Test use_sbd as a boolean

[mpagot]

Verification

SBD

Azure

• sle-15-SP5-Qesap-Azure-Byos-x86_64-BuildLATEST_AZURE_SLE15_5_BYOS-qesap_azure_fullypatch@64bit -> http://openqaworker15.qa.suse.cz/tests/249497

Inventory has

all:
  vars:
    use_sbd: true

All playbooks are called without any -e sue_sbd=* http://openqaworker15.qa.suse.cz/tests/249497/file/configure-qesap_azure_fullypatch.yaml

ansible:
  create:
    - registration.yaml -e reg_code='****' -e email_address='****'
    - fully-patch-system.yaml
    - pre-cluster.yaml
    - sap-hana-preconfigure.yaml -e use_reboottimeout=900
    - cluster_sbd_prep.yaml
    - sap-hana-storage.yaml
    - sap-hana-download-media.yaml
    - sap-hana-install.yaml
    - sap-hana-system-replication.yaml
    - sap-hana-system-replication-hooks.yaml
    - sap-hana-cluster.yaml

In the Ansible log http://openqaworker15.qa.suse.cz/tests/249497/logfile?filename=deploy-qesap_exec_ansible__profile.log.txt

DEBUG    OUTPUT: TASK [Variable use_sbd] ********************************************************
DEBUG    OUTPUT: task path: /root/qe-sap-deployment/ansible/playbooks/tasks/azure-cluster-hana.yaml:33
DEBUG    OUTPUT: Wednesday 18 October 2023  10:40:59 -0400 (0:00:01.807)       0:02:04.776 ***** 
DEBUG    OUTPUT: META: noop
DEBUG    OUTPUT: ok: [vmhana01] => {
DEBUG    OUTPUT:     "msg": "Cloud platform appears to be use_sbd:True use_sbd bool:True use_sbd string:False"
DEBUG    OUTPUT: }

All conditional tasks from ansible/playbooks/tasks/azure-cluster-bootstrap.yaml, like Slurp SBD config, are executed.
Conditional tasks Set stonith-timeout [sdb] and Set stonith-timeout [azure fencing] from ansible/playbooks/tasks/azure-cluster-hana.yaml are not executed.
Task file tasks/cluster-bootstrap.yaml is not included so none of the task from it are executed

crm status after the deployment

Status of pacemakerd: 'Pacemaker is running' (last updated 2023-10-18 14:43:25Z)
Cluster Summary:
  * Stack: corosync
  * Current DC: vmhana01 (version 2.1.5+20221208.a3f44794f-150500.6.5.8-2.1.5+20221208.a3f44794f) - partition with quorum
  * Last updated: Wed Oct 18 14:43:25 2023
  * Last change:  Wed Oct 18 14:43:17 2023 by root via crm_attribute on vmhana01
  * 2 nodes configured
  * 9 resource instances configured

Node List:
  * Online: [ vmhana01 vmhana02 ]

Full List of Resources:
  * stonith-sbd	(stonith:external/sbd):	 Started vmhana01
  * Clone Set: cln_azure-events [rsc_azure-events]:
    * Started: [ vmhana01 vmhana02 ]
  * Clone Set: cln_SAPHanaTopology_HDB_HA000 [rsc_SAPHanaTopology_HDB_HA000]:
    * Started: [ vmhana01 vmhana02 ]
  * Clone Set: msl_SAPHana_HDB_HA000 [rsc_SAPHana_HDB_HA000] (promotable):
    * rsc_SAPHana_HDB_HA000	(ocf::suse:SAPHana):	 FAILED vmhana01 (Monitoring)
    * Slaves: [ vmhana02 ]
  * rsc_socat_HDB_HA000	(ocf::heartbeat:azure-lb):	 Started vmhana02
  * Resource Group: g_ip_HDB_HA000:
    * rsc_ip_HA000	(ocf::heartbeat:IPaddr2):	 Started vmhana01

Failed Resource Actions:
  * rsc_SAPHana_HDB_HA000_monitor_61000 on vmhana01 'promoted' (8): call=50, status='complete', last-rc-change='Wed Oct 18 14:42:27 2023', queued=0ms, exec=4896ms
  

AWS

VR using qesap_reg_aws_gcp_sbd

• sle-15-SP5-Qesap-Aws-Byos-x86_64-BuildLATEST_AWS_SLE15_5_BYOS-qesap_aws_saptune_test@64bit -> http://openqaworker15.qa.suse.cz/tests/250537

GCP

VR using qesap_reg_aws_gcp_sbd

• sle-15-SP5-Qesap-Gcp-Byos-x86_64-BuildLATEST_GCE_SLE15_5_BYOS-qesap_gcp_saptune_test@64bit -> http://openqaworker15.qa.suse.cz/tests/250538

Native

AWS

using -e use_sbd=false

• sle-15-SP5-Qesap-Aws-Byos-x86_64-BuildLATEST_AWS_SLE15_5_BYOS-qesap_aws_fullypatch@64bit -> http://openqaworker15.qa.suse.cz/tests/249498

Inventory has

all:
  vars:
    use_sbd: false

Ansible playbooks are called as http://openqaworker15.qa.suse.cz/tests/249498/file/configure-qesap_aws_fullypatch.yaml

- sap-hana-cluster.yaml -e use_sbd=false

The Ansible log has : http://openqaworker15.qa.suse.cz/tests/249498/logfile?filename=deploy-qesap_exec_ansible__profile.log.txt

ERROR    OUTPUT:          TASK [Variable use_sbd] ********************************************************
ERROR    OUTPUT:          task path: /root/qe-sap-deployment/ansible/playbooks/tasks/cluster-bootstrap.yaml:221
ERROR    OUTPUT:          Wednesday 18 October 2023  11:40:25 -0400 (0:00:00.096)       0:00:49.478 ***** 
ERROR    OUTPUT:          ok: [vmhana01] => {
ERROR    OUTPUT:              "msg": "Cloud platform appears to be use_sbd:false use_sbd bool:False use_sbd string:True"
ERROR    OUTPUT:          }
ERROR    OUTPUT:          ok: [vmhana02] => {
ERROR    OUTPUT:              "msg": "Cloud platform appears to be use_sbd:false use_sbd bool:False use_sbd string:True"
ERROR    OUTPUT:          }

So TASKs from ansible/playbooks/tasks/cluster-bootstrap.yaml :

• [Enable SBD [sbd]]
• [Set stonith-timeout [sdb]]
• [Set stonith timeout [native - aws]]
• [Enable stonith]
• [Disable stonith action [aws]]

are all SKIPPED. Then Ansible FATALs at task Configure cluster IP [aws]

ERROR    OUTPUT:                      "_raw_params": "crm configure primitive rsc_ip_HDB_HDB00 ocf:suse:aws-vpc-move-ip params ip=192.168.1.10 routing_table=rtb-02e3b718a0f668552 interface=eth0 profile=default op start interval=0 timeout=180 op stop interval=0 timeout=180 op monitor interval=60 timeout=60",

...

ERROR    OUTPUT:              "rc": 1,

...

ERROR    OUTPUT:              "stderr_lines": [
ERROR    OUTPUT:                  "\u001b[33mWARNING\u001b[0m: (cluster_status) \twarning: Fencing and resource management disabled due to lack of quorum",
ERROR    OUTPUT:                  "\u001b[31mERROR\u001b[0m: (unpack_resources) \terror: Resource start-up disabled since no STONITH resources have been defined",
ERROR    OUTPUT:                  "\u001b[31mERROR\u001b[0m: (unpack_resources) \terror: Either configure some or disable STONITH with the stonith-enabled option",
ERROR    OUTPUT:                  "\u001b[31mERROR\u001b[0m: (unpack_resources) \terror: NOTE: Clusters with shared data need STONITH to ensure data integrity",
ERROR    OUTPUT:                  "\u001b[31mERROR\u001b[0m: crm_verify: Errors found during check: config not valid"
ERROR    OUTPUT:              ],

not using -e use_sbd=false

Obtained including also os-autoinst/os-autoinst-distri-opensuse#17978

GCP

using -e use_sbd=false

not using -e use_sbd=false

Obtained including also os-autoinst/os-autoinst-distri-opensuse#17978

VR: - sle-15-SP5-Qesap-Gcp-Byos-x86_64-BuildLATEST_GCE_SLE15_5_BYOS-qesap_gcp_sapconf_test@64bit ->
http://openqaworker15.qa.suse.cz/tests/246979

Inventory has

all:
  vars:
    use_sbd: false

and playbooks are called without any -e use_sbd= http://openqaworker15.qa.suse.cz/tests/246979/file/configure-qesap_gcp_sapconf.yaml

The test Enable SBD [sbd] in the Ansible log is now

ERROR    OUTPUT:          TASK [Enable SBD [sbd]] ********************************************************
ERROR    OUTPUT:          task path: /root/qe-sap-deployment/ansible/playbooks/tasks/cluster-bootstrap.yaml:225
ERROR    OUTPUT:          Friday 13 October 2023  10:38:33 -0400 (0:00:00.070)       0:00:40.621 ******** 
ERROR    OUTPUT:          skipping: [qesapval246979-vmhana01] => {
ERROR    OUTPUT:              "changed": false,
ERROR    OUTPUT:              "skip_reason": "Conditional result was False"
ERROR    OUTPUT:          }
ERROR    OUTPUT:          skipping: [qesapval246979-vmhana02] => {
ERROR    OUTPUT:              "changed": false,
ERROR    OUTPUT:              "skip_reason": "Conditional result was False"
ERROR    OUTPUT:          }

But Ansible fails later with

ERROR    OUTPUT:          TASK [Configure cluster IP [gcp]] **********************************************


...

ERROR    OUTPUT:          fatal: [qesapval246979-vmhana01]: FAILED! => {

...

ERROR    OUTPUT:                      "_raw_params": "crm configure primitive rsc_ip_HDB_HDB00 IPaddr2 params ip=10.0.0.12 cidr_netmask=32 nic=eth0 op monitor interval=3600s timeout=60s",


...

ERROR    OUTPUT:              "rc": 1,

...

ERROR    OUTPUT:              "stderr_lines": [
ERROR    OUTPUT:                  "\u001b[33mWARNING\u001b[0m: (cluster_status) \twarning: Fencing and resource management disabled due to lack of quorum",
ERROR    OUTPUT:                  "\u001b[31mERROR\u001b[0m: (unpack_resources) \terror: Resource start-up disabled since no STONITH resources have been defined",
ERROR    OUTPUT:                  "\u001b[31mERROR\u001b[0m: (unpack_resources) \terror: Either configure some or disable STONITH with the stonith-enabled option",
ERROR    OUTPUT:                  "\u001b[31mERROR\u001b[0m: (unpack_resources) \terror: NOTE: Clusters with shared data need STONITH to ensure data integrity",
ERROR    OUTPUT:                  "\u001b[31mERROR\u001b[0m: crm_verify: Errors found during check: config not valid"
ERROR    OUTPUT:              ],


ERROR    OUTPUT:              "stdout": "",

crm status after the deployment is not executed