| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
| |
The mux containerPort must be an int
The eventrouter REPLICAS must use the double quote/double brace
syntax documented here:
https://docs.openshift.org/latest/dev_guide/templates.html#writing-parameters
And must be escaped because it uses double braces, as explained here:
http://jinja.pocoo.org/docs/dev/templates/#escaping
|
|
|
|
|
| |
- all images logging and metrics change their default imagePullPolicy
from Always to IfNotPresent
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
We now use a CPU request to ensure logging infrastructure pods are
not capped by default for CPU usage. It is still important to ensure
we have a minimum amount of CPU.
We keep the use of the variables *_cpu_limit so that the existing
behavior is maintained.
Note that we don't want to cap an infra pod's CPU usage by default,
since we want to be able to use the necessary resources to complete
it's tasks.
Bug 1501960 (https://bugzilla.redhat.com/show_bug.cgi?id=1501960)
|
|
|
|
|
|
|
|
| |
Includes the following code review fixes, and a rebase with upstream/master
- mux.j2 template needed mux specific parameters for syslog configs
- fixed bug in IF condition not checking for true-ness
- updated README wording to remove ambiguity of payload_key's datatype
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Instead of the `openshift_logging_use_mux_client` boolean parameter,
use `openshift_logging_mux_client_mode` which will allow us to support
different mux client use cases:
The value `maximal` will cause Fluentd to perform as much of the
processing as possible at the local node. This currently means all of
the processing *except* for the Kubernetes metadata processing, which will
be done by mux. This is the currently recommended mode to use due to
current scaling issues.
The value `minimal` means that Fluentd will do *no* processing at all,
and send the raw logs to mux for processing. This is currently not
recommended to use due to current scaling issues. Ansible will warn
you if you try to use this mode.
`MUX_ALLOW_EXTERNAL` is no longer needed in the mux dc. mux now always
operates to process external logs. The ansible setting
`openshift_logging_mux_allow_external` is still required in order to
set up the mux service to accept connections from outside of the
cluster.
|
|
|
|
| |
(cherry picked from commit 231a7c4ce27ff944d9e14169062f487f57c94dd6)
|
|
|
|
| |
creeping
|
|
|
|
|
|
|
|
|
|
| |
"openshift_logging_fluentd_use_journal=false" nor omitted collects the log entries
https://bugzilla.redhat.com/show_bug.cgi?id=1466152
Do not set openshift_logging_fluentd_use_journal or USE_JOURNAL at
all unless it is explicitly set as an ansible param. It is almost
always better to let fluentd figure out which log driver docker
is using.
|
|\
| |
| | |
Sync environment variables FLUENTD/MUX_CPU_LIMIT FLUENTD/MUX_MEMORY_L…
|
| |
| |
| |
| | |
with the resource limit values.
|
|/
|
|
|
|
| |
https://bugzilla.redhat.com/show_bug.cgi?id=1465168
The fix is to pass the value through the `lower` filter. The value is
set in defaults/main.yaml so it should always have a value.
|
|\
| |
| | |
Merged by openshift-bot
|
| |
| |
| |
| | |
MUX_CPU_LIMIT, MUX_MEMORY_LIMIT.
|
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
| | |
and memory_limit.
Default values:
openshift_logging_fluentd_cpu_limit: 100m
openshift_logging_fluentd_memory_limit: 512Mi
openshift_logging_fluentd_buffer_queue_limit: 1024
openshift_logging_fluentd_buffer_size_limit: 16m
openshift_logging_fluentd_buffer_size_limit: 1m
openshift_logging_mux_cpu_limit: 500m
openshift_logging_mux_memory_limit: 1Gi
openshift_logging_mux_buffer_queue_limit: 1024
openshift_logging_mux_buffer_size_limit: 1m
|
|/ |
|
|
|
|
|
|
|
|
|
|
|
|
| |
When we currently create the set of logging `DeploymentConfig`s, we
create them with zero desired replicas. This causes the deployment to
immediately succeed as there is no work to be done. This inhibits our
ability to use nice CLI UX features like `oc rollout status` to monitor
the logging stack deployments. Instead, we should can create the configs
with the correct number of replicas in the first place and stop using
`oc scale` to bring them up after the fact.
Signed-off-by: Steve Kuznetsov <skuznets@redhat.com>
|
|
|