Latest upstream stable for el9. All four NVRs are already carried by the SO
prod repo, so no repo change is needed -- a so-repo-sync refresh is enough.
Tested on a managersearch and a heavynode (OL 9.8, 3.4.0), upgrading from
29.2.1/2.2.1 both by hand and through the state itself:
- The 29.8.1 RPM ships a byte-identical docker.service, so the full ExecStart
override in files/iptables-disabled.conf still resolves correctly and the
hand-written DOCKER/DOCKER-ISOLATION/DOCKER-USER chains came back
byte-identical on both nodes across upgrade, restart and reboot.
- update_holds re-pinned the versionlock from the old NVRs to the new ones
without intervention, so soup's path needs no change.
- docker-py 7.1.0 still creates sobridge and soauth (forced by removing both);
bridges keep their configured kernel names rather than br-<hash>.
- 29.6 changed how dynamic port allocation treats
net.ipv4.ip_local_reserved_ports; Strelka's 57314 is both published and
reserved, and docker-proxy still owns it with no bind errors.
- docker ps --format json gained a HealthStatus key. Additive, so so-status,
so-log-check and so-docker-prune all still parse it.
- containerd 2.3.6 ships the same config.toml, and it is %config(noreplace)
and unmodified on disk, so no .rpmnew and disabled_plugins=["cri"] survives.
- Zeek/Suricata/Strelka pipeline verified end-to-end with so-test: 111k packets
replayed, 0 capture loss, file extraction and ES ingest all landed.
The manifest unknown exclusion in so-log-check still fires on 29.8.1 -- it
comes from a tag lookup during the registry-to-registry image copy, not from
the 29.2.1 upgrade the old comment blamed -- so only the comment changes.
Result checks walked dispatch records oldest-first with a cap of five
lookups per pass, counting records whose push was still running. Five
long-running pushes therefore used every slot on every 15s pass and newer,
finished pushes were not reported until one cleared. Check the least
recently checked records first and back off on pushes that are still
running (30s for the first two minutes, then age/4 up to 5 minutes),
recording checked_at in the dispatch record.
Catch any exception when writing a dispatch record so a failed write
cannot skip intent cleanup and re-dispatch the same intents every pass.
Log both output streams when no jid is found, and stop logging a traceback
when a record has already been removed.
Scope the test's salt mock to the drainer import. Run from the repo root,
'salt' resolves to this repo's salt/ directory as a namespace package, so
setdefault left it in place and test_load_push_cfg failed.
Verified on a 3.4.0 managersearch + sensor: a pushed highstate with soc
and telegraf pushes dispatched into it all reported success, with 25
result lookups across the three pushes instead of one per record per pass.
_orch_failures assumed every level of a jobs.lookup_jid result was a dict.
A list or string at the top level, in return.data, in a step's changes, or
in changes.ret raised AttributeError. Because result checks run before the
drain and a record is only removed after it is evaluated, one such record
would have failed every 15s pass and stopped all pushes until it was removed
by hand. Guard each shape, and evaluate each record under its own exception
handler so an unreadable result is logged and dropped instead of blocking
the drainer. Per-step parsing moves to _step_failures.
Search stdout as well as stderr for the async jid, in case salt-run logging
is routed to stdout.
Close the RotatingFileHandler in test_make_logger_adds_handler_once to
avoid a ResourceWarning on Python 3.12+.
Verified on a 3.4.0 standalone: real failed and successful orchestration
results parse as before, a record whose evaluation raises is logged and
removed while the next record still reports, and a replicated SOC change
to telegraf.output (and its revert) is pushed, rendered and logged as
succeeded.
Josh Brower
Mike Reeves
Matthew Wright
Matthew Wright
Jorge Reyes
Jason Ertel
Jason Ertel
Josh Patterson
coreyogburn
Corey Ogburn