Bump version to 0.9.63

.version

@@ -1 +1 @@
-0.9.62
+0.9.63

CHANGELOG.md

@@ -1,5 +1,13 @@
 ## PVC Changelog
 
+###### [v0.9.63](https://github.com/parallelvirtualcluster/pvc/releases/tag/v0.9.63)
+
+  * Mentions Ganeti in the docs
+  * Increases API timeout back to 2s
+  * Adds .update-* configs to dpkg plugin
+  * Adds full/nearfull OSD warnings
+  * Improves size value handling for volumes
+
 ###### [v0.9.62](https://github.com/parallelvirtualcluster/pvc/releases/tag/v0.9.62)
 
   * [all] Adds an enhanced health checking, monitoring, and reporting system for nodes and clusters

README.md

@@ -9,7 +9,7 @@
 
 ## What is PVC?
 
-PVC is a Linux KVM-based hyperconverged infrastructure (HCI) virtualization cluster solution that is fully Free Software, scalable, redundant, self-healing, self-managing, and designed for administrator simplicity. It is an alternative to other HCI solutions such as Harvester, Nutanix, and VMWare, as well as to other common virtualization stacks such as ProxMox and OpenStack.
+PVC is a Linux KVM-based hyperconverged infrastructure (HCI) virtualization cluster solution that is fully Free Software, scalable, redundant, self-healing, self-managing, and designed for administrator simplicity. It is an alternative to other HCI solutions such as Ganeti, Harvester, Nutanix, and VMWare, as well as to other common virtualization stacks such as ProxMox and OpenStack.
 
 PVC is a complete HCI solution, built from well-known and well-trusted Free Software tools, to assist an administrator in creating and managing a cluster of servers to run virtual machines, as well as self-managing several important aspects including storage failover, node failure and recovery, virtual machine failure and recovery, and network plumbing. It is designed to act consistently, reliably, and unobtrusively, letting the administrator concentrate on more important things.
 

api-daemon/pvcapid/Daemon.py

@@ -27,7 +27,7 @@ from ssl import SSLContext, TLSVersion
 from distutils.util import strtobool as dustrtobool
 
 # Daemon version
-version = "0.9.62"
+version = "0.9.63"
 
 # API version
 API_VERSION = 1.0

client-cli/pvc/cli_lib/common.py

@@ -124,8 +124,8 @@ def call_api(
     data=None,
     files=None,
 ):
-    # Set the connect timeout to 1 seconds but extremely long (48 hour) data timeout
-    timeout = (1.05, 172800)
+    # Set the connect timeout to 2 seconds but extremely long (48 hour) data timeout
+    timeout = (2.05, 172800)
 
     # Craft the URI
     uri = "{}://{}{}{}".format(

client-cli/setup.py

@@ -2,7 +2,7 @@ from setuptools import setup
 
 setup(
     name="pvc",
-    version="0.9.62",
+    version="0.9.63",
     packages=["pvc", "pvc.cli_lib"],
     install_requires=[
         "Click",

daemon-common/ceph.py

@@ -73,6 +73,11 @@ byte_unit_matrix = {
     "G": 1024 * 1024 * 1024,
     "T": 1024 * 1024 * 1024 * 1024,
     "P": 1024 * 1024 * 1024 * 1024 * 1024,
+    "E": 1024 * 1024 * 1024 * 1024 * 1024 * 1024,
+    "Z": 1024 * 1024 * 1024 * 1024 * 1024 * 1024 * 1024,
+    "Y": 1024 * 1024 * 1024 * 1024 * 1024 * 1024 * 1024 * 1024,
+    "R": 1024 * 1024 * 1024 * 1024 * 1024 * 1024 * 1024 * 1024 * 1024,
+    "Q": 1024 * 1024 * 1024 * 1024 * 1024 * 1024 * 1024 * 1024 * 1024 * 1024,
 }
 
 # Matrix of human-to-metric values
@@ -83,6 +88,11 @@ ops_unit_matrix = {
     "G": 1000 * 1000 * 1000,
     "T": 1000 * 1000 * 1000 * 1000,
     "P": 1000 * 1000 * 1000 * 1000 * 1000,
+    "E": 1000 * 1000 * 1000 * 1000 * 1000 * 1000,
+    "Z": 1000 * 1000 * 1000 * 1000 * 1000 * 1000 * 1000,
+    "Y": 1000 * 1000 * 1000 * 1000 * 1000 * 1000 * 1000 * 1000,
+    "R": 1000 * 1000 * 1000 * 1000 * 1000 * 1000 * 1000 * 1000 * 1000,
+    "Q": 1000 * 1000 * 1000 * 1000 * 1000 * 1000 * 1000 * 1000 * 1000 * 1000,
 }
 
 
@@ -103,14 +113,18 @@ def format_bytes_tohuman(databytes):
 
 
 def format_bytes_fromhuman(datahuman):
-    # Trim off human-readable character
-    dataunit = str(datahuman)[-1]
-    datasize = int(str(datahuman)[:-1])
-    if not re.match(r"[A-Z]", dataunit):
+    if not re.search(r"[A-Za-z]+", datahuman):
         dataunit = "B"
         datasize = int(datahuman)
-    databytes = datasize * byte_unit_matrix[dataunit]
-    return databytes
+    else:
+        dataunit = str(re.match(r"[0-9]+([A-Za-z])[iBb]*", datahuman).group(1))
+        datasize = int(re.match(r"([0-9]+)[A-Za-z]+", datahuman).group(1))
+
+    if byte_unit_matrix.get(dataunit):
+        databytes = datasize * byte_unit_matrix[dataunit]
+        return databytes
+    else:
+        return None
 
 
 # Format ops sizes to/from human-readable units
@@ -731,22 +745,26 @@ def getVolumeInformation(zkhandler, pool, volume):
 
 
 def add_volume(zkhandler, pool, name, size):
-    # Add 'B' if the volume is in bytes
-    if re.match(r"^[0-9]+$", size):
-        size = "{}B".format(size)
-
     # 1. Verify the size of the volume
     pool_information = getPoolInformation(zkhandler, pool)
     size_bytes = format_bytes_fromhuman(size)
+    if size_bytes is None:
+        return (
+            False,
+            f"ERROR: Requested volume size '{size}' does not have a valid SI unit",
+        )
+
     if size_bytes >= int(pool_information["stats"]["free_bytes"]):
         return (
             False,
-            "ERROR: Requested volume size is greater than the available free space in the pool",
+            f"ERROR: Requested volume size '{format_bytes_tohuman(size_bytes)}' is greater than the available free space in the pool ('{format_bytes_tohuman(pool_information['stats']['free_bytes'])}')",
         )
 
     # 2. Create the volume
     retcode, stdout, stderr = common.run_os_command(
-        "rbd create --size {} {}/{}".format(size, pool, name)
+        "rbd create --size {} {}/{}".format(
+            format_bytes_tohuman(size_bytes), pool, name
+        )
     )
     if retcode:
         return False, 'ERROR: Failed to create RBD volume "{}": {}'.format(name, stderr)
@@ -766,7 +784,9 @@ def add_volume(zkhandler, pool, name, size):
         ]
     )
 
-    return True, 'Created RBD volume "{}/{}" ({}).'.format(pool, name, size)
+    return True, 'Created RBD volume "{}" of size "{}" in pool "{}".'.format(
+        name, format_bytes_tohuman(size_bytes), pool
+    )
 
 
 def clone_volume(zkhandler, pool, name_src, name_new):
@@ -813,28 +833,32 @@ def resize_volume(zkhandler, pool, name, size):
             name, pool
         )
 
-    # Add 'B' if the volume is in bytes
-    if re.match(r"^[0-9]+$", size):
-        size = "{}B".format(size)
-
     # 1. Verify the size of the volume
     pool_information = getPoolInformation(zkhandler, pool)
     size_bytes = format_bytes_fromhuman(size)
+    if size_bytes is None:
+        return (
+            False,
+            f"ERROR: Requested volume size '{size}' does not have a valid SI unit",
+        )
+
     if size_bytes >= int(pool_information["stats"]["free_bytes"]):
         return (
             False,
-            "ERROR: Requested volume size is greater than the available free space in the pool",
+            f"ERROR: Requested volume size '{format_bytes_tohuman(size_bytes)}' is greater than the available free space in the pool ('{format_bytes_tohuman(pool_information['stats']['free_bytes'])}')",
         )
 
     # 2. Resize the volume
     retcode, stdout, stderr = common.run_os_command(
-        "rbd resize --size {} {}/{}".format(size, pool, name)
+        "rbd resize --size {} {}/{}".format(
+            format_bytes_tohuman(size_bytes), pool, name
+        )
     )
     if retcode:
         return (
             False,
             'ERROR: Failed to resize RBD volume "{}" to size "{}" in pool "{}": {}'.format(
-                name, size, pool, stderr
+                name, format_bytes_tohuman(size_bytes), pool, stderr
             ),
         )
 
@@ -860,7 +884,7 @@ def resize_volume(zkhandler, pool, name, size):
             if target_vm_conn:
                 target_vm_conn.blockResize(
                     volume_id,
-                    format_bytes_fromhuman(size),
+                    size_bytes,
                     libvirt.VIR_DOMAIN_BLOCK_RESIZE_BYTES,
                 )
             target_lv_conn.close()
@@ -883,7 +907,7 @@ def resize_volume(zkhandler, pool, name, size):
     )
 
     return True, 'Resized RBD volume "{}" to size "{}" in pool "{}".'.format(
-        name, size, pool
+        name, format_bytes_tohuman(size_bytes), pool
     )
 
 

daemon-common/cluster.py

@@ -51,6 +51,8 @@ def getClusterHealth(zkhandler, node_list, vm_list, ceph_osd_list):
         "vm_stopped": 10,
         "osd_out": 50,
         "osd_down": 10,
+        "osd_full": 50,
+        "osd_nearfull": 10,
         "memory_overprovisioned": 50,
         "ceph_err": 50,
         "ceph_warn": 10,
@@ -110,6 +112,18 @@ def getClusterHealth(zkhandler, node_list, vm_list, ceph_osd_list):
                 f"cluster: Ceph OSD {ceph_osd['id']} in {up_texts[ceph_osd['stats']['up']].upper()} state"
             )
 
+        # Handle full or nearfull OSDs (>85%)
+        if ceph_osd["stats"]["utilization"] >= 90:
+            cluster_health_value -= health_delta_map["osd_full"]
+            cluster_health_messages.append(
+                f"cluster: Ceph OSD {ceph_osd['id']} is FULL ({ceph_osd['stats']['utilization']:.1f}% > 90%)"
+            )
+        elif ceph_osd["stats"]["utilization"] >= 85:
+            cluster_health_value -= health_delta_map["osd_nearfull"]
+            cluster_health_messages.append(
+                f"cluster: Ceph OSD {ceph_osd['id']} is NEARFULL ({ceph_osd['stats']['utilization']:.1f}% > 85%)"
+            )
+
     # Check for (n-1) overprovisioning
     #   Assume X nodes. If the total VM memory allocation (counting only running VMss) is greater than
     #   the total memory of the (n-1) smallest nodes, trigger this warning.

debian/changelog

@@ -1,3 +1,13 @@
+pvc (0.9.63-0) unstable; urgency=high
+
+  * Mentions Ganeti in the docs
+  * Increases API timeout back to 2s
+  * Adds .update-* configs to dpkg plugin
+  * Adds full/nearfull OSD warnings
+  * Improves size value handling for volumes
+
+ -- Joshua M. Boniface <joshua@boniface.me>  Fri, 28 Apr 2023 14:47:04 -0400
+
 pvc (0.9.62-0) unstable; urgency=high
 
   * [all] Adds an enhanced health checking, monitoring, and reporting system for nodes and clusters

docs/index.md

@@ -8,7 +8,7 @@
 
 ## What is PVC?
 
-PVC is a Linux KVM-based hyperconverged infrastructure (HCI) virtualization cluster solution that is fully Free Software, scalable, redundant, self-healing, self-managing, and designed for administrator simplicity. It is an alternative to other HCI solutions such as Harvester, Nutanix, and VMWare, as well as to other common virtualization stacks such as ProxMox and OpenStack.
+PVC is a Linux KVM-based hyperconverged infrastructure (HCI) virtualization cluster solution that is fully Free Software, scalable, redundant, self-healing, self-managing, and designed for administrator simplicity. It is an alternative to other HCI solutions such as Ganeti, Harvester, Nutanix, and VMWare, as well as to other common virtualization stacks such as ProxMox and OpenStack.
 
 PVC is a complete HCI solution, built from well-known and well-trusted Free Software tools, to assist an administrator in creating and managing a cluster of servers to run virtual machines, as well as self-managing several important aspects including storage failover, node failure and recovery, virtual machine failure and recovery, and network plumbing. It is designed to act consistently, reliably, and unobtrusively, letting the administrator concentrate on more important things.
 

node-daemon/plugins/dpkg

@@ -110,8 +110,8 @@ class MonitoringPluginScript(MonitoringPlugin):
 
         count_upgradable = len(list_upgradable)
 
-        # Get obsolete config files (dpkg-* or ucf-* under /etc)
-        retcode, stdout, stderr = pvc_common.run_os_command("/usr/bin/find /etc -type f -a \( -name '*.dpkg-*' -o -name '*.ucf-*' \)")
+        # Get obsolete config files (dpkg-*, ucf-*, or update-* under /etc)
+        retcode, stdout, stderr = pvc_common.run_os_command("/usr/bin/find /etc -type f -a \( -name '*.dpkg-*' -o -name '*.ucf-*' -o -name '*.update-*' \)")
 
         obsolete_conffiles = list()
         for conffile_line in stdout.split('\n'):

node-daemon/pvcnoded/Daemon.py

@@ -49,7 +49,7 @@ import re
 import json
 
 # Daemon version
-version = "0.9.62"
+version = "0.9.63"
 
 
 ##########################################################