/etc/collect.d/collect_containerization contains stale etcd conf file location

[Kaustubh Dhokte]

The /etc/collect.d/collect_containerization script looks at /etc/etcd/etcd.conf, when in actual fact we don't use this config file. Instead, the ETCD_LISTEN_CLIENT_URLS variable is passed to etcd via an environment variable. Ultimately it comes from /etc/default/etcd:

root@controller-0:/var/home/sysadmin# more /etc/default/etcd

Managed by Puppet
#[member]
ETCD_NAME="controller"
ETCD_DATA_DIR="/opt/etcd/24.09/controller.etcd"
ETCD_SNAPSHOT_COUNT=10000
ETCD_HEARTBEAT_INTERVAL=100
ETCD_ELECTION_TIMEOUT=1000
ETCD_QUOTA_BACKEND_BYTES=0
ETCD_LISTEN_CLIENT_URLS="https://192.168.205.1:2379,https://[127.0.0.1]:2379"

This will result in collect_containerization not properly saving the raw etcd datastore, which will make it harder to debug certain field issues.

Severity

Minor

Steps to Reproduce

Do a standard "collect"

Expected Behavior

Raw etcd info should be stored.

Actual Behavior

Raw etcd info not stored.

Reproducibility

Reproducible

System Configuration

all

Load info (eg: 2022-03-10_20-00-07)

All debian, I suspect.

Last Pass

Suspect it dates back to the switch to debian.

Timestamp/Logs

n/a

Alarms

n/a

Test Activity

debugging

Workaround

none

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.opendev.org/c/starlingx/utilities/+/923455

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.opendev.org/c/starlingx/utilities/+/923455
Committed: https://opendev.org/starlingx/utilities/commit/4ae191f23d997b383103b628bbcb1a5b36e6ffb2
Submitter: "Zuul (22348)"
Branch: master

commit 4ae191f23d997b383103b628bbcb1a5b36e6ffb2
Author: Kaustubh Dhokte <email address hidden>
Date: Thu Jul 4 00:12:51 2024 +0000

    Collect_containerization: correct etcd conf file path

    This change corrects etcd config file path in
    /etc/collect.d/collect_containerization.

    The file still uses etcd config file path used on centos i.e.
    /etc/etcd/etcd.conf. The same on Debian (new StarlingX base OS)
    is /etc/default/etcd. With this correction, now collectd
    can correctly dump etcd database.

    Test Plan:
    AIO-SX:
    PASS: Run 'collect --all', etcd database dumped correctly under
          var/extra.

    Closes-Bug: 2071866

    Change-Id: I85be6cdec4f050f1f34df21d772af12305f09f15
    Signed-off-by: Kaustubh Dhokte <email address hidden>