d03) Signed-off-by: Kir Kolyshkin <kolyshkin@gmail.com>
kolyshkin
added a commit
to kolyshkin/containerd-cgroups
that referenced
this pull request
Nov 6, 2024
Commit d4c949b ("Ignore error when starting transient unit that already exists" modified the code handling errors from startUnit to ignore UnitExists error. Apparently it was done so that kubelet can create the same pod slice over and over without hitting an error (see [1]). While it works for a pod slice to ensure it exists, it is a gross bug to ignore UnitExists when creating a container. In this case, the container init PID won't be added to the systemd unit (and to the required cgroup), and as a result the container will successfully run in a current user cgroup, without any cgroup limits applied. So, fix the code to only ignore UnitExists if we're not adding a process to the systemd unit. This way, kubelet will keep working as is, but runc will refuse to create containers which are not placed into a requested cgroup. [1] opencontainers/runc#1124 Signed-off-by: Kir Kolyshkin <kolyshkin@gmail.com>
kolyshkin
added a commit
to kolyshkin/containerd-cgroups
that referenced
this pull request
Feb 14, 2025
Commit d4c949b ("Ignore error when starting transient unit that already exists" modified the code handling errors from startUnit to ignore UnitExists error. Apparently it was done so that kubelet can create the same pod slice over and over without hitting an error (see [1]). While it works for a pod slice to ensure it exists, it is a gross bug to ignore UnitExists when creating a container. In this case, the container init PID won't be added to the systemd unit (and to the required cgroup), and as a result the container will successfully run in a current user cgroup, without any cgroup limits applied. So, fix the code to only ignore UnitExists if we're not adding a process to the systemd unit. This way, kubelet will keep working as is, but runc will refuse to create containers which are not placed into a requested cgroup. [1] opencontainers/runc#1124 Signed-off-by: Kir Kolyshkin <kolyshkin@gmail.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
6 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
If a unit already exists with the specified name, ignore the error.
This allows the call to be idempotent when doing slice management on systemd 231+.