Merge "Take into account actual compute.(ceph|lvm).ephemeral value"
diff --git a/README.rst b/README.rst
index bf4cbe2..c5606e8 100644
--- a/README.rst
+++ b/README.rst
@@ -714,6 +714,26 @@
compute:
max_concurrent_live_migrations: 1 # (1 is the default)
+Live migration with auto converge
+----------------------------------
+
+Auto converge throttles down CPU if a progress of on-going live migration is slow.
+https://docs.openstack.org/ocata/config-reference/compute/config-options.html
+
+.. code-block:: yaml
+
+ nova:
+ compute:
+ libvirt:
+ live_migration_permit_auto_converge: False # (False is the default)
+
+.. code-block:: yaml
+
+ nova:
+ controller:
+ libvirt:
+ live_migration_permit_auto_converge: False # (False is the default)
+
Enhanced logging with logging.conf
----------------------------------
@@ -772,6 +792,95 @@
<logger_name>:
level: WARNING
+Configure syslog parameters for libvirtd
+----------------------------------------
+
+To configure syslog parameters for libvirtd the below pillar structure should be used with values which are supported
+by libvirtd. These values might be known from the documentation.
+
+ nova:
+ compute:
+ libvirt:
+ logging:
+ level: 3
+ filters: '3:remote 4:event'
+ outputs: '3:syslog:libvirtd'
+ buffer_size: 64
+
+#################################################################
+#
+# Logging controls
+#
+
+# Logging level: 4 errors, 3 warnings, 2 information, 1 debug
+# basically 1 will log everything possible
+#log_level = 3
+
+# Logging filters:
+# A filter allows to select a different logging level for a given category
+# of logs
+# The format for a filter is one of:
+# x:name
+# x:+name
+# where name is a string which is matched against source file name,
+# e.g., "remote", "qemu", or "util/json", the optional "+" prefix
+# tells libvirt to log stack trace for each message matching name,
+# and x is the minimal level where matching messages should be logged:
+# 1: DEBUG
+# 2: INFO
+# 3: WARNING
+# 4: ERROR
+#
+# Multiple filter can be defined in a single @filters, they just need to be
+# separated by spaces.
+#
+# e.g. to only get warning or errors from the remote layer and only errors
+# from the event layer:
+#log_filters="3:remote 4:event"
+
+# Logging outputs:
+# An output is one of the places to save logging information
+# The format for an output can be:
+# x:stderr
+# output goes to stderr
+# x:syslog:name
+# use syslog for the output and use the given name as the ident
+# x:file:file_path
+# output to a file, with the given filepath
+# In all case the x prefix is the minimal level, acting as a filter
+# 1: DEBUG
+# 2: INFO
+# 3: WARNING
+# 4: ERROR
+#
+# Multiple output can be defined, they just need to be separated by spaces.
+# e.g. to log all warnings and errors to syslog under the libvirtd ident:
+#log_outputs="3:syslog:libvirtd"
+#
+
+# Log debug buffer size: default 64
+# The daemon keeps an internal debug log buffer which will be dumped in case
+# of crash or upon receiving a SIGUSR2 signal. This setting allows to override
+# the default buffer size in kilobytes.
+# If value is 0 or less the debug log buffer is deactivated
+#log_buffer_size = 64
+
+To configure logging parameters for qemu the below pillar structure and logging parameters should be used:
+
+ nova:
+ compute:
+ qemu:
+ logging:
+ handler: logd
+ virtlog:
+ enabled: true
+ level: 4
+ filters: '3:remote 3:event'
+ outputs: '4:syslog:virtlogd'
+ max_clients: 512
+ max_size: 2097100
+ max_backups: 2
+
Inject password to VM
---------------------
diff --git a/nova/compute.sls b/nova/compute.sls
index 0946d2e..33d30de 100644
--- a/nova/compute.sls
+++ b/nova/compute.sls
@@ -338,6 +338,26 @@
- require:
- pkg: nova_compute_packages
+{%- if compute.get('virtlog',{}).get('enabled', false) %}
+
+/etc/libvirt/virtlogd.conf:
+ file.managed:
+ - source: salt://nova/files/{{ compute.version }}/virtlogd.conf.{{ grains.os_family }}
+ - template: jinja
+ - require:
+ - pkg: nova_compute_packages
+
+/usr/sbin/virtlogd:
+ service.running:
+ - name: virtlogd
+ - enable: true
+ {%- if grains.get('noservices') %}
+ - onlyif: /bin/false
+ {%- endif %}
+ - watch:
+ - file: /etc/libvirt/virtlogd.conf
+{%- endif %}
+
virsh net-undefine default:
cmd.run:
- name: "virsh net-destroy default"
diff --git a/nova/files/mitaka/libvirtd.conf.Debian b/nova/files/mitaka/libvirtd.conf.Debian
index 4fab737..607cbcb 100644
--- a/nova/files/mitaka/libvirtd.conf.Debian
+++ b/nova/files/mitaka/libvirtd.conf.Debian
@@ -295,7 +295,9 @@
# Logging level: 4 errors, 3 warnings, 2 information, 1 debug
# basically 1 will log everything possible
#log_level = 3
-
+{%- if compute.libvirt.get('logging',{}).level is defined %}
+log_level = {{ compute.libvirt.logging.level }}
+{%- endif %}
# Logging filters:
# A filter allows to select a different logging level for a given category
# of logs
@@ -317,7 +319,9 @@
# e.g. to only get warning or errors from the remote layer and only errors
# from the event layer:
#log_filters="3:remote 4:event"
-
+{%- if compute.libvirt.get('logging',{}).filters is defined %}
+log_filters={{ compute.libvirt.logging.filters|yaml_dquote }}
+{%- endif %}
# Logging outputs:
# An output is one of the places to save logging information
# The format for an output can be:
@@ -336,16 +340,18 @@
# Multiple output can be defined, they just need to be separated by spaces.
# e.g. to log all warnings and errors to syslog under the libvirtd ident:
#log_outputs="3:syslog:libvirtd"
-#
-
+{%- if compute.libvirt.get('logging',{}).outputs is defined %}
+log_outputs={{ compute.libvirt.logging.outputs|yaml_dquote }}
+{%- endif %}
# Log debug buffer size: default 64
# The daemon keeps an internal debug log buffer which will be dumped in case
# of crash or upon receiving a SIGUSR2 signal. This setting allows to override
# the default buffer size in kilobytes.
# If value is 0 or less the debug log buffer is deactivated
#log_buffer_size = 64
-
-
+{%- if compute.libvirt.get('logging',{}).buffer_size is defined %}
+log_buffer_size = {{ compute.libvirt.logging.buffer_size }}
+{%- endif %}
##################################################################
#
# Auditing
@@ -400,4 +406,4 @@
# If set to 1, libvirtd will refuse to talk to clients that do not
# support keepalive protocol. Defaults to 0.
#
-#keepalive_required = 1
\ No newline at end of file
+#keepalive_required = 1
diff --git a/nova/files/mitaka/qemu.conf.Debian b/nova/files/mitaka/qemu.conf.Debian
index 5b039c4..a338e7d 100644
--- a/nova/files/mitaka/qemu.conf.Debian
+++ b/nova/files/mitaka/qemu.conf.Debian
@@ -484,3 +484,21 @@
"/dev/vfio/vfio",
{% endif %}
]
+
+# The backend to use for handling stdout/stderr output from
+# QEMU processes.
+#
+# 'file': QEMU writes directly to a plain file. This is the
+# historical default, but allows QEMU to inflict a
+# denial of service attack on the host by exhausting
+# filesystem space
+#
+# 'logd': QEMU writes to a pipe provided by virtlogd daemon.
+# This is the current default, providing protection
+# against denial of service by performing log file
+# rollover when a size limit is hit.
+#
+#stdio_handler = "logd"
+{%- if compute.get('qemu',{}).get('logging',{}).handler is defined %}
+stdio_handler = "{{ compute.qemu.logging.handler }}"
+{%- endif %}
diff --git a/nova/files/mitaka/virtlogd.conf.Debian b/nova/files/mitaka/virtlogd.conf.Debian
new file mode 100644
index 0000000..ecb5b2e
--- /dev/null
+++ b/nova/files/mitaka/virtlogd.conf.Debian
@@ -0,0 +1,78 @@
+{%- from "nova/map.jinja" import compute with context %}
+# Master virtlogd daemon configuration file
+#
+
+#################################################################
+#
+# Logging controls
+#
+
+# Logging level: 4 errors, 3 warnings, 2 information, 1 debug
+# basically 1 will log everything possible
+{%- if compute.get('virtlog',{}).level is defined %}
+log_level = {{ compute.virtlog.level }}
+{%- endif %}
+# Logging filters:
+# A filter allows to select a different logging level for a given category
+# of logs
+# The format for a filter is one of:
+# x:name
+# x:+name
+# where name is a string which is matched against source file name,
+# e.g., "remote", "qemu", or "util/json", the optional "+" prefix
+# tells libvirt to log stack trace for each message matching name,
+# and x is the minimal level where matching messages should be logged:
+# 1: DEBUG
+# 2: INFO
+# 3: WARNING
+# 4: ERROR
+#
+# Multiple filter can be defined in a single @filters, they just need to be
+# separated by spaces.
+#
+# e.g. to only get warning or errors from the remote layer and only errors
+# from the event layer:
+#log_filters="3:remote 4:event"
+{%- if compute.get('virtlog',{}).filters is defined %}
+log_filters={{ compute.virtlog.get('filters')|yaml_dquote }}
+{%- endif %}
+# Logging outputs:
+# An output is one of the places to save logging information
+# The format for an output can be:
+# x:stderr
+# output goes to stderr
+# x:syslog:name
+# use syslog for the output and use the given name as the ident
+# x:file:file_path
+# output to a file, with the given filepath
+# x:journald
+# ouput to the systemd journal
+# In all case the x prefix is the minimal level, acting as a filter
+# 1: DEBUG
+# 2: INFO
+# 3: WARNING
+# 4: ERROR
+#
+# Multiple output can be defined, they just need to be separated by spaces.
+# e.g. to log all warnings and errors to syslog under the virtlogd ident:
+#log_outputs="3:syslog:virtlogd"
+{%- if compute.get('virtlog',{}).outputs is defined %}
+log_outputs={{ compute.virtlog.get('outputs')|yaml_dquote }}
+{%- endif %}
+# The maximum number of concurrent client connections to allow
+# over all sockets combined.
+#max_clients = 1024
+{%- if compute.get('virtlog',{}).max_clients is defined %}
+max_clients = {{ compute.virtlog.max_clients }}
+{%- endif %}
+# Maximum file size before rolling over. Defaults to 2 MB
+#max_size = 2097152
+{%- if compute.get('virtlog',{}).max_size is defined %}
+max_size = {{ compute.virtlog.max_size }}
+{%- endif %}
+# Maximum number of backup files to keep. Defaults to 3,
+# not including the primary active file
+#max_backups = 3
+{%- if compute.get('virtlog',{}).max_backups is defined %}
+max_backups = {{ compute.virtlog.max_backups }}
+{%- endif %}
diff --git a/nova/files/mitaka/virtlogd.conf.RedHat b/nova/files/mitaka/virtlogd.conf.RedHat
new file mode 120000
index 0000000..d466306
--- /dev/null
+++ b/nova/files/mitaka/virtlogd.conf.RedHat
@@ -0,0 +1 @@
+virtlogd.conf.Debian
\ No newline at end of file
diff --git a/nova/files/newton/libvirtd.conf.Debian b/nova/files/newton/libvirtd.conf.Debian
index 4fab737..607cbcb 100644
--- a/nova/files/newton/libvirtd.conf.Debian
+++ b/nova/files/newton/libvirtd.conf.Debian
@@ -295,7 +295,9 @@
# Logging level: 4 errors, 3 warnings, 2 information, 1 debug
# basically 1 will log everything possible
#log_level = 3
-
+{%- if compute.libvirt.get('logging',{}).level is defined %}
+log_level = {{ compute.libvirt.logging.level }}
+{%- endif %}
# Logging filters:
# A filter allows to select a different logging level for a given category
# of logs
@@ -317,7 +319,9 @@
# e.g. to only get warning or errors from the remote layer and only errors
# from the event layer:
#log_filters="3:remote 4:event"
-
+{%- if compute.libvirt.get('logging',{}).filters is defined %}
+log_filters={{ compute.libvirt.logging.filters|yaml_dquote }}
+{%- endif %}
# Logging outputs:
# An output is one of the places to save logging information
# The format for an output can be:
@@ -336,16 +340,18 @@
# Multiple output can be defined, they just need to be separated by spaces.
# e.g. to log all warnings and errors to syslog under the libvirtd ident:
#log_outputs="3:syslog:libvirtd"
-#
-
+{%- if compute.libvirt.get('logging',{}).outputs is defined %}
+log_outputs={{ compute.libvirt.logging.outputs|yaml_dquote }}
+{%- endif %}
# Log debug buffer size: default 64
# The daemon keeps an internal debug log buffer which will be dumped in case
# of crash or upon receiving a SIGUSR2 signal. This setting allows to override
# the default buffer size in kilobytes.
# If value is 0 or less the debug log buffer is deactivated
#log_buffer_size = 64
-
-
+{%- if compute.libvirt.get('logging',{}).buffer_size is defined %}
+log_buffer_size = {{ compute.libvirt.logging.buffer_size }}
+{%- endif %}
##################################################################
#
# Auditing
@@ -400,4 +406,4 @@
# If set to 1, libvirtd will refuse to talk to clients that do not
# support keepalive protocol. Defaults to 0.
#
-#keepalive_required = 1
\ No newline at end of file
+#keepalive_required = 1
diff --git a/nova/files/newton/qemu.conf.Debian b/nova/files/newton/qemu.conf.Debian
index 5b039c4..a338e7d 100644
--- a/nova/files/newton/qemu.conf.Debian
+++ b/nova/files/newton/qemu.conf.Debian
@@ -484,3 +484,21 @@
"/dev/vfio/vfio",
{% endif %}
]
+
+# The backend to use for handling stdout/stderr output from
+# QEMU processes.
+#
+# 'file': QEMU writes directly to a plain file. This is the
+# historical default, but allows QEMU to inflict a
+# denial of service attack on the host by exhausting
+# filesystem space
+#
+# 'logd': QEMU writes to a pipe provided by virtlogd daemon.
+# This is the current default, providing protection
+# against denial of service by performing log file
+# rollover when a size limit is hit.
+#
+#stdio_handler = "logd"
+{%- if compute.get('qemu',{}).get('logging',{}).handler is defined %}
+stdio_handler = "{{ compute.qemu.logging.handler }}"
+{%- endif %}
diff --git a/nova/files/newton/virtlogd.conf.Debian b/nova/files/newton/virtlogd.conf.Debian
new file mode 100644
index 0000000..ecb5b2e
--- /dev/null
+++ b/nova/files/newton/virtlogd.conf.Debian
@@ -0,0 +1,78 @@
+{%- from "nova/map.jinja" import compute with context %}
+# Master virtlogd daemon configuration file
+#
+
+#################################################################
+#
+# Logging controls
+#
+
+# Logging level: 4 errors, 3 warnings, 2 information, 1 debug
+# basically 1 will log everything possible
+{%- if compute.get('virtlog',{}).level is defined %}
+log_level = {{ compute.virtlog.level }}
+{%- endif %}
+# Logging filters:
+# A filter allows to select a different logging level for a given category
+# of logs
+# The format for a filter is one of:
+# x:name
+# x:+name
+# where name is a string which is matched against source file name,
+# e.g., "remote", "qemu", or "util/json", the optional "+" prefix
+# tells libvirt to log stack trace for each message matching name,
+# and x is the minimal level where matching messages should be logged:
+# 1: DEBUG
+# 2: INFO
+# 3: WARNING
+# 4: ERROR
+#
+# Multiple filter can be defined in a single @filters, they just need to be
+# separated by spaces.
+#
+# e.g. to only get warning or errors from the remote layer and only errors
+# from the event layer:
+#log_filters="3:remote 4:event"
+{%- if compute.get('virtlog',{}).filters is defined %}
+log_filters={{ compute.virtlog.get('filters')|yaml_dquote }}
+{%- endif %}
+# Logging outputs:
+# An output is one of the places to save logging information
+# The format for an output can be:
+# x:stderr
+# output goes to stderr
+# x:syslog:name
+# use syslog for the output and use the given name as the ident
+# x:file:file_path
+# output to a file, with the given filepath
+# x:journald
+# ouput to the systemd journal
+# In all case the x prefix is the minimal level, acting as a filter
+# 1: DEBUG
+# 2: INFO
+# 3: WARNING
+# 4: ERROR
+#
+# Multiple output can be defined, they just need to be separated by spaces.
+# e.g. to log all warnings and errors to syslog under the virtlogd ident:
+#log_outputs="3:syslog:virtlogd"
+{%- if compute.get('virtlog',{}).outputs is defined %}
+log_outputs={{ compute.virtlog.get('outputs')|yaml_dquote }}
+{%- endif %}
+# The maximum number of concurrent client connections to allow
+# over all sockets combined.
+#max_clients = 1024
+{%- if compute.get('virtlog',{}).max_clients is defined %}
+max_clients = {{ compute.virtlog.max_clients }}
+{%- endif %}
+# Maximum file size before rolling over. Defaults to 2 MB
+#max_size = 2097152
+{%- if compute.get('virtlog',{}).max_size is defined %}
+max_size = {{ compute.virtlog.max_size }}
+{%- endif %}
+# Maximum number of backup files to keep. Defaults to 3,
+# not including the primary active file
+#max_backups = 3
+{%- if compute.get('virtlog',{}).max_backups is defined %}
+max_backups = {{ compute.virtlog.max_backups }}
+{%- endif %}
diff --git a/nova/files/newton/virtlogd.conf.RedHat b/nova/files/newton/virtlogd.conf.RedHat
new file mode 120000
index 0000000..d466306
--- /dev/null
+++ b/nova/files/newton/virtlogd.conf.RedHat
@@ -0,0 +1 @@
+virtlogd.conf.Debian
\ No newline at end of file
diff --git a/nova/files/ocata/libvirt.RedHat b/nova/files/ocata/libvirt.RedHat
new file mode 120000
index 0000000..f8f6638
--- /dev/null
+++ b/nova/files/ocata/libvirt.RedHat
@@ -0,0 +1 @@
+libvirt.Debian
\ No newline at end of file
diff --git a/nova/files/ocata/libvirtd.conf.Debian b/nova/files/ocata/libvirtd.conf.Debian
index 4fab737..1278aea 100644
--- a/nova/files/ocata/libvirtd.conf.Debian
+++ b/nova/files/ocata/libvirtd.conf.Debian
@@ -295,7 +295,9 @@
# Logging level: 4 errors, 3 warnings, 2 information, 1 debug
# basically 1 will log everything possible
#log_level = 3
-
+{%- if compute.libvirt.get('logging',{}).level is defined %}
+log_level = {{ compute.libvirt.logging.level }}
+{%- endif %}
# Logging filters:
# A filter allows to select a different logging level for a given category
# of logs
@@ -317,7 +319,9 @@
# e.g. to only get warning or errors from the remote layer and only errors
# from the event layer:
#log_filters="3:remote 4:event"
-
+{%- if compute.libvirt.get('logging',{}).filters is defined %}
+log_filters={{ compute.libvirt.logging.filters|yaml_dquote }}
+{%- endif %}
# Logging outputs:
# An output is one of the places to save logging information
# The format for an output can be:
@@ -336,15 +340,18 @@
# Multiple output can be defined, they just need to be separated by spaces.
# e.g. to log all warnings and errors to syslog under the libvirtd ident:
#log_outputs="3:syslog:libvirtd"
-#
-
+{%- if compute.libvirt.get('logging',{}).outputs is defined %}
+log_outputs={{ compute.libvirt.logging.outputs|yaml_dquote }}
+{%- endif %}
# Log debug buffer size: default 64
# The daemon keeps an internal debug log buffer which will be dumped in case
# of crash or upon receiving a SIGUSR2 signal. This setting allows to override
# the default buffer size in kilobytes.
# If value is 0 or less the debug log buffer is deactivated
#log_buffer_size = 64
-
+{%- if compute.libvirt.get('logging',{}).buffer_size is defined %}
+log_buffer_size = {{ compute.libvirt.logging.buffer_size }}
+{%- endif %}
##################################################################
#
@@ -400,4 +407,4 @@
# If set to 1, libvirtd will refuse to talk to clients that do not
# support keepalive protocol. Defaults to 0.
#
-#keepalive_required = 1
\ No newline at end of file
+#keepalive_required = 1
diff --git a/nova/files/ocata/libvirtd.conf.RedHat b/nova/files/ocata/libvirtd.conf.RedHat
new file mode 120000
index 0000000..2a7b101
--- /dev/null
+++ b/nova/files/ocata/libvirtd.conf.RedHat
@@ -0,0 +1 @@
+libvirtd.conf.Debian
\ No newline at end of file
diff --git a/nova/files/ocata/nova-compute.conf.Debian b/nova/files/ocata/nova-compute.conf.Debian
index 22ee1bc..aafb8d3 100644
--- a/nova/files/ocata/nova-compute.conf.Debian
+++ b/nova/files/ocata/nova-compute.conf.Debian
@@ -6384,6 +6384,9 @@
# * live_migration_permit_post_copy
# (boolean value)
#live_migration_permit_auto_converge=false
+{%- if compute.libvirt.live_migration_permit_auto_converge is defined %}
+live_migration_permit_auto_converge={{ compute.libvirt.live_migration_permit_auto_converge|lower }}
+{%- endif %}
#
# Determine the snapshot image format when sending to the image service.
diff --git a/nova/files/ocata/nova-controller.conf.Debian b/nova/files/ocata/nova-controller.conf.Debian
index e758573..42a7a0e 100644
--- a/nova/files/ocata/nova-controller.conf.Debian
+++ b/nova/files/ocata/nova-controller.conf.Debian
@@ -6345,6 +6345,9 @@
# * live_migration_permit_post_copy
# (boolean value)
#live_migration_permit_auto_converge=false
+{%- if controller.get('libvirt', {}).live_migration_permit_auto_converge is defined %}
+live_migration_permit_auto_converge={{ controller.libvirt.live_migration_permit_auto_converge|lower }}
+{%- endif %}
#
# Determine the snapshot image format when sending to the image service.
diff --git a/nova/files/ocata/qemu.conf.Debian b/nova/files/ocata/qemu.conf.Debian
index 6212dda..cb20491 100644
--- a/nova/files/ocata/qemu.conf.Debian
+++ b/nova/files/ocata/qemu.conf.Debian
@@ -497,3 +497,21 @@
"/dev/vfio/vfio",
{% endif %}
]
+
+# The backend to use for handling stdout/stderr output from
+# QEMU processes.
+#
+# 'file': QEMU writes directly to a plain file. This is the
+# historical default, but allows QEMU to inflict a
+# denial of service attack on the host by exhausting
+# filesystem space
+#
+# 'logd': QEMU writes to a pipe provided by virtlogd daemon.
+# This is the current default, providing protection
+# against denial of service by performing log file
+# rollover when a size limit is hit.
+#
+#stdio_handler = "logd"
+{%- if compute.get('qemu',{}).get('logging',{}).handler is defined %}
+stdio_handler = "{{ compute.qemu.logging.handler }}"
+{%- endif %}
diff --git a/nova/files/ocata/qemu.conf.RedHat b/nova/files/ocata/qemu.conf.RedHat
new file mode 120000
index 0000000..1d23f19
--- /dev/null
+++ b/nova/files/ocata/qemu.conf.RedHat
@@ -0,0 +1 @@
+qemu.conf.Debian
\ No newline at end of file
diff --git a/nova/files/ocata/virtlogd.conf.Debian b/nova/files/ocata/virtlogd.conf.Debian
new file mode 100644
index 0000000..ecb5b2e
--- /dev/null
+++ b/nova/files/ocata/virtlogd.conf.Debian
@@ -0,0 +1,78 @@
+{%- from "nova/map.jinja" import compute with context %}
+# Master virtlogd daemon configuration file
+#
+
+#################################################################
+#
+# Logging controls
+#
+
+# Logging level: 4 errors, 3 warnings, 2 information, 1 debug
+# basically 1 will log everything possible
+{%- if compute.get('virtlog',{}).level is defined %}
+log_level = {{ compute.virtlog.level }}
+{%- endif %}
+# Logging filters:
+# A filter allows to select a different logging level for a given category
+# of logs
+# The format for a filter is one of:
+# x:name
+# x:+name
+# where name is a string which is matched against source file name,
+# e.g., "remote", "qemu", or "util/json", the optional "+" prefix
+# tells libvirt to log stack trace for each message matching name,
+# and x is the minimal level where matching messages should be logged:
+# 1: DEBUG
+# 2: INFO
+# 3: WARNING
+# 4: ERROR
+#
+# Multiple filter can be defined in a single @filters, they just need to be
+# separated by spaces.
+#
+# e.g. to only get warning or errors from the remote layer and only errors
+# from the event layer:
+#log_filters="3:remote 4:event"
+{%- if compute.get('virtlog',{}).filters is defined %}
+log_filters={{ compute.virtlog.get('filters')|yaml_dquote }}
+{%- endif %}
+# Logging outputs:
+# An output is one of the places to save logging information
+# The format for an output can be:
+# x:stderr
+# output goes to stderr
+# x:syslog:name
+# use syslog for the output and use the given name as the ident
+# x:file:file_path
+# output to a file, with the given filepath
+# x:journald
+# ouput to the systemd journal
+# In all case the x prefix is the minimal level, acting as a filter
+# 1: DEBUG
+# 2: INFO
+# 3: WARNING
+# 4: ERROR
+#
+# Multiple output can be defined, they just need to be separated by spaces.
+# e.g. to log all warnings and errors to syslog under the virtlogd ident:
+#log_outputs="3:syslog:virtlogd"
+{%- if compute.get('virtlog',{}).outputs is defined %}
+log_outputs={{ compute.virtlog.get('outputs')|yaml_dquote }}
+{%- endif %}
+# The maximum number of concurrent client connections to allow
+# over all sockets combined.
+#max_clients = 1024
+{%- if compute.get('virtlog',{}).max_clients is defined %}
+max_clients = {{ compute.virtlog.max_clients }}
+{%- endif %}
+# Maximum file size before rolling over. Defaults to 2 MB
+#max_size = 2097152
+{%- if compute.get('virtlog',{}).max_size is defined %}
+max_size = {{ compute.virtlog.max_size }}
+{%- endif %}
+# Maximum number of backup files to keep. Defaults to 3,
+# not including the primary active file
+#max_backups = 3
+{%- if compute.get('virtlog',{}).max_backups is defined %}
+max_backups = {{ compute.virtlog.max_backups }}
+{%- endif %}
diff --git a/nova/files/ocata/virtlogd.conf.RedHat b/nova/files/ocata/virtlogd.conf.RedHat
new file mode 120000
index 0000000..d466306
--- /dev/null
+++ b/nova/files/ocata/virtlogd.conf.RedHat
@@ -0,0 +1 @@
+virtlogd.conf.Debian
\ No newline at end of file
diff --git a/nova/files/pike/libvirt.RedHat b/nova/files/pike/libvirt.RedHat
new file mode 120000
index 0000000..f8f6638
--- /dev/null
+++ b/nova/files/pike/libvirt.RedHat
@@ -0,0 +1 @@
+libvirt.Debian
\ No newline at end of file
diff --git a/nova/files/pike/libvirtd.conf.Debian b/nova/files/pike/libvirtd.conf.Debian
index 4fab737..6f4afb9 100644
--- a/nova/files/pike/libvirtd.conf.Debian
+++ b/nova/files/pike/libvirtd.conf.Debian
@@ -295,7 +295,9 @@
# Logging level: 4 errors, 3 warnings, 2 information, 1 debug
# basically 1 will log everything possible
#log_level = 3
-
+{%- if compute.libvirt.get('logging',{}).level is defined %}
+log_level = {{ compute.libvirt.logging.level }}
+{%- endif %}
# Logging filters:
# A filter allows to select a different logging level for a given category
# of logs
@@ -317,6 +319,9 @@
# e.g. to only get warning or errors from the remote layer and only errors
# from the event layer:
#log_filters="3:remote 4:event"
+{%- if compute.libvirt.get('logging',{}).filters is defined %}
+log_filters={{ compute.libvirt.logging.filters|yaml_dquote }}
+{%- endif %}
# Logging outputs:
# An output is one of the places to save logging information
@@ -337,6 +342,9 @@
# e.g. to log all warnings and errors to syslog under the libvirtd ident:
#log_outputs="3:syslog:libvirtd"
#
+{%- if compute.libvirt.get('logging',{}).outputs is defined %}
+log_outputs={{ compute.libvirt.logging.outputs|yaml_dquote }}
+{%- endif %}
# Log debug buffer size: default 64
# The daemon keeps an internal debug log buffer which will be dumped in case
@@ -344,7 +352,9 @@
# the default buffer size in kilobytes.
# If value is 0 or less the debug log buffer is deactivated
#log_buffer_size = 64
-
+{%- if compute.libvirt.get('logging',{}).buffer_size is defined %}
+log_buffer_size = {{ compute.libvirt.logging.buffer_size }}
+{%- endif %}
##################################################################
#
@@ -400,4 +410,4 @@
# If set to 1, libvirtd will refuse to talk to clients that do not
# support keepalive protocol. Defaults to 0.
#
-#keepalive_required = 1
\ No newline at end of file
+#keepalive_required = 1
diff --git a/nova/files/pike/libvirtd.conf.RedHat b/nova/files/pike/libvirtd.conf.RedHat
new file mode 120000
index 0000000..2a7b101
--- /dev/null
+++ b/nova/files/pike/libvirtd.conf.RedHat
@@ -0,0 +1 @@
+libvirtd.conf.Debian
\ No newline at end of file
diff --git a/nova/files/pike/nova-compute.conf.Debian b/nova/files/pike/nova-compute.conf.Debian
index 39a2133..e604d72 100644
--- a/nova/files/pike/nova-compute.conf.Debian
+++ b/nova/files/pike/nova-compute.conf.Debian
@@ -6415,6 +6415,9 @@
# * live_migration_permit_post_copy
# (boolean value)
#live_migration_permit_auto_converge=false
+{%- if compute.libvirt.live_migration_permit_auto_converge is defined %}
+live_migration_permit_auto_converge={{ compute.libvirt.live_migration_permit_auto_converge|lower }}
+{%- endif %}
#
# Determine the snapshot image format when sending to the image service.
diff --git a/nova/files/pike/nova-controller.conf.Debian b/nova/files/pike/nova-controller.conf.Debian
index b1b18d7..96e1d0f 100644
--- a/nova/files/pike/nova-controller.conf.Debian
+++ b/nova/files/pike/nova-controller.conf.Debian
@@ -6349,6 +6349,9 @@
# * live_migration_permit_post_copy
# (boolean value)
#live_migration_permit_auto_converge=false
+{%- if controller.get('libvirt', {}).live_migration_permit_auto_converge is defined %}
+live_migration_permit_auto_converge={{ controller.libvirt.live_migration_permit_auto_converge|lower }}
+{%- endif %}
#
# Determine the snapshot image format when sending to the image service.
diff --git a/nova/files/pike/qemu.conf.Debian b/nova/files/pike/qemu.conf.Debian
index 6212dda..cb20491 100644
--- a/nova/files/pike/qemu.conf.Debian
+++ b/nova/files/pike/qemu.conf.Debian
@@ -497,3 +497,21 @@
"/dev/vfio/vfio",
{% endif %}
]
+
+# The backend to use for handling stdout/stderr output from
+# QEMU processes.
+#
+# 'file': QEMU writes directly to a plain file. This is the
+# historical default, but allows QEMU to inflict a
+# denial of service attack on the host by exhausting
+# filesystem space
+#
+# 'logd': QEMU writes to a pipe provided by virtlogd daemon.
+# This is the current default, providing protection
+# against denial of service by performing log file
+# rollover when a size limit is hit.
+#
+#stdio_handler = "logd"
+{%- if compute.get('qemu',{}).get('logging',{}).handler is defined %}
+stdio_handler = "{{ compute.qemu.logging.handler }}"
+{%- endif %}
diff --git a/nova/files/pike/qemu.conf.RedHat b/nova/files/pike/qemu.conf.RedHat
new file mode 120000
index 0000000..1d23f19
--- /dev/null
+++ b/nova/files/pike/qemu.conf.RedHat
@@ -0,0 +1 @@
+qemu.conf.Debian
\ No newline at end of file
diff --git a/nova/files/pike/virtlogd.conf.Debian b/nova/files/pike/virtlogd.conf.Debian
new file mode 100644
index 0000000..ecb5b2e
--- /dev/null
+++ b/nova/files/pike/virtlogd.conf.Debian
@@ -0,0 +1,78 @@
+{%- from "nova/map.jinja" import compute with context %}
+# Master virtlogd daemon configuration file
+#
+
+#################################################################
+#
+# Logging controls
+#
+
+# Logging level: 4 errors, 3 warnings, 2 information, 1 debug
+# basically 1 will log everything possible
+{%- if compute.get('virtlog',{}).level is defined %}
+log_level = {{ compute.virtlog.level }}
+{%- endif %}
+# Logging filters:
+# A filter allows to select a different logging level for a given category
+# of logs
+# The format for a filter is one of:
+# x:name
+# x:+name
+# where name is a string which is matched against source file name,
+# e.g., "remote", "qemu", or "util/json", the optional "+" prefix
+# tells libvirt to log stack trace for each message matching name,
+# and x is the minimal level where matching messages should be logged:
+# 1: DEBUG
+# 2: INFO
+# 3: WARNING
+# 4: ERROR
+#
+# Multiple filter can be defined in a single @filters, they just need to be
+# separated by spaces.
+#
+# e.g. to only get warning or errors from the remote layer and only errors
+# from the event layer:
+#log_filters="3:remote 4:event"
+{%- if compute.get('virtlog',{}).filters is defined %}
+log_filters={{ compute.virtlog.get('filters')|yaml_dquote }}
+{%- endif %}
+# Logging outputs:
+# An output is one of the places to save logging information
+# The format for an output can be:
+# x:stderr
+# output goes to stderr
+# x:syslog:name
+# use syslog for the output and use the given name as the ident
+# x:file:file_path
+# output to a file, with the given filepath
+# x:journald
+# ouput to the systemd journal
+# In all case the x prefix is the minimal level, acting as a filter
+# 1: DEBUG
+# 2: INFO
+# 3: WARNING
+# 4: ERROR
+#
+# Multiple output can be defined, they just need to be separated by spaces.
+# e.g. to log all warnings and errors to syslog under the virtlogd ident:
+#log_outputs="3:syslog:virtlogd"
+{%- if compute.get('virtlog',{}).outputs is defined %}
+log_outputs={{ compute.virtlog.get('outputs')|yaml_dquote }}
+{%- endif %}
+# The maximum number of concurrent client connections to allow
+# over all sockets combined.
+#max_clients = 1024
+{%- if compute.get('virtlog',{}).max_clients is defined %}
+max_clients = {{ compute.virtlog.max_clients }}
+{%- endif %}
+# Maximum file size before rolling over. Defaults to 2 MB
+#max_size = 2097152
+{%- if compute.get('virtlog',{}).max_size is defined %}
+max_size = {{ compute.virtlog.max_size }}
+{%- endif %}
+# Maximum number of backup files to keep. Defaults to 3,
+# not including the primary active file
+#max_backups = 3
+{%- if compute.get('virtlog',{}).max_backups is defined %}
+max_backups = {{ compute.virtlog.max_backups }}
+{%- endif %}
diff --git a/nova/files/pike/virtlogd.conf.RedHat b/nova/files/pike/virtlogd.conf.RedHat
new file mode 120000
index 0000000..d466306
--- /dev/null
+++ b/nova/files/pike/virtlogd.conf.RedHat
@@ -0,0 +1 @@
+virtlogd.conf.Debian
\ No newline at end of file
diff --git a/nova/map.jinja b/nova/map.jinja
index c2334fc..4d06861 100644
--- a/nova/map.jinja
+++ b/nova/map.jinja
@@ -187,5 +187,16 @@
'services_failed_critical_threshold_percent': 0.6,
'computes_failed_warning_threshold_percent': 0.25,
'computes_failed_critical_threshold_percent': 0.5,
+ 'allocated_vcpus_minor_threshold': 0.9,
+ 'allocated_vcpus_major_threshold': 0.97,
+ 'allocated_ram_minor_threshold': 0.9,
+ 'allocated_ram_major_threshold': 0.97,
+ 'allocated_disk_minor_threshold': 0.9,
+ 'allocated_disk_major_threshold': 0.97,
+ 'ram_major_threshold': 0.85,
+ 'ram_critical_threshold': 0.95,
+ 'disk_major_threshold': 0.85,
+ 'disk_critical_threshold': 0.95,
+ 'endpoint_failed_major_threshold': 0.5,
},
}, grain='os_family', merge=salt['pillar.get']('nova:monitoring')) %}
diff --git a/nova/meta/prometheus.yml b/nova/meta/prometheus.yml
index 9029265..f701acd 100644
--- a/nova/meta/prometheus.yml
+++ b/nova/meta/prometheus.yml
@@ -34,6 +34,7 @@
{%- set major_threshold = monitoring.services_failed_critical_threshold_percent|float %}
{%- set minor_compute_threshold = monitoring.computes_failed_warning_threshold_percent|float %}
{%- set major_compute_threshold = monitoring.computes_failed_critical_threshold_percent|float %}
+{%- set major_endpoint_threshold = monitoring.endpoint_failed_major_threshold|float %}
{% raw %}
NovaAPIOutage:
if: >-
@@ -66,6 +67,29 @@
summary: "Host nova-api endpoint is not accessible"
description: >-
The host nova-api endpoint on the {{ $labels.host }} node is not accessible for at least 2 minutes.
+{%- endraw %}
+ NovaAPIServiceDownMajor:
+ if: >-
+ count(http_response_status{name=~"nova-api"} == 0) >= count(http_response_status{name=~"nova-api"}) * {{ major_endpoint_threshold }}
+ for: 2m
+ labels:
+ severity: major
+ service: nova
+ annotations:
+ summary: "{{major_endpoint_threshold * 100}}% of host nova-api endpoints are not accessible"
+ description: >-
+ {% raw %}{{ $value }} host nova-api endpoints are not accessible for at least 2 minutes (at least {% endraw %}{{major_endpoint_threshold * 100}}{% raw %}%).
+ NovaAPIServiceOutage:
+ if: >-
+ count(http_response_status{name=~"nova-api"} == 0) == count(http_response_status{name=~"nova-api"})
+ for: 2m
+ labels:
+ severity: critical
+ service: nova
+ annotations:
+ summary: "Host nova-api outage"
+ description: >-
+ All available host nova-api endpoints are not accessible for at least 2 minutes.
NovaServiceDown:
if: >-
openstack_nova_service_state == 0
@@ -79,7 +103,7 @@
{%- endraw %}
NovaServicesDownMinor:
if: >-
- count(openstack_nova_service_state{binary!~"nova-compute"} == 0) by (binary) >= on (binary) count(openstack_nova_service_state{binary!~"nova-compute"}) by (binary) * {{minor_threshold}} and count(openstack_nova_service_state{binary!~"nova-compute"} == 0) by (binary) < on (binary) count(openstack_nova_service_state{binary!~"nova-compute"}) by (binary) * {{major_threshold}}
+ count(openstack_nova_service_state{binary!~"nova-compute"} == 0) by (binary) >= on (binary) count(openstack_nova_service_state{binary!~"nova-compute"}) by (binary) * {{minor_threshold}}
labels:
severity: minor
service: nova
@@ -89,7 +113,7 @@
{{ $value }} {{ $labels.binary }} services are down {%- endraw %}(at least {{minor_threshold * 100}}%).
NovaComputeServicesDownMinor:
if: >-
- count(openstack_nova_service_state{binary="nova-compute"} == 0) >= count(openstack_nova_service_state{binary="nova-compute"}) * {{minor_compute_threshold}} and count(openstack_nova_service_state{binary="nova-compute"} == 0) < count(openstack_nova_service_state{binary="nova-compute"}) * {{major_compute_threshold}}
+ count(openstack_nova_service_state{binary="nova-compute"} == 0) >= count(openstack_nova_service_state{binary="nova-compute"}) * {{minor_compute_threshold}}
labels:
severity: minor
service: nova
@@ -127,99 +151,375 @@
summary: "{{ $labels.binary }} service outage"
description: >-
All {{ $labels.binary }} services are down.
- NovaTotalFreeVCPUsLow:
+{%- endraw -%}
+
+{%- set cpu_ratio = controller.get('cpu_allocation_ratio', 16.0) %}
+{%- set ram_ratio = controller.get('ram_allocation_ratio', 1.5) %}
+{%- set disk_ratio = controller.get('disk_allocation_ratio', 1.0) -%}
+
+{%- set alloc_cpu_minor_threshold = monitoring.allocated_vcpus_minor_threshold|float %}
+{%- set alloc_cpu_major_threshold = monitoring.allocated_vcpus_major_threshold|float %}
+{%- set alloc_ram_minor_threshold = monitoring.allocated_ram_minor_threshold|float %}
+{%- set alloc_ram_major_threshold = monitoring.allocated_ram_major_threshold|float %}
+{%- set alloc_disk_minor_threshold = monitoring.allocated_disk_minor_threshold|float %}
+{%- set alloc_disk_major_threshold = monitoring.allocated_disk_major_threshold|float -%}
+
+{%- set ram_major_threshold = monitoring.ram_major_threshold|float %}
+{%- set ram_critical_threshold = monitoring.ram_critical_threshold|float %}
+{%- set disk_major_threshold = monitoring.disk_major_threshold|float %}
+{%- set disk_critical_threshold = monitoring.disk_critical_threshold|float -%}
+
+ NovaHypervisorAllocatedVCPUsFullMinor:
if: >-
- (100.0 * openstack_nova_total_free_vcpus) / (openstack_nova_total_free_vcpus + openstack_nova_total_used_vcpus) < 10.0
- for: 1m
+ openstack_nova_used_vcpus >= openstack_nova_vcpus * {{ cpu_ratio }} * {{ alloc_cpu_minor_threshold }}
labels:
- severity: warning
+ severity: minor
service: nova
annotations:
- summary: "VCPU low limit for new instances"
- description: >-
- VPCU low limit for 1 minutes
- NovaTotalFreeMemoryLow:
+ summary: "{{ alloc_cpu_minor_threshold * 100 }}% of hypervisor VCPUs were allocated"
+ description: "{% raw %}{{ $value }} VCPUs on the {{ $labels.hostname }}{% endraw %} node were allocated (at least {{ alloc_cpu_minor_threshold * 100 }}%)."
+ NovaHypervisorAllocatedVCPUsFullMajor:
if: >-
- (100.0 * openstack_nova_total_free_ram) / (openstack_nova_total_free_ram + openstack_nova_total_used_ram) < 10.0
- for: 1m
+ openstack_nova_used_vcpus >= openstack_nova_vcpus * {{ cpu_ratio }} * {{ alloc_cpu_major_threshold }}
labels:
- severity: warning
+ severity: major
service: nova
annotations:
- summary: "Memory low limit for new instances"
- description: >-
- Memory low limit for 1 minutes
- NovaTotalFreeVCPUsShortage:
+ summary: "{{ alloc_cpu_major_threshold * 100 }}% of hypervisor VCPUs were allocated"
+ description: "{% raw %}{{ $value }} VCPUs on the {{ $labels.hostname }}{% endraw %} node were allocated (at least {{ alloc_cpu_major_threshold * 100 }}%)."
+ NovaHypervisorAllocatedVCPUsFullCritical:
if: >-
- (100.0 * openstack_nova_total_free_vcpus) / (openstack_nova_total_free_vcpus + openstack_nova_total_used_vcpus) < 2.0
- for: 1m
+ openstack_nova_used_vcpus >= openstack_nova_vcpus * {{ cpu_ratio }}
labels:
severity: critical
service: nova
annotations:
- summary: "VCPU shortage for new instances"
- description: >-
- VPCU shortage for 1 minutes
- NovaTotalFreeMemoryShortage:
+ summary: "No VCPUs available for allocation"
+ description: "All available VCPUs on the {% raw %}{{ $labels.hostname }}{% endraw %} node were allocated."
+ NovaHypervisorAllocatedMemoryFullMinor:
if: >-
- (100.0 * openstack_nova_total_free_ram) / (openstack_nova_total_free_ram + openstack_nova_total_used_ram) < 2.0
- for: 1m
+ openstack_nova_ram - openstack_nova_free_ram >= openstack_nova_ram * {{ ram_ratio }} * {{ alloc_ram_minor_threshold }}
+ labels:
+ severity: minor
+ service: nova
+ annotations:
+ summary: "{{ alloc_ram_minor_threshold * 100 }}% of hypervisor RAM was allocated"
+ description: "{% raw %}{{ $value }}MB of RAM on the {{ $labels.hostname }}{% endraw %} node was allocated (at least {{ alloc_ram_minor_threshold * 100 }}%)."
+ NovaHypervisorAllocatedMemoryFullMajor:
+ if: >-
+ openstack_nova_ram - openstack_nova_free_ram >= openstack_nova_ram * {{ ram_ratio }} * {{ alloc_ram_major_threshold }}
+ labels:
+ severity: major
+ service: nova
+ annotations:
+ summary: "{{ alloc_ram_major_threshold * 100 }}% of hypervisor RAM was allocated"
+ description: "{% raw %}{{ $value }}MB of RAM on the {{ $labels.hostname }}{% endraw %} node was allocated (at least {{ alloc_ram_major_threshold * 100 }}%)."
+ NovaHypervisorAllocatedMemoryFullCritical:
+ if: >-
+ openstack_nova_ram - openstack_nova_free_ram >= openstack_nova_ram * {{ ram_ratio }}
labels:
severity: critical
service: nova
annotations:
- summary: "Memory shortage for new instances"
- description: >-
- Memory shortage for 1 minutes
- NovaAggregatesFreeVCPUsLow:
+ summary: "No RAM available for allocation"
+ description: "All available RAM on the {% raw %}{{ $labels.hostname }}{% endraw %} node was allocated."
+ NovaHypervisorMemoryFullMajor:
if: >-
- (100.0 * openstack_nova_aggregate_free_vcpus) / (openstack_nova_aggregate_free_vcpus + openstack_nova_aggregate_used_vcpus) < 10.0
- for: 1m
+ openstack_nova_used_ram >= openstack_nova_ram * {{ ram_major_threshold }}
labels:
- severity: warning
+ severity: major
service: nova
- aggregate: "{{ $labels.aggregate }}"
annotations:
- summary: "VCPU low limit for new instances on aggregate {{ $labels.aggregate }}"
- description: >-
- VPCU low limit for 1 minutes on aggregate {{ $labels.aggregate }}
- NovaAggregatesFreeMemoryLow:
+ summary: "{{ ram_major_threshold * 100 }}% of hypervisor RAM was used"
+ description: "{% raw %}{{ $value }}MB of RAM on the {{ $labels.hostname }}{% endraw %} node was used (at least {{ ram_major_threshold * 100 }}%)."
+ NovaHypervisorMemoryFullCritical:
if: >-
- (100.0 * openstack_nova_aggregate_free_ram) / (openstack_nova_aggregate_free_ram + openstack_nova_aggregate_used_ram) < 10.0
- for: 1m
- labels:
- severity: warning
- service: nova
- aggregate: "{{ $labels.aggregate }}"
- annotations:
- summary: "Memory low limit for new instances on aggregate {{ $labels.aggregate }}"
- description: >-
- Memory low limit for 1 minutes on aggregate {{ $labels.aggregate }}
- NovaAggregatesFreeVCPUsShortage:
- if: >-
- (100.0 * openstack_nova_aggregate_free_vcpus) / (openstack_nova_aggregate_free_vcpus + openstack_nova_aggregate_used_vcpus) < 2.0
- for: 1m
+ openstack_nova_used_ram >= openstack_nova_ram * {{ ram_critical_threshold }}
labels:
severity: critical
service: nova
- aggregate: "{{ $labels.aggregate }}"
annotations:
- summary: "VCPU shortage for new instances on aggregate {{ $labels.aggregate }}"
- description: >-
- VPCU shortage for 1 minutes on aggregate {{ $labels.aggregate }}
- NovaAggregatesFreeMemoryShortage:
+ summary: "{{ ram_critical_threshold * 100 }}% of hypervisor RAM was used"
+ description: "{% raw %}{{ $value }}MB of RAM on the {{ $labels.hostname }}{% endraw %} node was used (at least {{ ram_critical_threshold * 100 }}%)."
+ NovaHypervisorAllocatedDiskFullMinor:
if: >-
- (100.0 * openstack_nova_aggregate_free_ram) / (openstack_nova_aggregate_free_ram + openstack_nova_aggregate_used_ram) < 2.0
- for: 1m
+ openstack_nova_disk - openstack_nova_free_disk >= openstack_nova_disk * {{ disk_ratio }} * {{ alloc_disk_minor_threshold }}
+ labels:
+ severity: minor
+ service: nova
+ annotations:
+ summary: "{{ alloc_disk_minor_threshold * 100 }}% of hypervisor disk space was allocated"
+ description: "{% raw %}{{ $value }}GB of disk space on the {{ $labels.hostname }}{% endraw %} node was allocated (at least {{ alloc_disk_minor_threshold * 100 }}%)."
+ NovaHypervisorAllocatedDiskFullMajor:
+ if: >-
+ openstack_nova_disk - openstack_nova_free_disk >= openstack_nova_disk * {{ disk_ratio }} * {{ alloc_disk_major_threshold }}
+ labels:
+ severity: major
+ service: nova
+ annotations:
+ summary: "{{ alloc_disk_major_threshold * 100 }}% of hypervisor disk space was allocated"
+ description: "{% raw %}{{ $value }}GB of disk space on the {{ $labels.hostname }}{% endraw %} node was allocated (at least {{ alloc_disk_major_threshold * 100 }}%)."
+ NovaHypervisorAllocatedDiskFullCritical:
+ if: >-
+ openstack_nova_disk - openstack_nova_free_disk >= openstack_nova_disk * {{ disk_ratio }}
labels:
severity: critical
service: nova
- aggregate: "{{ $labels.aggregate }}"
annotations:
- summary: "Memory shortage for new instances on aggregate {{ $labels.aggregate }}"
- description: >-
- Memory shortage for 1 minutes on aggregate {{ $labels.aggregate }}
-{%- endraw %}
+ summary: "No disk space available for allocation"
+ description: "All available disk space on the {% raw %}{{ $labels.hostname }}{% endraw %} node was allocated."
+ NovaHypervisorDiskFullMajor:
+ if: >-
+ openstack_nova_used_disk >= openstack_nova_disk * {{ disk_major_threshold }}
+ labels:
+ severity: major
+ service: nova
+ annotations:
+ summary: "{{ disk_major_threshold * 100 }}% of hypervisor disk space was used"
+ description: "{% raw %}{{ $value }}GB of disk space on the {{ $labels.hostname }}{% endraw %} node was used (at least {{ disk_major_threshold * 100 }}%)."
+ NovaHypervisorDiskFullCritical:
+ if: >-
+ openstack_nova_used_disk >= openstack_nova_disk * {{ disk_critical_threshold }}
+ labels:
+ severity: critical
+ service: nova
+ annotations:
+ summary: "{{ disk_critical_threshold * 100 }}% of hypervisor disk space was used"
+ description: "{% raw %}{{ $value }}GB of disk space on the {{ $labels.hostname }}{% endraw %} node was used (at least {{ disk_critical_threshold * 100 }}%)."
+ NovaAggregateAllocatedVCPUsFullMinor:
+ if: >-
+ openstack_nova_aggregate_used_vcpus >= openstack_nova_aggregate_vcpus * {{ cpu_ratio }} * {{ alloc_cpu_minor_threshold }}
+ labels:
+ severity: minor
+ service: nova
+ annotations:
+ summary: "{{ alloc_cpu_minor_threshold * 100 }}% of aggregate VCPUs were allocated"
+ description: "{% raw %}{{ $value }} VCPUs on the {{ $labels.aggregate }}{% endraw %} aggregate were allocated (at least {{ alloc_cpu_minor_threshold * 100 }}%)."
+ NovaAggregateAllocatedVCPUsFullMajor:
+ if: >-
+ openstack_nova_aggregate_used_vcpus >= openstack_nova_aggregate_vcpus * {{ cpu_ratio }} * {{ alloc_cpu_major_threshold }}
+ labels:
+ severity: major
+ service: nova
+ annotations:
+ summary: "{{ alloc_cpu_major_threshold * 100 }}% of aggregate VCPUs were allocated"
+ description: "{% raw %}{{ $value }} VCPUs on the {{ $labels.aggregate }}{% endraw %} aggregate were allocated (at least {{ alloc_cpu_major_threshold * 100 }}%)."
+ NovaAggregateAllocatedVCPUsFullCritical:
+ if: >-
+ openstack_nova_aggregate_used_vcpus >= openstack_nova_aggregate_vcpus * {{ cpu_ratio }}
+ labels:
+ severity: critical
+ service: nova
+ annotations:
+ summary: "No VCPUs available for allocation"
+ description: "All available VCPUs on the {% raw %}{{ $labels.aggregate }}{% endraw %} aggregate were allocated."
+ NovaAggregateAllocatedMemoryFullMinor:
+ if: >-
+ openstack_nova_aggregate_ram - openstack_nova_aggregate_free_ram >= openstack_nova_aggregate_ram * {{ ram_ratio }} * {{ alloc_ram_minor_threshold }}
+ labels:
+ severity: minor
+ service: nova
+ annotations:
+ summary: "{{ alloc_ram_minor_threshold * 100 }}% of aggregate RAM was allocated"
+ description: "{% raw %}{{ $value }}MB of RAM on the {{ $labels.aggregate }}{% endraw %} aggregate was allocated (at least {{ alloc_ram_minor_threshold * 100 }}%)."
+ NovaAggregateAllocatedMemoryFullMajor:
+ if: >-
+ openstack_nova_aggregate_ram - openstack_nova_aggregate_free_ram >= openstack_nova_aggregate_ram * {{ ram_ratio }} * {{ alloc_ram_major_threshold }}
+ labels:
+ severity: major
+ service: nova
+ annotations:
+ summary: "{{ alloc_ram_major_threshold * 100 }}% of aggregate RAM was allocated"
+ description: "{% raw %}{{ $value }}MB of RAM on the {{ $labels.aggregate }}{% endraw %} aggregate was allocated (at least {{ alloc_ram_major_threshold * 100 }}%)."
+ NovaAggregateAllocatedMemoryFullCritical:
+ if: >-
+ openstack_nova_aggregate_ram - openstack_nova_aggregate_free_ram >= openstack_nova_aggregate_ram * {{ ram_ratio }}
+ labels:
+ severity: critical
+ service: nova
+ annotations:
+ summary: "No RAM available for allocation"
+ description: "All available RAM on the {% raw %}{{ $labels.aggregate }}{% endraw %} aggregate was allocated."
+ NovaAggregateMemoryFullMajor:
+ if: >-
+ openstack_nova_aggregate_used_ram >= openstack_nova_aggregate_ram * {{ ram_major_threshold }}
+ labels:
+ severity: major
+ service: nova
+ annotations:
+ summary: "{{ ram_major_threshold * 100 }}% of aggregate RAM was used"
+ description: "{% raw %}{{ $value }}MB of RAM on the {{ $labels.aggregate }}{% endraw %} aggregate was used (at least {{ ram_major_threshold * 100 }}%)."
+ NovaAggregateMemoryFullCritical:
+ if: >-
+ openstack_nova_aggregate_used_ram >= openstack_nova_aggregate_ram * {{ ram_critical_threshold }}
+ labels:
+ severity: critical
+ service: nova
+ annotations:
+ summary: "{{ ram_critical_threshold * 100 }}% of aggregate RAM was used"
+ description: "{% raw %}{{ $value }}MB of RAM on the {{ $labels.aggregate }}{% endraw %} aggregate was used (at least {{ ram_critical_threshold * 100 }}%)."
+ NovaAggregateAllocatedDiskFullMinor:
+ if: >-
+ openstack_nova_aggregate_disk - openstack_nova_aggregate_free_disk >= openstack_nova_aggregate_disk * {{ disk_ratio }} * {{ alloc_disk_minor_threshold }}
+ labels:
+ severity: minor
+ service: nova
+ annotations:
+ summary: "{{ alloc_disk_minor_threshold * 100 }}% of aggregate disk space was allocated"
+ description: "{% raw %}{{ $value }}GB of disk space on the {{ $labels.aggregate }}{% endraw %} aggregate was allocated (at least {{ alloc_disk_minor_threshold * 100 }}%)."
+ NovaAggregateAllocatedDiskFullMajor:
+ if: >-
+ openstack_nova_aggregate_disk - openstack_nova_aggregate_free_disk >= openstack_nova_aggregate_disk * {{ disk_ratio }} * {{ alloc_disk_major_threshold }}
+ labels:
+ severity: major
+ service: nova
+ annotations:
+ summary: "{{ alloc_disk_major_threshold * 100 }}% of aggregate disk space was allocated"
+ description: "{% raw %}{{ $value }}GB of disk space on the {{ $labels.aggregate }}{% endraw %} aggregate was allocated (at least {{ alloc_disk_major_threshold * 100 }}%)."
+ NovaAggregateAllocatedDiskFullCritical:
+ if: >-
+ openstack_nova_aggregate_disk - openstack_nova_aggregate_free_disk >= openstack_nova_aggregate_disk * {{ disk_ratio }}
+ labels:
+ severity: critical
+ service: nova
+ annotations:
+ summary: "No disk space available for allocation"
+ description: "All available disk space on the {% raw %}{{ $labels.aggregate }}{% endraw %} aggregate was allocated."
+ NovaAggregateDiskFullMajor:
+ if: >-
+ openstack_nova_aggregate_used_disk >= openstack_nova_aggregate_disk * {{ disk_major_threshold }}
+ labels:
+ severity: major
+ service: nova
+ annotations:
+ summary: "{{ disk_major_threshold * 100 }}% of aggregate disk space was used"
+ description: "{% raw %}{{ $value }}GB of disk space on the {{ $labels.aggregate }}{% endraw %} aggregate was used (at least {{ disk_major_threshold * 100 }}%)."
+ NovaAggregateDiskFullCritical:
+ if: >-
+ openstack_nova_aggregate_used_disk >= openstack_nova_aggregate_disk * {{ disk_critical_threshold }}
+ labels:
+ severity: critical
+ service: nova
+ annotations:
+ summary: "{{ disk_critical_threshold * 100 }}% of aggregate disk space was used"
+ description: "{% raw %}{{ $value }}GB of disk space on the {{ $labels.aggregate }}{% endraw %} aggregate was used (at least {{ disk_critical_threshold * 100 }}%)."
+ NovaTotalAllocatedVCPUsFullMinor:
+ if: >-
+ openstack_nova_total_used_vcpus >= openstack_nova_total_vcpus * {{ cpu_ratio }} * {{ alloc_cpu_minor_threshold }}
+ labels:
+ severity: minor
+ service: nova
+ annotations:
+ summary: "{{ alloc_cpu_minor_threshold * 100 }}% of cloud VCPUs were allocated"
+ description: "{% raw %}{{ $value }}{% endraw %} VCPUs in the cloud were allocated (at least {{ alloc_cpu_minor_threshold * 100 }}%)."
+ NovaTotalAllocatedVCPUsFullMajor:
+ if: >-
+ openstack_nova_total_used_vcpus >= openstack_nova_total_vcpus * {{ cpu_ratio }} * {{ alloc_cpu_major_threshold }}
+ labels:
+ severity: major
+ service: nova
+ annotations:
+ summary: "{{ alloc_cpu_major_threshold * 100 }}% of cloud VCPUs were allocated"
+ description: "{% raw %}{{ $value }}{% endraw %} VCPUs in the cloud were allocated (at least {{ alloc_cpu_major_threshold * 100 }}%)."
+ NovaTotalAllocatedVCPUsFullCritical:
+ if: >-
+ openstack_nova_total_used_vcpus >= openstack_nova_total_vcpus * {{ cpu_ratio }}
+ labels:
+ severity: critical
+ service: nova
+ annotations:
+ summary: "No VCPUs available for allocation"
+ description: "All available VCPUs in the cloud were allocated."
+ NovaTotalAllocatedMemoryFullMinor:
+ if: >-
+ openstack_nova_total_ram - openstack_nova_total_free_ram >= openstack_nova_total_ram * {{ ram_ratio }} * {{ alloc_ram_minor_threshold }}
+ labels:
+ severity: minor
+ service: nova
+ annotations:
+ summary: "{{ alloc_ram_minor_threshold * 100 }}% of cloud RAM was allocated"
+ description: "{% raw %}{{ $value }}MB{% endraw %} of RAM in the cloud was allocated (at least {{ alloc_ram_minor_threshold * 100 }}%)."
+ NovaTotalAllocatedMemoryFullMajor:
+ if: >-
+ openstack_nova_total_ram - openstack_nova_total_free_ram >= openstack_nova_total_ram * {{ ram_ratio }} * {{ alloc_ram_major_threshold }}
+ labels:
+ severity: major
+ service: nova
+ annotations:
+ summary: "{{ alloc_ram_major_threshold * 100 }}% of cloud RAM was allocated"
+ description: "{% raw %}{{ $value }}MB{% endraw %} of RAM in the cloud was allocated (at least {{ alloc_ram_major_threshold * 100 }}%)."
+ NovaTotalAllocatedMemoryFullCritical:
+ if: >-
+ openstack_nova_total_ram - openstack_nova_total_free_ram >= openstack_nova_total_ram * {{ ram_ratio }}
+ labels:
+ severity: critical
+ service: nova
+ annotations:
+ summary: "No RAM available for allocation"
+ description: "All available RAM in the cloud was allocated."
+ NovaTotalMemoryFullMajor:
+ if: >-
+ openstack_nova_total_used_ram >= openstack_nova_total_ram * {{ ram_major_threshold }}
+ labels:
+ severity: major
+ service: nova
+ annotations:
+ summary: "{{ ram_major_threshold * 100 }}% of cloud RAM was used"
+ description: "{% raw %}{{ $value }}MB{% endraw %} of RAM in the cloud was used (at least {{ ram_major_threshold * 100 }}%)."
+ NovaTotalMemoryFullCritical:
+ if: >-
+ openstack_nova_total_used_ram >= openstack_nova_total_ram * {{ ram_critical_threshold }}
+ labels:
+ severity: critical
+ service: nova
+ annotations:
+ summary: "{{ ram_critical_threshold * 100 }}% of cloud RAM was used"
+ description: "{% raw %}{{ $value }}MB{% endraw %} of RAM in the cloud was used (at least {{ ram_critical_threshold * 100 }}%)."
+ NovaTotalAllocatedDiskFullMinor:
+ if: >-
+ openstack_nova_total_disk - openstack_nova_total_free_disk >= openstack_nova_total_disk * {{ disk_ratio }} * {{ alloc_disk_minor_threshold }}
+ labels:
+ severity: minor
+ service: nova
+ annotations:
+ summary: "{{ alloc_disk_minor_threshold * 100 }}% of cloud disk space was allocated"
+ description: "{% raw %}{{ $value }}GB{% endraw %} of disk space in the cloud was allocated (at least {{ alloc_disk_minor_threshold * 100 }}%)."
+ NovaTotalAllocatedDiskFullMajor:
+ if: >-
+ openstack_nova_total_disk - openstack_nova_total_free_disk >= openstack_nova_total_disk * {{ disk_ratio }} * {{ alloc_disk_major_threshold }}
+ labels:
+ severity: major
+ service: nova
+ annotations:
+ summary: "{{ alloc_disk_major_threshold * 100 }}% of cloud disk space was allocated"
+ description: "{% raw %}{{ $value }}GB{% endraw %} of disk space in the cloud was allocated (at least {{ alloc_disk_major_threshold * 100 }}%)."
+ NovaTotalAllocatedDiskFullCritical:
+ if: >-
+ openstack_nova_total_disk - openstack_nova_total_free_disk >= openstack_nova_total_disk * {{ disk_ratio }}
+ labels:
+ severity: critical
+ service: nova
+ annotations:
+ summary: "No disk space available for allocation"
+ description: "All available disk space in the cloud was allocated."
+ NovaTotalDiskFullMajor:
+ if: >-
+ openstack_nova_total_used_disk >= openstack_nova_total_disk * {{ disk_major_threshold }}
+ labels:
+ severity: major
+ service: nova
+ annotations:
+ summary: "{{ disk_major_threshold * 100 }}% of cloud disk space was used"
+ description: "{% raw %}{{ $value }}GB{% endraw %} of disk space in the cloud was used (at least {{ disk_major_threshold * 100 }}%)."
+ NovaTotalDiskFullCritical:
+ if: >-
+ openstack_nova_total_used_disk >= openstack_nova_total_disk * {{ disk_critical_threshold }}
+ labels:
+ severity: critical
+ service: nova
+ annotations:
+ summary: "{{ disk_critical_threshold * 100 }}% of cloud disk space was used"
+ description: "{% raw %}{{ $value }}GB{% endraw %} of disk space in the cloud was used (at least {{ disk_critical_threshold * 100 }}%)."
{%- endif %}
NovaErrorLogsTooHigh:
{%- set log_threshold = monitoring.error_log_rate.warn|float %}
@@ -231,7 +531,8 @@
service: nova
annotations:
summary: "High number of errors in Nova logs"
- description: "The rate of errors in Nova logs over the last 5 minutes is too high on the {{ $labels.host }} node (current value={{ $value }}, threshold={%- endraw %}{{ log_threshold }})."
+ description: "The average per-second rate of errors in Nova logs on the {{ $labels.host }} node is {{ $value }} (as measured over the last 5 minutes)."
+{%- endraw %}
{%- if is_compute and exporters is defined %}
{%- raw %}
LibvirtDown: