scenarios: added a scenario for detecting NFS name resolution failures (

#876)

Fixes #716

Signed-off-by: Mustafa Kemal Gilor <[email protected]>

hotsos/core/issues/issue_types.py

@@ -118,6 +118,10 @@ class NetworkWarning(IssueTypeBase):
     pass
 
 
+class NFSNameResolutionError(IssueTypeBase):
+    pass
+
+
 class RabbitMQWarning(IssueTypeBase):
     pass
 

hotsos/defs/scenarios/storage/nfs/nfs-detect-mount-name-resolution-failure.yaml

@@ -0,0 +1,15 @@
+checks:
+  has_nfs_name_resolution_failures:
+    input:
+      path: ['var/log/syslog', 'var/log/boot.log']
+    expr: '.*mount.nfs.*Failed to resolve server (.*): (Name or service not known|Temporary failure in name resolution).*'
+conclusions:
+  nfs_name_resolution_failures:
+    decision: has_nfs_name_resolution_failures
+    raises:
+      type: NFSNameResolutionError
+      message: >-
+        Name resolution failures have been detected for the following NFS mounts:
+        {names}
+      format-dict:
+        names: '@checks.has_nfs_name_resolution_failures.search.results_group_1:unique_comma_join'

hotsos/defs/tests/scenarios/storage/nfs/nfs-name-or-service-not-known.yaml

@@ -0,0 +1,12 @@
+target-name: nfs-detect-mount-name-resolution-failure.yaml
+data-root:
+  files:
+    var/log/syslog: |
+      Aug 23 07:27:31 juju-0c7157-trilio-test-16 mount[809]: mount.nfs: Failed to resolve server my.great.nfsserver.org: Name or service not known
+      Oct  9 11:28:22 mail mount[714]: mount.nfs4: Failed to resolve server my.another.great.nfs.org: Temporary failure in name resolution
+  copy-from-original:
+    - sos_commands/date/date
+raised-issues:
+  NFSNameResolutionError: >-
+    Name resolution failures have been detected for the following NFS mounts:
+    my.another.great.nfs.org, my.great.nfsserver.org

hotsos/defs/tests/scenarios/storage/nfs/nfs-temporary-name-resolution-failure.yaml

@@ -0,0 +1,14 @@
+target-name: nfs-detect-mount-name-resolution-failure.yaml
+data-root:
+  files:
+    var/log/boot.log: |
+      mount.nfs: Failed to resolve server filesrv23: Temporary failure in name resolution
+      mountall: mount /epgbatch_data [623] terminated with status 32
+      mount.nfs: Failed to resolve server filesrv24: Temporary failure in name resolution
+      mountall: mount /epgbatch_data [632] terminated with status 32
+  copy-from-original:
+    - sos_commands/date/date
+raised-issues:
+  NFSNameResolutionError: >-
+    Name resolution failures have been detected for the following NFS mounts:
+    filesrv23, filesrv24

tests/unit/storage/test_nfs.py

@@ -0,0 +1,22 @@
+
+from hotsos.core.config import HotSOSConfig
+
+from .. import utils
+
+
+class StorageNFSTestsBase(utils.BaseTestCase):
+
+    def setUp(self):
+        super().setUp()
+        HotSOSConfig.plugin_name = 'storage'
+
+
+@utils.load_templated_tests('scenarios/storage/nfs')
+class TestNFSScenarios(StorageNFSTestsBase):
+    """
+    Scenario tests can be written using YAML templates that are auto-loaded
+    into this test runner. This is the recommended way to write tests for
+    scenarios. It is however still possible to write the tests in Python if
+    required. See https://hotsos.readthedocs.io/en/latest/contrib/testing.html
+    for more information.
+    """