sesdev-1.3.0+1590413709.g4ad4e03-bp152.1.1

- Update to 1.3.0+1590413709.g4ad4e03:
  + upstream 1.3.0 release (2020-05-25)
  * octopus/ses7: added "--stop-before-ceph-orch-apply" function (PR #301)
  * Implement RGW deployment in octopus, ses7 (PR #314)
  * ceph_salt_deployment: do not force user to change dashboard pw (PR #315)
  * makecheck: possibly prophylactically downgrade libudev1 (PR #317)
  * contrib/standalone.sh: --no-stop-on-failure option (PR #318)
  * ceph_salt_deployment: make use of 'cephadm' role (PR #319)
  * octopus/ses7: removed "--deploy-mons", "--deploy-mgrs", "--deploy-osds",
    "--deploy-mdss" (replaced by "--stop-before-ceph-orch-apply") (PR #301)
  * seslib: drop Containers module from SES7 deployment (PR #303)
  * provision.sh: remove curl RPM from the environment (PR #311)
  * Fixed "sesdev create caasp4" default deployment by disabling multi-master (PR #302)
  * ceph_salt_deployment: do not deploy MDS if no mds roles present (PR #313)
  * caasp: do not install salt (PR #320)
  * supportconfig: handle both scc and nts tarball prefixes (PR
  * seslib: convert certain public methods into private (PR #309)
  * caasp4: rename "storage" role to "nfs" and drop it from default 4-node
    deployment (PR #310)

- Update to 1.2.0+1588616857.gaa3df4c:
  + upstream 1.2.0 release (2020-05-04)
  * deepsea_deployment: pre-create Stage 4 pools (PR #298)
  * setup.cfg: do not break Tumbleweed RPM install (PR #297)
  * ceph_salt_deployment.sh: rip out time sync code (PR #289)
  * Rename '--stop-before-ceph-salt-deploy' to '--stop-before-ceph-salt-apply'
    (PR #290)

- Update to 1.1.12+1588084576.g8b2afa1:
  + upstream 1.1.12 release (2020-04-28)
  * provision.sh: enable autorefresh on all repos (PR #288)
  * seslib: fix "Unused argument" linter warning (PR #286)
  * provision.sh.j2: Properly prepare CaaSP nodes (PR #283)
  * ceph_salt_deployment: really sync clocks (PR #285)
  * Rename 'ceph-salt deploy' to 'ceph-salt apply' (PR #280)

- Update to 1.1.11+1587722868.gc7e5625:
  + upstream 1.1.11 release (2020-04-24)
  * Implement feature "sesdev create makecheck" (PR #236)
  * Deployment: silently add "master" and "bootstrap" roles (PR #254)
  * sesdev: implement "sesdev list --format json" and PrettyTable-based
    "sesdev list" (PR #259)
  * qa: fix maybe_wait_for_mdss test (PR #262)
  * provision.sh: remove Non-OSS repos on openSUSE (PR #265)
  * sesdev: support Click 6.7 (PR #268)
  * seslib: more robust dashboard tunnelling (PR #274)
  * deepsea: "openattic" role needs Stage 4 (PR #275)
  * provision.sh: refactor while loops and fix packaging-related breakage
    (PR #279)
  * Remove "disable cephadm bootstrap" functionality (PR #261)
  * Rename --deepsea-cli and --ceph-salt-deploy options to --salt
    (PR #267)

- Update to 1.1.10+1587124098.ga2f8189:
  + upstream 1.1.10 release (2020-04-17)
  * ceph_salt_deployment: sync clocks after deployment (PR #187)
  * ceph_salt_deployment: fetch GitHub PRs when needed (PR #197)
  * New deployment type "pacific" using packages/containers built from
    upstream Ceph "master" branch (PR #200)
  * README.md: Needed packages for all major distros (PR #209)
  * sesdev: systemically vet roles on create (PR #215)
  * templates: extend --qa-test to DeepSea-deployed versions (PR #230)
  * sesdev: make deployment_id argument optional (PR#235)
  * sesdev: Globbing for "stop", "start", and "destroy" (PR #238)
  * new --synced-folder option to NFS mount directories (PR #247)
  * ses7, octopus, pacific: Deploy MDSs (PR #258)
  * deepsea_deployment: run Stage 4 only if justified by roles (PR #205)
  * seslib: install sesdev-generated keypair under non-default name (PR #207)
  * Jenkinsfile.integration: Retry jenkins slave deletion (PR #214)
  * ceph_salt_deployment: use --prefix /usr with "pip install" (PR #221)
  * deepsea_deployment: check if drive_groups.yml exists (PR #229)
  * seslib: force ses5 prometheus node to master (PR #232)
  * provision.sh: reinstall certain packages from the update repos (PR #243)
  * ceph_salt_deployment: do not provision client-only nodes (PR #246)
  * ceph_salt_deployment: move OSD deployment to sesdev (PR #186)
  * ceph_salt_deployment: use sesdev to deploy MONs and MGRs (PR #189)
  * sesdev: uniform --force/--non-interactive (PR #201)
  * ceph_salt_deployment: make it easier to install ceph-salt from source (PR #210)
  * seslib: make all str-type settings default to the empty string (PR #217)
  * ceph_salt_deployment: Use quoted string to set bootstrap configs (PR #224)
  * ceph_salt_deployment: use lower-case on config nodes (PR #226)
  * ceph_salt_deployment: do not refresh/sync pillar data (PR #227)
  * templates: move test.ping try-wait into a separate file (PR #234)

- Update to 1.1.9+1585227448.g3a7e36e:
  + upstream 1.1.9 release (2020-03-26)
  * config.yaml: enable setting of repo priority in version_os_repo_mapping (PR #163)
  * provision.sh: add SUSE:CA repo on ses5 (PR #166)
  * Implement feature: "sesdev ssh DEP_ID NODE_ID COMMAND" (PR #175)
  * Implement feature: "sesdev supportconfig DEPLOYMENT_ID NODE_ID" (PR #176)
  * Implement feature: --encrypted-osds (PR #192)
  * provision.sh: do not fail ses5 deployment if ntp not installed (PR #173)
  * ceph_salt_deployment.sh: adapt Drive Group string to new syntax (PR #178)
  * qa/health-ok.sh: wait for OSD nodes to show up (PR #180)
  * Vagrantfile,sesdev.spec: require vagrant > 2.2.2 (PR #167)
  * Use "filesystems:ceph:octopus:upstream" for default cephadm/container build (PR #170)
  * Set "osd crush chooseleaf type = 0" via bootstrap ceph.conf in very small clusters (PR #183)
  * ceph_salt_deployment.sh: Fetch github PRs when installing from src (PR #190)

- Update to 1.1.8+1584097308.ga30c06e:
  + upstream 1.1.8 release (2020-03-13)
  * seslib: remove host's virtual networks on destroy (PR #102)
  * Jenkinsfile.integration for PR testing (PR #118, PR #154)
  * prometheus and alertmanager tunnels (PR #148)
  * tests: enable unit testing via tox (PR #151)
  * seslib: stop printing misleading device names (PR #150)
  * seslib: Set admin roles for octopus (PR #158)
  * Change "admin" role semantics and make roles configurable (PR #161)
  * Allow user to override parts of OS_REPOS, VERSION_OS_REPO_MAPPING, and
    IMAGE_PATHS (PR #146)

- Update to 1.1.7+1583767117.gc963107:
  + upstream 1.1.7 release (2020-03-09)
  * all PRs mentioned in this changelog since 1.1.6, plus
  * spec: always install sesdev-qa RPM along with sesdev (PR #144)

- Update to 1.1.7+1583767117.gc963107:
  + upstream 1.1.7 release (2020-03-09)
  * all PRs mentioned in this changelog since 1.1.6, plus
  * spec: always install sesdev-qa RPM along with sesdev (PR #144)

- Update to 1.1.6+1583488232.gfdcc651:
  + qa: fix path to qa scripts for RPM case (PR #141)
  + ceph_salt_deployment: run "ceph-salt status" after "ceph-salt config ls"
    (PR #138)

- Update to 1.1.6+1583428626.ga138314:
  + upstream 1.1.6 release (2020-03-05)
  * sesdev: add --non-interactive option to "sesdev create" (PR #125)
  * Implement "sesdev qa-test" command (PR #129)
  * qa: assert "ceph versions" matches "ceph --version" (PR #131)
  * seslib: correct downstream container for "sesdev create {ses7,octopus}" (PR #130)
  * provision.sh: remove Python 2 so it doesn't pollute the environment (PR #133)
  * Rename --ceph-container-image to --image-path (PR #115)
  * provision: install "command-not-found", "supportutils", etc. in test environments (PR #123)
  * Return with a non-zero exit code in a failure case (PR #127)
  * seslib: rename --deploy-bootstrap to --cephadm-bootstrap (PR #137)

- Update to 1.1.5+1582717868.g68df753:
  + upstream 1.1.5 release (2020-02-26)
  * sesdev.spec: use standard ordering of sections
  * sesdev: give the user a way to specify --no-deploy-... (PR #120)
  * seslib: fix --no-deploy-mgrs option not working (PR #122)

- Update to 1.1.4+1582707984.gdb87191
  + upstream 1.1.4 release (2020-02-26)
  * sesdev.spec: properly package /usr/share/sesdev directory
    (follow-on fix for PR #112)

- Update to 1.1.3+1582639489.g0e91afa:
  + upstream 1.1.3 release (2020-02-25)
  * Rename ceph-bootstrap to ceph-salt (PR#114)
  * Migrate ceph-bootstrap-qa to sesdev (part 2) (PR#112)
  * provision: remove which RPM from test environment (PR#113)
  * ceph_salt_deployment: disable system update and reboot (PR#117)
  * seslib: by default, a mgr for every mon (PR#111)

- Update to 1.1.2+1581962442.g190d64e:
  + upstream 1.1.2 release (2020-02-17)
  * Implement "vagrant box list" and "vagrant box remove" (PR#69)
  * Allow user to specify custom private key file for remote libvirt (PR#71)
  * spec: add Fedora-specific Requires (PR#77)
  * Pillar is now automatically configured by ceph-bootstrap (PR#78)
  * Implement "sesdev scp" feature (PR#101)
  * Implement "sesdev create caasp4" feature (PR#103)
  * Miscellaneous spec file cleanups and bugfixes (PR#72)
  * several fixes for octopus/ses7 deployment (PR#76)
  * Remove any orphaned images after destroy (PR#81)
  * seslib: fix Ceph repos for ses5, ses6, ses7 (PR#83)
  * tools/run_async: decode stderr bytes (PR#88)
  * libvirt/network: autostart networks per default (PR#93)
  * Fix NTP issue that was causing SES5 deployment to fail (PR#108)
  * Increase chances of getting the latest ses7 packages (PR#84)
  * ceph_bootstrap_deployment: log cephadm and ceph-bootstrap version (PR#86)
  * ceph_bootstrap: restart salt-master after ceph-bootstrap installation (PR#87)
  * seslib: add SES7 Internal Media when --qa-test given (PR#90)

- Update to 1.1.1+1580550501.gd6782ab:
  + changes from upstream SHA1 d6782abc3002a5a8ec299b1897ad6166a786bafa:
  * Revamp --num-disks handling (PR #65)
  * doc: globally change 'rjfd' to 'SUSE' (PR #66)
  * ceph_bootstrap_deployment: "ceph-bootstrap -ldebug deploy" (PR #68)
  * checkin.sh: use NoSource
  * spec:
    = drop Group
    = streamline %autosetup
    = fix Fedora conditional

- Update to 1.1.1+1580302614.g0e79816:
  + upstream 1.1.1 release (2020-01-29)
  * Octopus and SES7 deployment with ceph-bootstrap (PR #28)
  * Implement --repo-priority / --no-repo-priority (PR #19)
  * Add option for predefined libvirt networks (PR #39)
  * qa: initial qa integration (PR #46)
  * When deploying ses5 with explicit --roles, do not add openattic (PR #23)
  * Enable display of manpages (PR #33)
  * ceph_bootstrap_deployment.sh: Also set -e (PR #35)
  * vagrant: generate random serial number for each attached disk (PR #50)
  * ceph_bootstrap_deployment: ensure minions are responding (PR #52)
  * Add node info to SSH tunnel command (PR #27)
  * Display amount of deployed VMs in status output (PR #31)
  * Enable linter via travis (PR #38)
  * ceph-bootstrap: 'ceph-salt-formula' moved to 'ceph-bootstrap' (PR #44)
  * ceph_bootstrap: use "ceph-bootstrap deploy" command to run ceph-salt formula (PR #62)

- Update to 1.1.0+1574848386.g395e92f:
  + Add Github Action to publish to OBS (PR #10).
  + SUMA deployment in `octopus` version (PR #14).
  + Handle Ctrl+C on deployment creation (PR #8).
  + seslib: fixed 100% cpu usage when deploying cluster (PR #16).
  + Updated README.md on how to use an editable Python venv (PR #15).
  + cli: `list` subcommand now shows the version of each deployment (PR #12).
  + octopus and ses7 versions now use the OBS repo
    filesystems:ceph:master:upstream that is updated in a daily basis (PR #13).

- Update to 1.0.3+1573815567.g0a8bfe0:
  + README instructions about libvirt configuration.
  + CLI subcommand `info` replaced by `show`.
  + fix typo in `sesdev start --help` command.

-Update to 1.0.2+1573810156.g0c0dc21:
  + fix missing libvirt engine jinja template in setup.py.

- Update to 1.12.1+1619444076.g04475be:
  + upstream 1.12.1 release (2021-04-26)
    + deployment: Remove redundant node network code (PR #592)
    + contrib/nukevolz.sh: primitive script to recover wasted disk space (PR #593)
    + zypper: remove python3-base if it's installed (PR #594)
    + templates: disallow global ID reclaim (PR #596)

- Update to 1.12.0+1616368506.g9bba186:
  + upstream 1.12.0 release (2021-03-22)
  * provision: enable coredumps (PR #566)
  * sesdev: support custom repo URLs pointing to .repo file (PR #571)
  * cephadm: add 'period update' in radosgw deployment (PR #579)
  * tox: add support for Python 3.8 (PR #582)
  * sesdev.spec: fix missing qa/common/rgw.sh (PR #572)
  * contrib/upgrade-demo-ses6-to-ses7.sh: fix (PR #573)
  * zypper: be more careful when installing systemd-coredump (PR #577)
  * sesdev: replace fancy list comprehension with simpleton code (PR #581)
  * deployment: saner processing of --deepsea-repo, --deepsea-branch (PR #584)
  * zypper.j2: remove Python 2 even on SLE-12-SP3 (PR #585)
  * provision.sh: persist systemd journal in SES7 only (PR #569)
  * ceph-salt: no cephadm for nodes with no sesdev roles (PR #570)
  * constant.py: drop nfs from octopus default roles (PR #587)
  * nautilus_pre_stage_4.sh.j2: adapt to syntax change in Nautilus 14.2.17 (PR #589)
  * qa/common/json.sh: adapt to change in ceph status json format (PR #590)

- Update to 1.11.0+1607012263.g73dea97:
  + upstream 1.11.0 release (2020-12-03)
  * Implement --rgw-ssl option for deploying RGWs with SSL enabled (PR #557)
  * zypper: add repos when needed (PR #558)
  * ses7: use product container image when --product given (PR #561)
  * Handle image_paths_product, image_paths_devel overrides in config.yaml (PR #564)
  * seslib/constant.py: add SES7 Update repo (PR #565)

- Update to 1.10.0+1605283435.g8e610fb:
  + upstream 1.10.0 release (2020-11-13)
  * contrib/standalone.sh: add --fqdn test cases (PR #523)
  * Implement --no-apparmor option (PR #543)
  * Expose Ceph Dashboard on CaaSP SES deployment (PR #553)
  * doc/README.md: add common pitfall "Failed to intialize libnetcontrol" (PR #554)
  * spec: pin sesdev-qa version to sesdev version (PR #551)
  * constant,zypper: use download.nue.suse.com instead of download.suse.de (PR #552)
  * deployment_day_2.sh: create CephFS using documented procedure (PR #555)

- Update to 1.9.1+1603459604.g8c9ed2c
  + upstream 1.9.1 release (2020-10-23)
  * sesdev --devel flag: let click deduct type (PR #550)

- Update to 1.9.0+1603448566.gcb3362c
  + upstream 1.9.0 release (2020-10-23)
  * constant: add missing SLE-15-SP[12] Product repos (PR #530)
  * Implement "sesdev upgrade" subcommand (PR #531)
  * Implement "sesdev reboot DEP_ID NODE" command (PR #532)
  * contrib/upgrade-demo-ses6-to-ses7.sh: add upgrade demo script (PR #535)
  * provision.sh: add helper_scripts Jinja template (PR #540)
  * exceptions: have SesDevException return optional exit code (PR #538)
  * deployment: wait for rebooted node to complete boot sequence (PR #539)
  * sesdev: fix --filestore regression (PR #545)
  * cephadm/deployment_day_2.sh.j2: do not fail on defective --dry-run (PR #546)
  * setup.cfg: include seslib/templates/cephadm/ directory in packaging (PR #548)
  * templates/zypper: do not install supportutils-plugin-ses in SES{5,6,7} (PR #526)
  * deepsea/nautilus_pre_stage_0.sh: refrain from patching DeepSea (PR #547)

- Update to 1.8.0+1602174240.g45df4dd
  + upstream 1.8.0 release (2020-10-08)
  * Add support for IPv6 (PR #56)
  * contrib/deepsea_drive_replace.sh: new test script (PR #306)
  * ceph_salt: deploy and smoke-test Prometheus (PR #418)
  * ceph-salt: deploy and smoke-test Grafana (PR #475)
  * ceph-salt: deploy and smoke-test alertmanager (PR #480)
  * ceph-salt: deploy and smoke-test node-exporter (PR #496)
  * Support official Vagrant boxes from Vagrant Cloud (PR #476)
  * new msgr2 secure options (PR #481)
  * Add ability to deploy Ceph on Ubuntu "Bionic Beaver" 18.04 (PR #486)
  * add label for ses deployment on caasp (PR #488)
  * Implement --no-provision option to deploy "bare" VMs (PR #490)
  * Implement "sesdev link" for linking two clusters (PR #498)
  * Implement "sesdev status" subcommand (PR #502)
  * Implement "sesdev show DEP_ID --nodes-with-role=ROLE" (PR #506)
  * Make "sesdev box list" and "sesdev box remove" accept globs (PR #517)
  * create: implement --fqdn option, specify FQDN when appropriate (PR #519, PR #522)
  * replace --deploy/--no-deploy option with --provision/--no-provision (PR #490)
  * caasp4: remove anti-affinity for mons for single node (PR #495)
  * Revert "ceph_salt_deployment: add 1-minute grace period" (PR #521)
  * ceph-salt: set some network-related config params explicitly (PR #482)
  * deployment: do not assert if starting an existing deployment (PR #493)
  * deployment: do not populate public_network with "0/24" (PR #512)
  * qa: wait for grafanas to show up before counting them (PR #514)
  * qa: wait longer for cluster to become healthy (PR #515)
  * Add makecheck-specific repos only in "sesdev create makecheck" (PR #520)
  * zypper: install rbd-nbd on all Leap 15.2/SLE-15-SP2 nodes (PR #484)
  * Update pacific repos (PR #489)
  * qa/nfs: do not attempt to mount NFS export on "pacific" (PR #492)
  * deployment: use "vagrant up --provision" to start cluster (PR #501)
  * cleanup: move cephadm Day 2 provisioner to its own subdirectory (PR #513)
  * tox: run pip with --use-feature=2020-resolver (PR #516)

- Update to 1.7.0+1599999069.g8f2e732
  + upstream 1.7.0 release (2020-09-13)
  * ceph_salt_deployment: do "ceph orch apply --dry-run" (PR #406)
  * single node CaaSP cluster (PR #427)
  * If --ssd option is given, attempt to make first additional disk
    non-rotational (PR #448)
  * Implement "sesdev box remove --all" feature (PR #451)
  * qa: support openSUSE Tumbleweed (PR #457)
  * Drop ceph-salt '/system_update' config (PR #447)
  * replace-mgr-modules: refrain from introducing Python 2 (PR #431)
  * deepsea: add ganesha roles to policy.cfg (PR #437)
  * deepsea: properly recognize deprecated ganesha role (PR #442)
  * ceph_salt_deployment: prevent Vagrant 2.2.10+ from deleting master
    node (PR #452)
  * provision.sh: disable host checking when SSHing within the cluster
    (PR #455)
  * ceph_salt_deployment: explicitly create mds service (PR #458)
  * contrib/standalone.sh: adapt to create always returning 0 (PR #460)
  * deepsea_deployment: really stop before Stage 0 (PR #462)
  * ceph-salt: Fix ceph image path config (PR #465)
  * ceph_salt_deployment: always wait for OSDs to appear (PR #466)
  * ceph_salt_deployment: deploy MDSs according to documentation
    (PR #468)
  * ceph_salt_deployment: extend OSD deployment timeout (PR #432)
  * caasp: set num_disks in same way as for ceph deployments (PR #434)
  * vet_configuration: vet caasp4 roles more carefully (PR #435)
  * cephadm: allow preparation of cluster for manual deployment (PR #436)
  * caasp: update to caasp 4.5 (PR #438)
  * deployment: disks even when explicit storage role not given (PR #439)
  * cleanup: move code out of Deployment/_generate_nodes() (PR #441)
  * ceph-salt: enable user to control which nodes get "admin" role
    (PR #443)
  * cleanup: streamline unit testing and code linting (PR #444)
  * ceph_salt_deployment: tolerate additional bootstrap MONs/MGRs that
    we didn't ask for (PR #449)
  * ceph_salt_deployment: expose ceph-salt errors early (PR #456)
  * ceph-salt: bootstrap minion no longer required to have admin role
    (PR #461)
  * Split ceph_salt_deployment.sh into "Day 1" and "Day 2" scripts
    (PR #471)

- Update to 1.6.1+1597147109.g204bf79
  + upstream 1.6.1 release (2020-08-11)
  * ceph_salt_deployment: implement "--stop-before-ceph-orch-apply" (PR #415)
  * settings: rename straggler version_os_repo_mappings (PR #419)
  * setup.cfg: fix "options.package_data" file list (PR #424)
    (fixes a packaging regression introduced in 1.6.0)
  * sesdev: non-create, non-box functions in alphabetical order (PR #414)
  * Unify naming of role-related constants (PR #416)
  * CaaSP: remove hard dependency on loadbalancer (PR #422)

- Update to 1.6.0+1595760860.gc353ea1
  + upstream 1.6.0 release (2020-07-26)
  * sesdev: implement --dry-run for create commands (PR #384)
  * ceph_salt_deployment: use mgr/nfs CLI to deploy NFS Ganesha (PR #385)
  * 'sesdev replace-mgr-modules' should also replace 'bin/cephadm' (PR #387)
  * qa: enable dashboard branding test on {octopus,ses7} (PR #392)
  * zypper: remove rsync if it's installed in the Vagrant Box (PR #395)
  * ceph_salt_deployment: "ceph -s" when OSDs fail to come up (PR #409)
  * deployment: run supportconfig with 1-hour timeout (PR #410)
  * provision.sh: persist the journal (PR #411)
  * ses5: stop patching srv/salt/ceph/time/ntp/default.sls (PR #407)
  * split seslib/__init__.py into several smaller files (PR #377, #380)
  * provision.sh: add all repos before installing packages (PR #381)
  * Refactor seslib/templates/provision.sh (PR #383)
  * deployment/status: aggregate global parameters (PR #393)
  * Improve the "show" subcommand and deployment configuration listing (PR #399)
  * Change --repo-priority default from "True" to "False" (PR #400)
  * seslib: revamp custom_repo (PR #402)
  * seslib/deployment: try to destroy the whole cluster at once (PR #403)
  * log: print log messages to the screen under certain circumstances (PR #404)
  * Rename "version_os_repo_mapping" to "version_devel_repos" (PR #408)
  * ceph_salt_deployment.sh: extend OSD timeout (PR #413)
  * qa/common/rgw: fix curl try-wait (PR #379)
  * sesdev: use Luminous roles for SES5 (PR #382)
  * ceph_salt_deployment: fix off-by-one error in OSDs wait loop (PR #394)
  * sesdev: fix --non-interactive/--force handling (PR #395)
  * deployment: generate static networks on create only (PR #401)

- Update to 1.5.0+1593783262.g6d822da:
  + upstream 1.5.0 release (2020-07-03)
  * cephadm iSCSI deployment (PR #300)
  * "--filestore" option to deploy OSDs with FileStore (PR #341)
  * "--devel/--product" option and add-repo subcommand (PR #351)
  * qa: superficial test for presence of dashboard branding (PR #374)
  * qa: add IGW to existing tests (PR #372)
  * provisioning: set fqdn through vagrant and leave /etc/hosts alone (PR #199)
  * no longer support nautilus deployment in Tumbleweed (PR #352)
  * ceph_salt_deployment.sh: reduce number of "ceph orch apply" calls (PR #363)
  * sesdev: raise exception if --roles combined with --single-node (PR #376)
  * cleanup: replace "grep | wc --lines" with "grep --count" (PR #375)
  * makecheck: reasonable defaults for Ceph repo/branch (PR #277)
  * provision.sh: set up SSH keys earlier (PR #361)
  * provision.sh: avoid endless while loop (PR #365)
  * sesdev: sanitize makecheck deployment IDs (PR #371)
  * doc: Update the example Deployment id in sesdev help (PR #378)
  * qa/systemctl_test: better error message when FSID omitted (PR #373)
  * qa/rgw: run curl command in try-wait loop to ping RGW (PR #370)
  * qa: do not wait for non-existent daemons to start (PR #368)
  * qa: get RGW port from /etc/ceph/ceph.conf (PR #353)

- Update to 1.4.0+1592681394.g28f5853:
  + upstream 1.4.0 release (2020-06-20)
  * added "replace-mgr-modules" subcommand (PR #24)
  * added "replace-ceph-salt" subcommand (PR #331)
  * added NFS (Ganesha) server deployment in {octopus,ses7,pacific} (PR #337)
  * qa: added curl-based RGW smoke test (PR #344)
  * qa: added systemctl-based smoke test (PR #347)
  * removed "--use-deepsea" option for {ses7,octopus,pacific} (PR #334)
  * ceph_salt_deployment: fix use_salt=True deployment (PR #326)
  * sync_clocks.sh: put chronyc calls in try_wait (PR #328)
  * qa: tolerate +1 MGRs in number_of_nodes_actual_vs_expected_test (PR #332)
  * ceph_salt_deployment: deploy OSDs from YAML (ServiceSpec) file (PR #203)
  * makecheck: possibly prophylactically downgrade libncurses6 (PR #325)
  * explicit "admin" role no longer allowed (PR #330)
  * seslib: move boilerplate ssh options into a staticmethod (PR #333)
  * deployment IDs are now vetted for correctness (PR #335)
  * seslib: generate comma-separated lists of nodes with each role (PR #345)
  * If not provided explicitly via the "--domain" option, new deployments now
    default to ".test" instead of ".com" as the cluster TLD (PR #350)