28 April 2016
Alerting on Lambda Metrics
You can now alert on the metrics about your AWS Lambda functions.
5 August 2015
Improved S3 Metrics
Collection of S3 metadata on the size of buckets and number of objects per bucket are now using the new Cloudwatch metrics for that data as opposed to counting in your bucket. This will lead to accurate counts for buckets with over 500,000 objects.
6 July 2015
Stackdriver agent updated to collectd 5.5.0
We have updated the base for the Stackdriver agent to be collectd 5.5.0 which has a variety of bug fixes and improvements over the previous 5.3.0 version.
9 June 2015
Added Favorite Dashboards and Groups
Users can now mark individual dashboards and groups as favorites which will place them higher up in the navigation.
14 May 2015
More RDS Events
You will now see information in your event log about when your RDS instances have maintenance events or failover events. This can be helpful in better understanding the cause of various performance or available problems
More EBS Metrics
If you are using EBS volumes, you can now see more metrics including information about the queue length for those volumes and, for SSD PIOPS volumes, information on how much of the provisioned throughput you are getting.
21 April 2015
New Main Console
We enhanced the overall layout and visual design of the default Stackdriver dashboard.
- You can now configure the charts that you want to highlight on the home page.
- Uptime Checks are now displayed on the main console in the top left-hand corner of the screen.
- The incidents widget is wider, allowing you to access more information related to active incidents at a glance.
- You can expand/collapse any of the widgets on the page by clicking the empty space to the right of the section name.
Support for Lambda metrics on custom dashboards
You can now get the metrics for your Lambda functions on custom dashboards in Stackdriver. Update your IAM role for Stackdriver to allow the permissions
and then you will be able to have graphs on your custom dashboards of the number of invocations, number of errors, number of throttled invocations as well as the minimum, maximum and average duration for your functions' runtime.
5 March 2015
When creating or editing charts you now get an instant preview. The new dialog resizes to fit on various devices.
26 February 2015
We now have a mobile friendly header navigation. Clicking the menu icon now overlays the menu on top of the screen instead of pushing everything off the page.
24 February 2015
If you are using Stackdriver to monitor your Elasticsearch cluster, you can now monitor more metrics to help you understand the health of your cluster. Take a look at the updated agent configuration. This now lets you set up an alerting policy if any member of your Elasticsearch cluster thinks that there are nodes missing.
12 February 2015
Support for Elasticache Tags
13 January 2015
Kinesis Dashboards and Alerting
Stackdriver now collects your Kinesis metrics (be sure to update your IAM policy to include Kinesis). You may chart Kinesis Stream metrics on dashboards or see them on the service page. You may also create alerting policies on these metrics.
18 December 2014
Uptime Monitoring for AppEngine Modules
Uptime monitoring can now be set up to refer to an individual AppEngine module. This allows for easier configuration of those checks and seeing the status of an AppEngine Uptime Check from the individual resource page.
8 December 2014
The Stackdriver agent has been updated to contain a plugin that can be used for monitoring metrics from ZooKeeper and then using those metrics for graphing or alerting.
4 December 2014
CloudFront Dashboards and Alerting
You may now chart CloudFront metrics on custom dashboards or see them on the service page. You may also create alerting policies on CloudFront metrics.
2 December 2014
Added the ability to copy existing policies
Users can now create copies of existing policies for easy modification from the policy overview and detail pages.
17 November 2014
Improved Instance Custom Metric List
Filtering supports regex queries and persists per-browser. Metrics can be pinned for quick access from the "Pinned" view option.
12 November 2014
Full Support for Amazon's Frankfurt (eu-central-1) region
3 November 2014
Uptime Checks via the API
28 October 2014
Uptime Checks on Homepage
Additionally, resource uptime checks on the endpoints page have been updated to a more concise format.
24 October 2014
Suggested Policies from Endpoints
7 October 2014
Load Balancer Tags
2 October 2014
Custom Time Intervals on Dashboard
24 September 2014
Report Settings Page
15 September 2014
HBase 0.98 Support
12 September 2014
Google Cloud Status Updates in Event Feed
11 September 2014
Cloud Console Links on Resource Pages
Instance (AWS and GCP only), Block Storage Volume (AWS and GCP only), AppEngine App, CloudSQL, ELB, DynamoDB, ElastiCache, RDS, Redshift and SQS.
10 September 2014
Enterprise Linux 7 Support
2 September 2014
Events for AppEngine changes
Events for Google Cloud Project permissions changes
29 August 2014
Set Policies to trigger only if all the conditions are violated on a single resource
26 August 2014
Set tags for instances from the extractor
[tags] tag1=value tag2=someotherthing
and then restarting the stackdriver-extractor service.
25 August 2014
Support for Negative Numbers
You can now send and alert on negative numbers for Custom metrics within Stackdriver.
Instance Custom Wildcard Metric Charts
You can now use wildcard selectors such as project.prod.metric.*.count to graph collections of custom metrics being submitted by different instances. This feature is only enabled for users who are submitting custom metrics associated with an instance.
23 August 2014
Service Metrics Grouped Under Instances
Previously when adding charts to custom dashboards, all metrics related to instances were displayed together. They are now grouped and ordered by service.
20 August 2014
Google Compute Engine Instance Costs
Cost information for for Google Compute Engine instances is now available as an instance metric when creating custom dashboards and is also displayed on the instances page.
Links to AppEngine Logs
You can now navigate from AppEngine versions in Stackdriver to the logs viewer for the logs associated with the AppEngine module and version
18 August 2014
You can now make dashboards fullscreen by clicking on the fullscreen button from a custom dashboard. In fullscreen mode, the header and control bar are hidden.
13 August 2014
Alerting notifications via Slack Channels
You can now connect your Slack channels to Stackdriver for alerting notifications. Check under Account Settings > Notifications > Slack to configure any channels you may have.
12 August 2014
API for acknowledging and closing Incidents
You may now acknowledge or close any incident via the API. Details may be found at -http://docs.stackdriver.com/api/0.2/alerting.html#update-incidentstate-v0-2
8 August 2014
Support for Apache Scoreboard metrics
Metrics from the Apache scoreboard are now exposed for hosts using the apache agent plugin. You can see these metrics on any instance with the apache plugin installed and can also use them for custom dashboards or alerts
28 July 2014
Support for HBase and Kafka
3 July 2014
Support for AWS t2 Instances
Stackdriver now has support for the t2 instance family launched by Amazon yesterday. With this support, you can view the balance and usage of CPU credits on an individual instance page as well as in custom dashboards or alerting. Read more about the new instance types here
30 June 2014
You may enable a Stackdriver light theme under: Account Settings > Profile > Theme Preference.
18 June 2014
Maintenance Mode Events.
You may now see details about resources going in and out of maintenance mode in the Event Feed. You may filter them either by the resource to which it was applied or by the 'State' filter.
Maintenance Mode terminated resources.
6 June 2014
API support for retrieving custom metrics.
You can now query the Stackdriver API for a list of your custom metrics.
Maintenance mode for RDS Instances.
6 June 2014
Comments while switching Maintenance mode on.
3 June 2014
Login with a Google Account
Longer Windows for Absence of Metric Alerts
2 June 2014
Support for multiple Customer Accounts.
Stackdriver now allows you to associate a user with multiple customer accounts. As a result, a user does not need to have separate logins with different email addresses. An account administrators can now invite a user (email address) to be part of their customer account even if the user is associated with another customer. Users, who have multiple customer accounts associated, will now be able to select customer account they want to work on at login time. While logged into Stackdriver, they'll also be able to switch accounts. See details here.
28 May 2014
Support for retrieving Incidents in Alerting API
You can now retrieve incidents using the Alerting API. Incidents list endpoint returns a paginated list of incidents (in current implementation, ordered by started_at timestamp). You can filter the list by incident state, policy, resource or resource type. The API also provides an endpoint for retrieving details of a single incident.
Stackdriver Agent updates
The Stackdriver extractor portion of the agent has been updated to support proxies and has fixes for python 2.4.
23 May 2014
Added reason field for Maintenance mode
When setting maintenance mode on a resource you can now add a reason to describe why the resource is being set into maintenance mode.
RDS Events added to the Event Log
Stackdriver now shows you RDS Events for backups, Security Group and Parameter Group changes. You may see these in the Event Log on either the main dashboard or the RDS dashboard.
A new type filter, Configuration, has been added to the Event Log to help you track RDS Security Group and Parameter Group changes.
21 May 2014
Alerting on MySQL metrics
1 May 2014
Graph Latencies of Generic Endpoints
Support for CloudFormation event from Amazon CloudTrail
Added support for CloudFormation events from Amazon CloudTrail. You'll now see stack creation, deletion, update and update cancelation events in Stackdriver. Instructions to set up CloudTrail integration with Stackdriver are here.
30 April 2014
Load Balancer Attributes
29 April 2014
Timed Maintenance Mode for Instances and Resources via the API
mode on instances and load balancers. When the expiration time
passes, the resource is removed from maintenance mode. See the API
documentation for details:
Support for Elastic Load Balancer event from Amazon CloudTrail
Added support for Elastic Load Balancer events from Amazon CloudTrail. With CloudTrail integration enabled, you'll now see ELB creation, deletion, instance registration, instance deregistration, policy change and configuration change events in Stackdriver. Like other CloudTrail events, there is a delay between the actual events taking place and being visible in CloudTrail and in Strackdriver. If you haven't already, here is how to set up CloudTrail integration - http://support.stackdriver.com/customer/portal/articles/1491719-integrating-with-cloudtrail.
28 April 2014
Ubuntu 14.04 (Trusty Tahr) support
The Stackdriver agent is now built for and supports Ubuntu 14.04 LTS (Trusty Tahr). Installation instructions can be found here
API Support for Managing Custom Dashboards
You can now use the Stackdriver API to manage custom dashboards. You
can list, create, and delete dashboards. See the API documentation
for details: http://docs.stackdriver.com/
25 April 2014
Running resource counts for groups
You may now see the running count over time, for various resources belonging to a group. The resources that we count are ElastiCache, Instances, Load Balancers, Databases, SNS Topics, SQS Queues and Dynamo Tables.
Click the Running Resources tab in charts sections of the group dashboard to view the charts.
22 April 2014
Export Chart Data
Stackdriver now supports downloading of chart data in JSON format. The gear icon on upper right corner of each chart contains an "Export JSON" option that will save the data displayed currently by the chart as a JSON file.
Time Zone Preference
Stackdriver now supports configuring a preferred time zone on a per user basis. The setting can be changed at https://app.stackdriver.com/settings/profile. The default for all users is to use the browsers local time.
New API Endpoints - List Instances and Load Balancers
You can now query a list of instances or load balancers via the Stackdriver API. See the endpoints https://api.stackdriver.com/
If you have large numbers of resources, not that you may need to
17 April 2014
v0.2.0 of the Stackdriver statsd backend
A new version of the stackdriver-statsd-backend has been published to NPM (npm install stackdriver-statsd-backend) and GitHub (https://github.com/Stackdriver/stackdriver-statsd-backend). Among the new features:
- Add support for the statsd set datatype, which counts the number of distinct values sent for a metric in the collection period.
- Parameters to disable any of the default sub-metrics that we send for counters, gauges, and timers.
- Added support for sending the percentiles configured to compute in config.percentThreshold by setting config.stackdriver.sendTimerPercentiles.
- (beta) Added special case values for the source parameter. Setting detect-aws will attempt to pull a local AWS instance ID; setting detect-gce will attempt to pull a local GCE instance ID from the local metadata services.
- (advanced) Added config.stackdriver.sourceFromPrefix to allow for using a single statsd to accept metrics from multiple instances. The config.stackdriver.sourcePrefixSeparator (default to pipe | ) will set the character, anything before will be set as the source param (instance ID), anything after will be the metric name.
14 April 2014
Maintenance Mode for Load Balancers
You may now set load balancers (ELBs) into maintenance mode so that they will not produce alerts. To enable or disable maintenance, see the button on the details page for the ELB. You may also use the API (http://docs.stackdriver.com/api/0.2/loadbalancers.html) to control maintenance mode for ELBs.
4 April 2014
Running Instance counts for groups
You may now see the running instances over time for a group. Click the Running Resources tab in charts sections of the group dashboard to view the chart.
31 March 2014
Default Thresholds for Alerting Policies
We are now providing a suggested default threshold when you create a threshold alert policy for any metric. These thresholds take into account the past behavior of the resources that you are setting up the metric. If you notice any suggestions that seem particularly wrong, let us know so that we can continue to improve the algorithm!
Endpoint Port Monitoring
Our uptime monitoring solution now supports the option for you to monitor UDP/TPC ports, in addition to the HTTP/HTTPS monitoring we supported already!
To use it, simply select "Services" in the main navigation, and then "Endpoints" at the bottom of the Services menu to get to the endpoint listing page. Click the "Add an Endpoint" link at the top of the list, and select "UDP/TCP" from the Endpoint Type select field at the top of the form. Then simply choose a resource or group in our system...or enter a generic end point for us to check, enter the protocol and port to monitor, and tell us what text that should be expected in the response. Then tell us how often to check for you...it's that simple!
28 March 2014
New Alerting Notification Option - Team
Aside from alerting individuals via email and SMS, you may now alert a Team.
A Team is a group of users that may be created in the Account Settings page. Users may be added to or removed from a Team via the Users page in the Account Settings. Each user may also see the teams to which they belong in their Profile page.
27 March 2014
New Alerting Condition - Group Aggregate Threshold
Alerting has been upgraded to support alerting on a metric across a group.
26 March 2014
Support for monitoring CouchDB
An update to the agent has been published which supports monitoring CouchDB. See how to configure it here.
24 March 2014
Support for monitoring varnish
An update to the agent has been published which supports monitoring the Varnish web cache accelerator. See how to configure it here.
20 March 2014
New Uptime Monitoring Locations
We have added two new locations for uptime monitoring -- these locations are in Singapore and Oregon. This makes the set of probe locations Virginia, Dallas, Amsterdam, Singapore and Oregon.
18 March 2014
Security Groups for ELBs
If you are running your ELB inside of a VPC, you are able to control access to the ELB by adding it to a security group just like you can do with instances. This information is now collected and displayed in the Stackdriver UI for your load balancers.
17 March 2014
Generic Uptime Monitoring
In addition to monitoring HTTP/HTTPs endpoints on your instances and load balancers, you can now set up a check on any publicly accessible HTTP/HTTPs site in Stackdriver. To use this, go to the Endpoint list from the services menu, select 'Add an Endpoint Check' and then choose a 'Generic Endpoint'. You can also set up a policy to alert on when these endpoints aren't available.
14 March 2014
Server Monitoring Beta
In addition to monitoring your instances in the cloud, we now have beta support for monitoring servers anywhere whether they are in a data center or hosted by a provider we do not support. This support is currently built for Red Hat Enterprise Linux 6.x and compatible Linux distributions and Ubuntu 12.04 LTS. If you are interested in joining the beta, let us know.
12 March 2014
New Alerting System
You should have received an e-mail detailing when you will be migrated from the current system to the new system.
Specific health checks can now be configured for a policy. For example, you can now create policies which only take action when there is a scheduled maintenance window event or a degraded hardware event. A list of health checks and their descriptions can be found in this KB article.
Multiple Condition Policies
Alerting policies can now specify multiple conditions which define when an alert should be triggered. Conditions can be defined such that ALL or ANY must be met before triggering an incident. Elite customers can add up to 6 separate conditions to a policy.
When alerting on groups, you can now specify what % (or #) of members of the group must be in violation. For example you can create a condition which triggers if only 30% or more of the group is in violation.
You can list, read, delete, pause, and resume policies through the API.
11 March 2014
Create Custom Graphs of Endpoint Latency
You can now create graphs on custom dashboards or group dashboards of the latency of endpoints monitored by Stackdriver. Select the "Instance Endpoint" or "Load Balancer Endpoint" resource type. You can then filter based on the resource that the endpoint is monitoring with the usual filters of name and group.
10 March 2014
Uptime Monitoring User Agent Change
Checks with Stackdriver's uptime monitoring feature now use a user agent of Stackdriver_terminus_bot(http://www.stackdriver.com) rather than a generic python/requests user agent.
5 March 2014
Partial API Support for Policies
Additional functionality is now available through the API. Policies can now be put in maintenance mode, in the same way that instances can. The API also allows the listing and deletion of policies. Details can be found at http://docs.stackdriver.com/api .
4 March 2014
Change to Full-Screen mode for charts
Putting charts into "Full-Screen mode" used to both expand the chart to fill the browser window AND trigger the browser application to switch into "Full Screen mode" as well. For some users forcing the browser application into Full Screen mode was undesirable. We've changed the Full-Screen mode for charts to now expand to fill the current browser window area below the primary navigation, but it will no longer force the browser application into "Full Screen mode".
Charts in the new Full-Screen mode still behave as before (zoom, hover data-tips, etc.), but users will now continue to have access to all browser menus items, toolbars, and the Stackdriver application primary navigation at the top of the page. Users can now navigate to other areas of the application without needing to exit the chart Full-Screen mode.
3 March 2014
Removal of Custom Metrics
For cases where you send a custom metric accidentally or to test the API, you can now schedule the removal those metrics so that they are not visible for creating custom graphs or alerting policies. You can read the information on how to do this here
20 February 2014
Support for Amazon VPC
You can now view a list of all your VPCs as well as the related instances, security groups and subnets, under the infrastructure tab in the app. Security groups on the instance details page now contain information about the egress rules (if there are any present).
Group Criteria to Filter by VPC Subnet
You can now use VPC subnet as a criteria to filter on when creating a group.
Custom CloudWatch Metrics
Stackdriver now supports collecting custom metrics that you send to CloudWatch. By default, any metric that is sent to CloudWatch with a namespace of Stackdriver is collected. If you wish to collect additional namespaces, you can add them from your app settings.
Note that these metrics are treated the same as other custom metrics in Stackdriver.
19 February 2014
Alert Notifications via SNS
You can now use an SNS topic in your account as a target for alert notifications. Set up the topic from settings and then select it just like you can any of email, SMS, PagerDuty, etc. This allows another path to do custom handling of your alerts in a programatic fashion.
17 February 2014
Added Amazon CloudTrail Identity and Access Management (IAM) Messages to Activity Log
Added messages to activity log for creation and deletion of IAM users, roles, and groups.
New REST API Functionality
The REST API supports several new interfaces:
- Update a group.
- Get the list of metrics for an instance.
- Get the last hour of data for a given instance and metric.
Maintenance Mode for Instances in the UI
Maintenance mode can be controlled in the UI. On the "Instance Details" page, the "Actions" section includes a button to toggle the configuration of maintenance mode. If an instance is in maintenance mode, the "Instance Details" page shows a banner prominently at the top of the page to show the settings.
14 February 2014
Maintenance Mode for Instances in REST API
The REST API now includes functionality to put instances into maintenance mode. When a resource is in maintenance mode, new alerts for the resource will not be generated. You can use this functionality to silence alerts while performing maintenance on your system.
See the knowledge base for a complete description of maintenance mode: /customer/portal/articles/1491772-maintenance-mode-for-alerting
See the documentation for the REST API for detailed usage:
10 February 2014
Role-based API Keys
Due to the introduction of the REST API, API keys now have roles associated with them. All historical keys have been assigned a "Submit Data" role. This allows the holder to submit custom metrics or agent data to Stackdriver. There is also a "Read-Only" role that provides access to read data and metadata through the REST API. Finally, the "Admin" role allows the holder to access the create and update APIs.
07 February 2014
Support for Amazon RedShift
You can now create charts in groups and custom dashboards for your RedShift Clusters. RedShift monitoring is included free of charge during the month of February.
Available metrics are:
- Average Network Receive Throughput
- Average Network Transmit Throughput
- Average Percentage Disk Space Used
- Average Read IOPS
- Average Read Latency
- Average Read Throughput
- Average Write IOPS
- Average Write Latency
- Average Write Throughput
- CPU Utilization
- Database Connections
- Health Status
- Maintenance Mode
*Note that if you have set up a custom IAM policy with minimal permissions, you will need to add support in your role for the 'redshift:List*', 'redshift:Describe*', and 'redshift:Get*' actions to be able to access your RedShift data in Stackdriver.
05 February 2014
New RDS and ElastiCache Metrics
Custom dashboards and alerting policies can be created for the following new metrics:
- AverageMemoryUsage - The percentage of memory in use.
- AverageDiskUsage - The percentage of total disk space in use.
- AverageFreeableMemoryPercentage - The freeable memory available as a percentage of total memory.
- AverageBytesUsedForCachePercentage - (Redis) Cache bytes allocated as a percentage of total memory.
- AverageBytesUsedForCacheItemsPercentage - (Memcache) Bytes allocated to cache items as a percentage of total memory
- AverageUnusedMemoryPercentage - (Memcache) Unused bytes available to the cache as a percentage of total memory.
Proxy Support in Agent
The newest version of the agent package (5.3.0-130) now supports an easier way to set up use of an HTTP proxy for cases where you have instances in a VPC that does not allow for outbound HTTPs connections. To use this, add your proxy to the config file where you placed your agent api key (platform specific) in the PROXY_URL variable.
03 February 2014
API Alpha Release
Stackdriver is rolling out a REST API to help better manage your account through scriptable interfaces.
- current documentation can be found at http://docs.stackdriver.com/api
- python client library can be found at https://github.com/Stackdriver/stackdriver-client-python
- List Users
- List Groups
- Create Group
- Delete Group
- Resolve resource id
28 January 2014
Statistics for MongoDB secondary nodes
The Stackdriver agent will now report statistics for MongoDB secondary nodes.
26 January 2014
You can now access an overview of all of the endpoints which are being checked by Stackdriver at https://app.stackdriver.com/endpoints.
17 January 2014
Support for Amazon SNS
You can now create charts in groups and custom dashboards for your SNS Topics. SNS monitoring is included free of charge during the month of February.
Available metrics are:
- Average Size of Published Messages
- Maximum Size of Published Messages
- Minimum Size of Published Messages
- Number of Messages Published
- Number of Notifications Delivered
- Number of Notifications Failed
- SNS Count
*Note that if you have set up a custom IAM policy with minimal permissions, you will need to add support in your role for the sns:List* and sns:Get* actions to be able to access your SNS data in Stackdriver.
Microsoft IIS and SQL Server Monitoring Support
An updated version of the Windows agent (v0.4) is available which provides support for monitoring IIS and SQL Server. Installing the updated version from https://repo.stackdriver.com/windows/Stackdriverinstaller-0.4.exe will start collecting these metrics and make them available on the view for the instance
CentOS and Red Hat Enterprise Linux 5.x Agent Support
The Linux agent is now built for Red Hat Enterprise Linux 5.x and CentOS 5.x. Installation instructions can be found here.
26 December 2013
Custom Dashboard Improvements
We've launched a few improvements to custom dashboard pages today.
- When aggregating a metric across resources, you can now choose to use maximum, minimum, median and 95th percentile in addition to the previously supported aggregation functions of sum and average.
- Custom dashboards can be set to have a layout with 4, 5 or 6 columns in addition to the 1, 2 or 3 previously supported. This is great for widescreen monitors to be able to show more graphs on the screen!
- You can now copy a dashboard by clicking on the cog on the top of the dashboard and selecting "Copy Dashboard". Note that this creates a completely disconnected copy but should make it easier to build out sets of dashboards for similar use cases.
20 December 2013
See Policies Which Apply to a Resource
You can now see the policies which apply to any individual resource whether because they are directly applied to that policy, they apply to all resources of that type or that the policy applies to any group that the resource is a member of. To see them, click the icon next to the resource name on any individual resource page.
19 December 2013
Global Resource Search
There is now a search box in the upper right corner of the nav on the main homepage that allows you to search for and go to the details page for any resource quickly and easily. Feedback on the functionality is much appreciated and we are looking to make it more universal through the application in the near future.
16 December 2013
Alert on PostgreSQL metrics
Customers monitoring PostgreSQL via the Stackdriver agent can now select to create alerts based on some of those metrics.
13 December 2013
Beta Support for Google Compute Engine
We have launched beta support for monitoring Google Compute Engine. We currently monitor the state of instances, volumes and snapshots in your GCE environment and can get metrics about the instances via the Stackdriver agent. Read more about setting it up
Improved Policy View from Group Dashboards
When clicking the show policy icon from a group dashboard, you now also will see the policies which apply to all of your resources as well as any policies which apply to parent groups of the selected group. This should help to make it easier to manage and see the policies which apply to your resources.
12 December 2013
Group Criteria to Filter by AMI
You can now use the AMI of an instance as a criteria to filter on when creating a group.
11 December 2013
Improved Anomaly Event Cards
Anomaly cards in the event feeds and activity logs now include additional information regarding the anomaly including an embedded chart.
09 December 2013
Agent status in Settings
You can now see the status of how many of your instances currently have the agent running as well as the ones which don't have the agent installed. Check it out at https://app.stackdriver.com/settings/agent
06 December 2013
Share Your Stackdriver Charts
We have added an ability to share charts from Stackdriver via an embeddable iframe. Click the share button from the settings for a chart to get the embed code or read more about sharing charts in the documentation.
Latency chart for uptime checks
You can now view the latency reported by uptime checks over time. These charts will display at the top of any instance or load balancer page where a check is configured.
02 December 2013
Test Feature for Uptime Checks
A test button has been added to the create uptime check dialog so a user can test their endpoint before saving it
27 November 2013
RabbitMQ Agent Plugin
Stackdriver now supports direct monitoring of RabbitMQ queues. Documentation can be found here
25 November 2013
Stackdriver now supports the ability to notify external web services via webhooks. Documentation can be found here
14 November 2013
New PostgreSQL RDS and C3 Instances Support
Stackdriver supports both the newly announced PostgreSQL RDS and C3 Instances
13 November 2013
Organizing your infrastructure into logical groups just got more powerful with the introduction of subgroups.
- You can add subgroups to any group using the groups list on the main page or on any group page up to five levels deep.
- Subgroups filters apply in addition to its parent filters so that resources can be further categorized within a group.
- Subgroups can be accessed on a groups page as a clickable dropdown.
Group/Resource Heath at a Glance
- If a group's health is in a warning or error state we now show the number of unhealthy resources to the right of the group in the groups list which can be expanded to to reveal which resources are currently in an unhealthy state
Support for Amazon CloudTrail
- Stackdriver has added support for Amazons new auditing service CloudTrail. After configuring Stackdriver to connect to your CloudTrail logs, you can view the new CloudTrail events in your feeds across the site.
Redesigned Activity Log
- In support of the new CloudTrail events Stackdriver has completely redesigned its activity log to support faceted navigation and previews of events over time. You can find the page at: https://app.stackdriver.com/feed
Riak plugin for Stackdriver agent
- We now have a Riak plugin for the Stackdriver agent. You can find information on how to install it here. Please note, you must have the agent installed and upgraded to the latest version for this to work.
11 November 2013
- Users can now configure monitoring of any HTTP endpoints exposed on any Instance, Load Balancer or Group page. Documentation
Policy Based Action Automation
- Following configuration from the knowledge-base users can now set up actions to be automatically triggered when an incident is created.
20 October 2013
Monthly Cost Included in Instance List
- The instance list now includes the estimated monthly cost for each instance.
18 October 2013
Multiple Notifications for Alerting Policies
- Policies now support notifying multiple targets when an alert goes off. Configuring policies.
15 October 2013
Elasticsearch Information in Agent
- The agent has been updated to support Elasticsearch monitoring. Configuring the plugin will cause Elasticsearch to appear in the Services menu, and Elasticsearch charts to appear on the relevant instance details pages.
11 October 2013
Availability Zone metrics for ELB
- For ELBs, you can now see a breakdown of Average Latency, Requests, and Responses by Code per availability zone for an individual load balancer. Access the individual load balancer page and click on the By Availability Zone tab to see these metrics.
9 October 2013
Process Information in Agent
- The agent has been updated to send information about the number of processes in various states (running, zombie, sleeping, paging, stopped, and blocked). You can view this information on the details page for an instance or show in custom graphs or set up an alerting policy.
2 October 2013
Hipchat and Campfire Notifications
- Added support for publishing incident status updates to Hipchat or Campfire rooms. These can be configured on your account settings page and selected as a notification type for any alerting policy.
30 September 2013
Annotation Events API
- Added an API endpoint allowing for the creation of arbitrary annotations in the event timeline. These can be global or bound to an instance.
25 September 2013
Security Groups List Page
- We now have a page listing basic information for security groups
4 September 2013
Redis ElastiCache support
- We have basic support now for ElastiCache clusters using the redis engine as announced today by Amazon
27 August 2013
User Roles in Stackdriver
- Stackdriver now supports three basic types of user roles -- Admin, User and Read-only. Read more details
02 August 2013
statsd Output to Stackdriver
- Using our new statsd backend allows for sending aggregated metrics from statsd to Stackdriver. These can be custom metrics or custom instance metrics, and used in any custom dashboard.
26 July 2013
- Full support for ElastiCache is now available in the application. This includes a dashboard to view your ElastiCache clusters, details on each cluster, alerting based on the cloudwatch metrics and display of changes to ElastiCache clusters in your event feed.
- Support for service discovery has been added. Stackdriver will now discover instances running common services such as Apache, Nginx, MySQL, MongoDB, Redis, and more. Learn more here.
22 July 2013
AWS ElastiCache Metrics
- We are now collecting elasticache metrics for all AWS customers using it. The metrics can be viewed from graphs created inside a custom dashboard.
Metric Absence Detection
- Health check policies can now be configured to alert on the absence of a metric. You can configure the policy to alert if a metric isn't seen for longer than a configurable window. This can be applied to a single resource, group of resources, or all of a resource type just like existing policies. Look for the checkbox on the second step in configuring a health check policy.
Event Feed Added to the Resource Pages
- You can now see the events related to a particular resource on the relevant resource page.
Cluster Outlier Detection
- Stackdriver now analyzes your clusters to identify resources that are behaving differently from the others. You can learn about any issues that we find on the Insights page under the Events menu.
18 July 2013
Custom Metrics Library for Java
- We've made a library to send custom metrics to Stackdriver from within your Java applications. The source code can be found in GitHub. Read more about sending custom metrics to Stackdriver
9 July 2013
MongoDB Metric Support Added to the Agent
- The agent can now be configured to send metrics for your MongoDB instance. You must first install the agent or upgrade it to the latest version if you already have it installed. The MongoDB plugin then can be configured to start receiving metrics on your dashboard.
2 July 2013
Events on Graphs
- Events now show up as icons on graphs so you can correlate events to related metrics. Hover over the icon to get as summary of the events. Read more about events on graphs on the blog.
1 July 2013
Custom Metrics for Instances
- Custom metrics can now be explicitly associated with an Instance resource. Read more about sending custom metrics to Stackdriver
27 June 2013
Stackdriver Agent available for Debian Squeeze and Wheezy
- The Stackdriver agent is now available for Debian Squeeze and Wheezy in addition to the previously supported Linux distributions. See here for more info on installing the agent
- Stackdriver analyzes your infrastructure for anomalous behavior. The current analysis studies your load balancers, databases, and instances for sudden changes in key metrics. When a metric demonstrates a sudden spike or a step to a new "normal" level, the issue is reported as an "Insight" in the application.
20 June 2013
- Charts can enter fullscreen mode on supported browsers.
Auto-Refresh Refreshes Faster
- The auto-refresh for charts now refreshes every minute so that data being sent from the agent or custom metrics are visible more quickly
Additional CPU related metrics available
- If you have the agent installed, you can now graph and alert on CPU load (1min, 5min and 15min) and CPU Steal percentage in addition to overall CPU utilization. CPU load, especially the 5 or 15 min version, are a good indicator of system utilization that is less variable than pure utilization. CPU Steal percentage is a helpful indicator to discover when your instances are running on a heavily loaded host or have a noisy neighbor.
18 June 2013
Stackdriver Agent available for Ubuntu Lucid (10.04)
- The Stackdriver agent is now available for Ubuntu Lucid (10.04) in addition to the previously supported Linux distributions. See here for more info on installing the agent
Two way Pagerduty sync
- Stackdriver now supports updating its incident status based on the status of incidents resolved in Pagerduty.
17 June 2013
Multiple Email Recipients for Policy Alerting
- You may now enter multiple e-mail addresses when creating or editing a policy
- Hitting comma, semicolon or the enter key when adding an email address will add the current email to the form and allow you to immediately start typing the next email address
- Clicking on the x next to an e-mail address removes it from the form
14 June 2013
Code Deploy Events
- Stackdriver now supports ingesting and displaying code deploy events.
- Learn how to send us code deploy events here.
<p font-family:="" sans-serif;="" font-style:="" normal;="" font-variant:="" normal;"="">11 June 2013
Group charts updates
Charts on Group Dashboards now have the same features as charts on Custom Dashboards, which include:
- reordering charts
- adding multiple metrics
- using wildcards for adding custom metrics
6 June 2013
- Pagerduty services can be added through your account settings or inline with policy creation
- New incidents will be sent to the pagerduty service you configured and will be automatically resolved if the resource recovers
Stackdriver Agent Availability
- We are now providing binary packages of collectd with a simpler installation path for agent metrics for an initial set of Linux distributions. Information on setting those up can be found here
- If using the packaged agent configuration, then your metrics are now sent in a form that will let you alert on them
- The packaged agent also includes plugins for monitoring Apache, nginx, memcached, redis and MySQL. If you'd like to configure these, follow these instructions
3 June 2013
- Autoscaling groups have been added under the services tab.
- Configure IAM to add autoscaling permissions. /customer/portal/articles/1491727-what-are-the-minimal-permissions-needed-for-stackd.
29 May 2013
Defining Groups as Clusters
- Added the ability to define a group as a cluster. A cluster is a group of nodes that share a common configuration and role within an application. Incidents will be logged when any node's performance or configuration are different from the others in its cluster.
24 May 2013
Glob matching for multiple custom metrics on Custom Dashboards
- Added the ability for charts to pull custom metrics dynamically by configuring them with wildcards.
Groups Based on Instances behind a Load Balancer
- When creating a group, select LB Instances as the filter criteria, then select one of your load balancers. Amazon Elastic Load Balancer(s) are currently supported.
- The group will contain the instances behind that load balancer, as well as any ELBs that those instances belong to and any volumes attached to those instances.
22 May 2013
Aggregate Graph Metrics on Custom Dashboards
- Enables you to display the SUM or AVERAGE of all data series in a chart (rather than the values of each data series).
- For example, you can now view the average CPU of a group on a single chart.
Reorder charts on Custom Dashboards
- Drag and drop reordering of charts on custom dashboards.
21 May 2013
- Provides you with insight into metrics that deviated significantly from their normal performance range.
- This feature is accessible via Events-->Insights (view)
Multiple lines per chart
- Enables you to add more than one metric to charts on custom dashboards.
AMI inventory page
- Provides you with a list of all AMIs in your AWS account (view)
Instance health alerts
- Generates notifications when one of your instances is impaired. An instance can enter an impaired state for many reasons, such as failing a health check, being scheduled for retirement, etc.
- You can access the new health alerting policies via https://app.stackdriver.com/policy