mgr/dashboard: cephadm e2e job: display info on error & other improvements

- Fix: ensure that on_error trap is called (display more info on error).
- Set static IPs to VMs.
- Remove domain in cluster definition to avoid side effects of potential dns misconfiguration.
- Minor improvements.

Fixes: https://tracker.ceph.com/issues/53991

Signed-off-by: Alfonso Martínez <[email protected]>

Author: alfonsomthd

src/pybind/mgr/dashboard/ci/cephadm/bootstrap-cluster.sh

@@ -22,9 +22,9 @@ fsid=$(cat /etc/ceph/ceph.conf | grep fsid | awk '{ print $3}')
 cephadm_shell="cephadm shell --fsid ${fsid} -c /etc/ceph/ceph.conf -k /etc/ceph/ceph.client.admin.keyring"
 
 {% for number in range(1, nodes) %}
-  ssh-copy-id -f -i /etc/ceph/ceph.pub  -o StrictHostKeyChecking=no root@{{ prefix }}-node-0{{ number }}.{{ domain }}
+  ssh-copy-id -f -i /etc/ceph/ceph.pub  -o StrictHostKeyChecking=no root@{{ prefix }}-node-0{{ number }}
   {% if expanded_cluster is defined %}
-    ${cephadm_shell} ceph orch host add {{ prefix }}-node-0{{ number }}.{{ domain }}
+    ${cephadm_shell} ceph orch host add {{ prefix }}-node-0{{ number }}
   {% endif %}
 {% endfor %}
 

src/pybind/mgr/dashboard/ci/cephadm/ceph_cluster.yml

@@ -1,8 +1,10 @@
 parameters:
  nodes: 4
+ node_ip_offset: 100
  pool: ceph-dashboard
  network: ceph-dashboard
- domain: cephlab.com
+ gateway: 192.168.100.1
+ netmask: 255.255.255.0
  prefix: ceph
  numcpus: 1
  memory: 2048
@@ -22,9 +24,12 @@ parameters:
  reserveip: true
  reservedns: true
  sharedkey: true
- domain: {{ domain }}
  nets:
-  - {{ network }}
+  - name: {{ network }}
+    ip: 192.168.100.{{ node_ip_offset + number }}
+    gateway: {{ gateway }}
+    mask: {{ netmask }}
+    dns: {{ gateway }}
  disks: {{ disks }}
  pool: {{ pool }}
  sharedfolders: [{{ ceph_dev_folder }}]

src/pybind/mgr/dashboard/ci/cephadm/start-cluster.sh

@@ -1,6 +1,6 @@
 #!/usr/bin/env bash
 
-set -ex
+set -eEx
 
 cleanup() {
     set +x
@@ -18,15 +18,15 @@ on_error() {
         printf "\n\nERROR $1 thrown on line $2\n\n"
         printf "\n\nCollecting info...\n\n"
         printf "\n\nDisplaying MGR logs:\n\n"
-        kcli ssh -u root -- ceph-node-00 'cephadm logs -n $(cephadm ls | grep -Eo "mgr\.ceph[0-9a-z.-]+" | head -n 1)'
+        kcli ssh -u root -- ceph-node-00 'cephadm logs -n \$(cephadm ls | grep -Eo "mgr\.ceph[0-9a-z.-]+" | head -n 1) -- --no-tail --no-pager'
         for vm_id in 0 1 2
         do
             local vm="ceph-node-0${vm_id}"
             printf "\n\nDisplaying journalctl from VM ${vm}:\n\n"
             kcli ssh -u root -- ${vm} 'journalctl --no-tail --no-pager -t cloud-init' || true
             printf "\n\nEnd of journalctl from VM ${vm}\n\n"
             printf "\n\nDisplaying container logs:\n\n"
-            kcli ssh -u root -- ${vm} 'podman logs --names --since 30s $(podman ps -aq)' || true
+            kcli ssh -u root -- ${vm} 'podman logs --names --since 30s \$(podman ps -aq)' || true
         done
         printf "\n\nTEST FAILED.\n\n"
     fi
@@ -80,7 +80,7 @@ while [[ -z $(kcli ssh -u root -- ceph-node-00 'journalctl --no-tail --no-pager
     kcli list vm
     if [[ ${CLUSTER_DEBUG} != 0 ]]; then
         kcli ssh -u root -- ceph-node-00 'podman ps -a'
-        kcli ssh -u root -- ceph-node-00 'podman logs --names --since 30s $(podman ps -aq)'
+        kcli ssh -u root -- ceph-node-00 'podman logs --names --since 30s \$(podman ps -aq)'
     fi
     kcli ssh -u root -- ceph-node-00 'journalctl -n 100 --no-pager -t cloud-init'
 done

src/pybind/mgr/dashboard/frontend/cypress/integration/orchestrator/workflow/02-create-cluster-add-host.e2e-spec.ts

@@ -6,12 +6,7 @@ import {
 describe('Create cluster add host page', () => {
   const createCluster = new CreateClusterWizardHelper();
   const createClusterHostPage = new CreateClusterHostPageHelper();
-  const hostnames = [
-    'ceph-node-00.cephlab.com',
-    'ceph-node-01.cephlab.com',
-    'ceph-node-02.cephlab.com',
-    'ceph-node-[01-03].cephlab.com'
-  ];
+  const hostnames = ['ceph-node-00', 'ceph-node-01', 'ceph-node-02', 'ceph-node-[01-03]'];
   const addHost = (hostname: string, exist?: boolean, pattern?: boolean, labels: string[] = []) => {
     cy.get('button[data-testid=table-action-button]').click();
     createClusterHostPage.add(hostname, exist, false, labels);

src/pybind/mgr/dashboard/frontend/cypress/integration/orchestrator/workflow/04-create-cluster-create-osds.e2e-spec.ts

@@ -20,11 +20,7 @@ describe('Create cluster create osds page', () => {
 
   describe('when Orchestrator is available', () => {
     it('should create OSDs', () => {
-      const hostnames = [
-        'ceph-node-00.cephlab.com',
-        'ceph-node-02.cephlab.com',
-        'ceph-node-03.cephlab.com'
-      ];
+      const hostnames = ['ceph-node-00', 'ceph-node-02', 'ceph-node-03'];
       for (const hostname of hostnames) {
         osds.create('hdd', hostname, true);
 

src/pybind/mgr/dashboard/frontend/cypress/integration/orchestrator/workflow/06-cluster-check.e2e-spec.ts

@@ -7,12 +7,7 @@ describe('when cluster creation is completed', () => {
   const services = new ServicesPageHelper();
   const hosts = new HostsPageHelper();
 
-  const hostnames = [
-    'ceph-node-00.cephlab.com',
-    'ceph-node-01.cephlab.com',
-    'ceph-node-02.cephlab.com',
-    'ceph-node-03.cephlab.com'
-  ];
+  const hostnames = ['ceph-node-00', 'ceph-node-01', 'ceph-node-02', 'ceph-node-03'];
 
   beforeEach(() => {
     cy.login();

src/pybind/mgr/dashboard/frontend/cypress/integration/orchestrator/workflow/08-hosts.e2e-spec.ts

@@ -5,12 +5,7 @@ describe('Host Page', () => {
   const hosts = new HostsPageHelper();
   const services = new ServicesPageHelper();
 
-  const hostnames = [
-    'ceph-node-00.cephlab.com',
-    'ceph-node-01.cephlab.com',
-    'ceph-node-02.cephlab.com',
-    'ceph-node-03.cephlab.com'
-  ];
+  const hostnames = ['ceph-node-00', 'ceph-node-01', 'ceph-node-02', 'ceph-node-03'];
 
   beforeEach(() => {
     cy.login();