qa: update to tentacle

Signed-off-by: Patrick Donnelly <pdonnell@ibm.com>

doc/dev/release-checklists.rst

@@ -104,12 +104,13 @@ Code cleanup
 QA suite
 --------
 
-- [?] create qa/workunits/test_telemetry_(X-1).sh
-- [?] create qa/workunits/test_telemetry_(X-1)_x.sh
-- [ ] create qa/suites/upgrade/(X-1)-x
-- [ ] remove qa/suites/upgrade/(X-3)-x-*
-- [ ] create qa/releases/X.yaml
-- [ ] create qa/suites/rados/thrash-old-clients/1-install/(X-1).yaml
+- [x] create qa/workunits/test_telemetry_(X-1).sh
+- [x] create qa/workunits/test_telemetry_(X-1)_x.sh
+- [x] create qa/suites/upgrade/(X-1)-x
+- [x] remove qa/suites/upgrade/(X-3)-x-*
+- [x] update qa/fs/upgrade/ to remove (X-3) and add (X-1); check with fs team to confirm / help
+- [x] create qa/releases/X.yaml
+- [x] create qa/suites/rados/thrash-old-clients/1-install/(X-1).yaml
 
 
 ceph-build

qa/releases/tentacle.yaml

@@ -0,0 +1,6 @@
+tasks:
+- exec:
+    osd.0:
+      - ceph osd require-osd-release tentacle
+      - ceph osd set-require-min-compat-client tentacle
+- ceph.healthy:

qa/suites/fs/upgrade/upgraded_client/tasks/0-install/squid.yaml

@@ -1,9 +1,9 @@
 meta:
 - desc: |
-   install ceph/quincy latest
+   install ceph/squid latest
 tasks:
 - install:
-    branch: quincy
+    branch: squid
     exclude_packages:
       - cephadm
       - ceph-mgr-cephadm
@@ -27,7 +27,7 @@ tasks:
       - NetworkManager
       - iproute
       - util-linux
-- print: "**** done installing quincy"
+- print: "**** done installing squid"
 - ceph:
     log-ignorelist:
       - overall HEALTH_
@@ -49,5 +49,5 @@ tasks:
         ms bind msgr2: false
 - exec:
     osd.0:
-      - ceph osd set-require-min-compat-client quincy
+      - ceph osd set-require-min-compat-client squid
 - print: "**** done ceph"

qa/suites/rados/thrash-old-clients/1-install/tentacle.yaml

@@ -4,7 +4,7 @@ overrides:
       - \(MON_DOWN\)
 tasks:
 - install:
-    branch: quincy
+    branch: tentacle
     exclude_packages:
       - ceph-mgr-dashboard
       - ceph-mgr-diskprediction-local

qa/suites/upgrade/quincy-x/filestore-remove-check/.qa

@@ -1 +0,0 @@
-../.qa

qa/suites/upgrade/quincy-x/filestore-remove-check/0-cluster/openstack.yaml

@@ -1,6 +0,0 @@
-openstack:
-  - machine:
-      disk: 100 # GB
-  - volumes: # attached to each instance
-      count: 4
-      size: 30 # GB

qa/suites/upgrade/quincy-x/filestore-remove-check/0-cluster/start.yaml

@@ -1,46 +0,0 @@
-meta:
-- desc: |
-   Run ceph on one nodes,
-   Use xfs beneath the osds. upgrade to reef
-   should fail to start the osds with filestore
-overrides:
-  ceph:
-    mon_bind_msgr2: false
-    mon_bind_addrvec: false
-    mon-health-to-clog: false
-    wait-for-healthy: false
-    wait-for-osds-up: false
-    wait-for-scrub: false
-    skip_stop_pg_num_changes: true
-    fs: xfs
-    log-ignorelist:
-      - overall HEALTH_
-      - \(MON_DOWN\)
-      - \(MGR_DOWN\)
-      - slow request
-      - \(MON_MSGR2_NOT_ENABLED\)
-      - \(OSD_DOWN\)
-      - \(OSD_HOST_DOWN\)
-      - \(POOL_APP_NOT_ENABLED\)
-      - OSD_DOWN
-      - mons down
-      - mon down
-      - MON_DOWN
-      - out of quorum
-      - PG_DEGRADED
-      - Reduced data availability
-      - Degraded data redundancy
-      - OSDMAP_FLAGS
-      - OSD_ROOT_DOWN
-
-    conf:
-      global:
-        enable experimental unrecoverable data corrupting features: "*"
-        mon warn on msgr2 not enabled: false
-      mon:
-        mon warn on osd down out interval zero: false
-roles:
-- - mon.a
-  - mgr.x
-  - osd.0
-  - osd.1

qa/suites/upgrade/quincy-x/filestore-remove-check/1-ceph-install/quincy.yaml

@@ -1,29 +0,0 @@
-meta:
-- desc: install ceph/quincy latest
-tasks:
-- install:
-    exclude_packages:
-      - ceph-mgr-cephadm
-      - cephadm
-      - libcephfs-dev
-    branch: quincy
-- print: "**** done install quincy"
-- ceph:
-    create_rbd_pool: false
-    conf:
-      global:
-        bluestore_warn_on_legacy_statfs: false
-        bluestore warn on no per pool omap: false
-        mon pg warn min per osd: 0
-      mon:
-        mon_warn_on_insecure_global_id_reclaim: false
-        mon_warn_on_insecure_global_id_reclaim_allowed: false
-- exec:
-    osd.0:
-      - ceph osd require-osd-release quincy
-- print: "**** done ceph"
-overrides:
-  ceph:
-    conf:
-      mon:
-        mon warn on osd down out interval zero: false

qa/suites/upgrade/quincy-x/filestore-remove-check/2 - upgrade.yaml

@@ -1,19 +0,0 @@
-meta:
-- desc: |
-   install upgrade ceph/-x on cluster
-   restart : mons, osd.*
-tasks:
-- print: "**** start install.upgrade of nodes"
-- install.upgrade:
-    all:
-- print: "**** done install.upgrade of nodes"
-- print: "**** start ceph.restart of all osds"
-- ceph.restart:
-    daemons: [osd.0,osd.1,osd.2]
-    mon-health-to-clog: false
-    wait-for-healthy: false
-    wait-for-osds-up: false
-    wait-for-scrub: false
-    skip_stop_pg_num_changes: true
-    expected-failure: "FileStore has been deprecated and is no longer supported"
-- print: "**** done ceph.restart of all mons and osds"

qa/suites/upgrade/quincy-x/filestore-remove-check/objectstore/filestore-xfs.yaml

@@ -1,14 +0,0 @@
-overrides:
-  ceph:
-    fs: xfs
-    conf:
-      osd:
-        osd objectstore: filestore
-        osd sloppy crc: true
-  ceph-deploy:
-    fs: xfs
-    filestore: True
-    conf:
-      osd:
-        osd objectstore: filestore
-        osd sloppy crc: true

qa/suites/upgrade/quincy-x/filestore-remove-check/ubuntu_latest.yaml

@@ -1 +0,0 @@
-.qa/distros/supported/ubuntu_latest.yaml

qa/suites/upgrade/quincy-x/stress-split/2-first-half-tasks/.qa

@@ -1 +0,0 @@
-../.qa/

qa/suites/upgrade/quincy-x/stress-split/3-stress-tasks/.qa

@@ -1 +0,0 @@
-../.qa/

qa/suites/upgrade/quincy-x/stress-split/4-second-half-tasks/.qa

@@ -1 +0,0 @@
-../.qa/

qa/suites/upgrade/squid-x/parallel/0-start.yaml

@@ -51,3 +51,4 @@ overrides:
       - FS_DEGRADED
       - OSDMAP_FLAGS
       - OSD_UPGRADE_FINISHED
+      - PG_DEGRADED

qa/suites/upgrade/squid-x/parallel/1-tasks.yaml

@@ -5,21 +5,20 @@ overrides:
       - telemetry module includes new collections
 tasks:
 - install:
-    branch: quincy
+    branch: squid
     exclude_packages:
       - ceph-volume
 - print: "**** done install task..."
-- print: "**** done start installing quincy cephadm ..."
+- print: "**** done start installing squid cephadm ..."
 - cephadm:
-    image: quay.ceph.io/ceph-ci/ceph:quincy
-    cephadm_branch: quincy
-    cephadm_git_url: https://github.com/ceph/ceph
+    image: quay.ceph.io/ceph-ci/ceph:squid
+    compiled_cephadm_branch: squid
     conf:
       osd:
         #set config option for which cls modules are allowed to be loaded / used
         osd_class_load_list: "*"
         osd_class_default_list: "*"
-- print: "**** done end installing quincy cephadm ..."
+- print: "**** done end installing squid cephadm ..."
 
 - print: "**** done start cephadm.shell ceph config set mgr..."
 - cephadm.shell:
@@ -27,12 +26,12 @@ tasks:
       - ceph config set mgr mgr/cephadm/use_repo_digest true --force
 - print: "**** done cephadm.shell ceph config set mgr..."
 
-- print: "**** done start telemetry quincy..."
+- print: "**** done start telemetry squid..."
 - workunit:
     clients:
       client.0:
-        - test_telemetry_quincy.sh
-- print: "**** done end telemetry quincy..."
+        - test_telemetry_squid.sh
+- print: "**** done end telemetry squid..."
 
 - print: "**** done start parallel"
 - parallel:
@@ -44,5 +43,5 @@ tasks:
 - workunit:
     clients:
       client.0:
-        - test_telemetry_quincy_x.sh
+        - test_telemetry_squid_x.sh
 - print: "**** done end telemetry x..."

qa/suites/upgrade/squid-x/parallel/overrides/ignorelist_health.yaml

@@ -0,0 +1,20 @@
+overrides:
+  ceph:
+    log-ignorelist:
+      - \(MDS_ALL_DOWN\)
+      - \(MDS_UP_LESS_THAN_MAX\)
+      - \(OSD_SLOW_PING_TIME
+      - reached quota
+      - overall HEALTH_
+      - \(CACHE_POOL_NO_HIT_SET\)
+      - \(POOL_FULL\)
+      - \(SMALLER_PGP_NUM\)
+      - \(SLOW_OPS\)
+      - \(CACHE_POOL_NEAR_FULL\)
+      - \(POOL_APP_NOT_ENABLED\)
+      - \(PG_AVAILABILITY\)
+      - \(OBJECT_MISPLACED\)
+      - slow request
+      - \(MON_DOWN\)
+      - noscrub
+      - nodeep-scrub

qa/suites/upgrade/squid-x/parallel/workload/rados_api.yaml

@@ -5,7 +5,7 @@ workload:
   full_sequential:
     - print: "**** done start rados_api.yaml"
     - workunit:
-        branch: quincy
+        branch: squid
         clients:
           client.0:
             - cls

qa/suites/upgrade/squid-x/parallel/workload/rados_loadgenbig.yaml

@@ -5,7 +5,7 @@ workload:
   full_sequential:
     - print: "**** done start rados_loadgenbig.yaml"
     - workunit:
-        branch: quincy
+        branch: squid
         clients:
           client.0:
             - rados/load-gen-big.sh

qa/suites/upgrade/squid-x/parallel/workload/rbd_import_export.yaml

@@ -5,7 +5,7 @@ workload:
   full_sequential:
     - print: "**** done start rbd_import_export.yaml"
     - workunit:
-        branch: quincy
+        branch: squid
         clients:
           client.1:
               - rbd/import_export.sh

qa/suites/upgrade/squid-x/parallel/workload/test_rbd_api.yaml

@@ -5,7 +5,7 @@ workload:
   full_sequential:
     - print: "**** done start test_rbd_api.yaml"
     - workunit:
-        branch: quincy
+        branch: squid
         clients:
           client.0:
               - rbd/test_librbd.sh

qa/suites/upgrade/squid-x/parallel/workload/test_rbd_python.yaml

@@ -10,7 +10,7 @@ workload:
   full_sequential:
     - print: "**** done start test_rbd_python.yaml"
     - workunit:
-        branch: quincy
+        branch: squid
         clients:
           client.0:
             - rbd/test_librbd_python.sh

qa/suites/upgrade/squid-x/stress-split/1-start.yaml

@@ -22,14 +22,13 @@ overrides:
       - OSD_UPGRADE_FINISHED
 tasks:
 - install:
-    branch: quincy
+    branch: squid
     exclude_packages:
       - ceph-volume
 
 - cephadm:
-    image: quay.ceph.io/ceph-ci/ceph:quincy
-    cephadm_branch: quincy
-    cephadm_git_url: https://github.com/ceph/ceph
+    image: quay.ceph.io/ceph-ci/ceph:squid
+    compiled_cephadm_branch: squid
     conf:
       osd:
         #set config option for which cls modules are allowed to be loaded / used
@@ -76,37 +75,33 @@ first-half-sequence:
       - ceph config set mgr mgr/cephadm/daemon_cache_timeout 60
       - ceph config set global log_to_journald false --force
 
-      - ceph orch upgrade start --image quay.ceph.io/ceph-ci/ceph:$sha1
-      - ceph orch ps
+      - echo wait for mgr daemons to upgrade
+      # upgrade the mgr daemons first
+      - ceph orch upgrade start --image quay.ceph.io/ceph-ci/ceph:$sha1 --daemon-types mgr
+      - while ceph orch upgrade status | jq '.in_progress' | grep true && ! ceph orch upgrade status | jq '.message' | grep Error ; do ceph orch ps ; ceph versions ; ceph orch upgrade status ; sleep 30 ; done
 
       - echo wait for minority of mons to upgrade
-      - while ! ceph mon versions | grep $sha1 && ceph orch upgrade status | jq '.in_progress' | grep true && ! ceph orch upgrade status | jq '.message' | grep Error; do sleep 2 ; done
-      - ceph orch ps
-      - ceph orch upgrade pause
+      # upgrade 1 of 3 mon daemons, then wait 60 seconds
+      - ceph orch upgrade start --image quay.ceph.io/ceph-ci/ceph:$sha1 --daemon-types mon --limit 1
+      - while ceph orch upgrade status | jq '.in_progress' | grep true && ! ceph orch upgrade status | jq '.message' | grep Error ; do ceph orch ps ; ceph versions ; ceph orch upgrade status ; sleep 30 ; done
       - sleep 60
-      - ceph orch upgrade resume
 
       - echo wait for majority of mons to upgrade
-      - "while ! ceph mon versions | grep $sha1 | egrep ': [23]' && ceph orch upgrade status | jq '.in_progress' | grep true && ! ceph orch upgrade status | jq '.message' | grep Error; do sleep 2 ; done"
-      - ceph orch ps
-      - ceph orch upgrade pause
+      # upgrade one more mon daemon (to get us to 2/3 upgraded) and wait 60 seconds
+      - ceph orch upgrade start --image quay.ceph.io/ceph-ci/ceph:$sha1 --daemon-types mon --limit 1
+      - while ceph orch upgrade status | jq '.in_progress' | grep true && ! ceph orch upgrade status | jq '.message' | grep Error ; do ceph orch ps ; ceph versions ; ceph orch upgrade status ; sleep 30 ; done
       - sleep 60
-      - ceph orch upgrade resume
 
       - echo wait for all mons to upgrade
-      - "while ! ceph mon versions | grep $sha1 | grep ': 3' && ceph orch upgrade status | jq '.in_progress' | grep true && ! ceph orch upgrade status | jq '.message' | grep Error; do sleep 2 ; done"
-      - ceph orch ps
-      - ceph orch upgrade pause
+      # upgrade final mon daemon and wait 60 seconds
+      - ceph orch upgrade start --image quay.ceph.io/ceph-ci/ceph:$sha1 --daemon-types mon
+      - while ceph orch upgrade status | jq '.in_progress' | grep true && ! ceph orch upgrade status | jq '.message' | grep Error ; do ceph orch ps ; ceph versions ; ceph orch upgrade status ; sleep 30 ; done
       - sleep 60
-      - ceph orch upgrade resume
 
       - echo wait for half of osds to upgrade
-      - "while ! ceph osd versions | grep $sha1 | egrep ': [45678]' && ceph orch upgrade status | jq '.in_progress' | grep true && ! ceph orch upgrade status | jq '.message' | grep Error; do sleep 2 ; done"
-      - ceph orch upgrade pause
-      - ceph orch ps
-
-      - ceph orch ps
-      - ceph versions
+      # upgrade 4 of the 8 OSDs
+      - ceph orch upgrade start --image quay.ceph.io/ceph-ci/ceph:$sha1 --daemon-types osd --limit 4
+      - while ceph orch upgrade status | jq '.in_progress' | grep true && ! ceph orch upgrade status | jq '.message' | grep Error ; do ceph orch ps ; ceph versions ; ceph orch upgrade status ; sleep 30 ; done
 
 
 #################
@@ -131,11 +126,12 @@ second-half-sequence:
     - cephadm.shell:
         env: [sha1]
         mon.a:
-          - ceph orch upgrade resume
           - sleep 60
 
           - echo wait for upgrade to complete
-          - while ceph orch upgrade status | jq '.in_progress' | grep true ; do ceph orch ps ; ceph versions ; sleep 30 ; done
+          # upgrade whatever is left
+          - ceph orch upgrade start --image quay.ceph.io/ceph-ci/ceph:$sha1
+          - while ceph orch upgrade status | jq '.in_progress' | grep true && ! ceph orch upgrade status | jq '.message' | grep Error ; do ceph orch ps ; ceph versions ; ceph orch upgrade status ; sleep 30 ; done
 
           - echo upgrade complete
           - ceph orch ps

qa/suites/upgrade/squid-x/stress-split/2-first-half-tasks/rbd-cls.yaml

@@ -3,7 +3,7 @@ meta:
    run basic cls tests for rbd
 first-half-tasks:
 - workunit:
-    branch: quincy
+    branch: squid
     clients:
       client.0:
         - cls/test_cls_rbd.sh

qa/suites/upgrade/squid-x/stress-split/2-first-half-tasks/rbd-import-export.yaml

@@ -3,7 +3,7 @@ meta:
    run basic import/export cli tests for rbd
 first-half-tasks:
 - workunit:
-    branch: quincy
+    branch: squid
     clients:
       client.0:
         - rbd/import_export.sh

qa/suites/upgrade/squid-x/stress-split/2-first-half-tasks/rbd_api.yaml

@@ -3,7 +3,7 @@ meta:
    librbd C and C++ api tests
 first-half-tasks:
 - workunit:
-     branch: quincy
+     branch: squid
      clients:
         client.0:
            - rbd/test_librbd.sh

qa/suites/upgrade/squid-x/stress-split/3-stress-tasks/rbd-cls.yaml

@@ -3,7 +3,7 @@ meta:
    run basic cls tests for rbd
 stress-tasks:
 - workunit:
-    branch: quincy
+    branch: squid
     clients:
       client.0:
         - cls/test_cls_rbd.sh

qa/suites/upgrade/squid-x/stress-split/3-stress-tasks/rbd-import-export.yaml

@@ -3,7 +3,7 @@ meta:
    run basic import/export cli tests for rbd
 stress-tasks:
 - workunit:
-    branch: quincy
+    branch: squid
     clients:
       client.0:
         - rbd/import_export.sh

qa/suites/upgrade/squid-x/stress-split/3-stress-tasks/rbd_api.yaml

@@ -3,7 +3,7 @@ meta:
    librbd C and C++ api tests
 stress-tasks:
 - workunit:
-     branch: quincy
+     branch: squid
      clients:
         client.0:
            - rbd/test_librbd.sh

qa/suites/upgrade/squid-x/stress-split/4-second-half-tasks/rbd-import-export.yaml

@@ -3,7 +3,7 @@ meta:
    run basic import/export cli tests for rbd
 second-half-tasks:
 - workunit:
-    branch: quincy
+    branch: squid
     clients:
       client.0:
         - rbd/import_export.sh

qa/suites/upgrade/squid-x/stress-split/overrides/ignorelist_health.yaml

@@ -0,0 +1,20 @@
+overrides:
+  ceph:
+    log-ignorelist:
+      - \(MDS_ALL_DOWN\)
+      - \(MDS_UP_LESS_THAN_MAX\)
+      - \(OSD_SLOW_PING_TIME
+      - reached quota
+      - overall HEALTH_
+      - \(CACHE_POOL_NO_HIT_SET\)
+      - \(POOL_FULL\)
+      - \(SMALLER_PGP_NUM\)
+      - \(SLOW_OPS\)
+      - \(CACHE_POOL_NEAR_FULL\)
+      - \(POOL_APP_NOT_ENABLED\)
+      - \(PG_AVAILABILITY\)
+      - \(OBJECT_MISPLACED\)
+      - slow request
+      - \(MON_DOWN\)
+      - noscrub
+      - nodeep-scrub