Skip to content

Ask to remove podman if detected during the install. - #11

Closed
lennysh wants to merge 1 commit into
intelligentwolf:masterfrom
lennysh:ask-to-remove-podman-if-detected
Closed

lennysh wants to merge 1 commit into
intelligentwolf:masterfrom
lennysh:ask-to-remove-podman-if-detected

Conversation

@lennysh

@lennysh lennysh commented May 8, 2026

Copy link
Copy Markdown

Ask to remove podman if detected during the install. Docker-ce cannot be installed on the host if podman is installed.

… be installed on the host if podman is installed.
intelligentwolf added a commit that referenced this pull request Aug 1, 2026
…ile there is still room to act, and the installer resolves podman conflicts

Security (wolf1, 2026-08-01): a security tool that is installed but cannot
do its job now surfaces in System Check instead of silently rotting.
- scan_broken_security_controls: failed units (chkrootkit, rkhunter, AIDE/
  dailyaidecheck, clamav, auditd, fail2ban), alert delivery impossible
  (MAILTO set but the tool's mail command missing — checked even when the
  unit is green; chkrootkit exited 127 nightly for weeks this way), and
  uninitialised baselines (chkrootkit log.expected, AIDE database,
  rkhunter --propupd). All config keys and transports verified against the
  Debian packaging, not guessed. Posture category, never Compromise.
- scan_tmp_binaries: package-owned + digest-clean files are structurally
  suppressed (dpkg -S/-V, rpm -qf/-V) so a chronyd-class false positive
  cannot happen; the executable-bit gate stays, so corosync's /dev/shm
  blackbox data files stay invisible to it.

/boot (wolf1): the >99% suppression alerted only after dpkg was already
broken mid-unpack. New predictive boot_partition analyzer sizes the floor
from the kernels actually on the partition (< 2 kernel sets free = High,
< 1 = Critical) and detects orphaned kernels — including the running
kernel whose package was removed while booted, which is why autoremove
never had a handle on the pile. Remediation spells out the grub.cfg-
truncation-safe cleanup order.

Installer:
- podman detected before a Docker CE install now prompts for removal
  (docker-ce packages conflict; the podman-docker shim also masquerades
  as /usr/bin/docker). Never removes unattended. (PR #11, lennysh)
- agent stops for upgrade are breadcrumbed in the supervisor log so
  'agent exited (code 0)' lines stop reading as unexplained flaps (klas).

UI: the Storage Array drive-health card now states WHY it is unavailable
(pre-25.8.0 node, expired session, transport error) with a retry button,
instead of silently omitting itself.

Co-Authored-By: CodeWolf <paul@wolf.uk.com>
Co-Authored-By: Wolf Software Systems Ltd <paul@wolf.uk.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants