Topics marked with * relate to HQ Enterprise-only features.
- Auto-Discovery of Resources
- Auto Scan
- File Scan
- Initiating Auto-Discovery of Resources and Adding Them to the HQ Inventory
- Manually Adding Resources that Are Not Auto-Discovered
- Solving Auto-Discovery Problems
Feedback is welcome. Click Add Comment at the bottom of the page.
Hyperic HQ provides an intelligent and efficient auto-discovery feature to collect resource-specific details about the environment and include them in the HQ inventory. Auto-discovery discovers all of the resources running on the machines in your environment, from operating systems (platforms) to servers (such as Apache web servers) to services (such as Apache virtual hosts). For each of these discovered resources, HQ also discovers, collects, and stores metrics.
After a scan finishes, users can add the discovered resources to the inventory. The procedure for this is described below. Once the resources are in the HQ inventory, you can group resources into applications or other logical groups, and therefore more effectively manage their inventory in HQ.
The scan process also detects key changes to resources that are already in inventory. For instance, if a platform's fully qualified domain name (FQDN) changes, the next scan detects the change and presents it as modified property in the Platform Type section of the auto-discovery results.
There are two types of Auto-Discovery scans:
You can control the discovery and addition of resources as follows:
- Initiating auto-discovery of resources and adding them to the HQ inventory
- Manually adding resources to the HQ inventory when HQ doesn't discover them automatically
An auto-discovery "Auto Scan" requires no end-user intervention. Auto Scans typically do not take long and are not CPU-intensive. Auto Scans are automatically run by the HQ Agent every 15 minutes as well as each time an Agent is started.
The implementation of Auto Scan varies by platform. On Unix-type platforms, the Auto Scan is performed using a process-table scan, which looks for processes that match a given pattern. On Windows platforms, a simple registry scan is performed, looking for registry keys that installed products register during their installation process.
An auto-discovery "File Scan" scans the file system for installed products. File Scan must be initiated by the user using the New Auto-Discovery link found on all platform pages (in the Tools Menu). File Scan is more configurable than Auto Scan: users can specify which machines should be included in the auto-discovery, which directories should be searched, and which types of servers should be discovered.
File Scan can be performed on demand or can be scheduled. The scheduling functionality allows for recurring scans so that, for example, auto-discovery occurs on the last day of every month at 11:55 PM. Users can check the status of completed, in-progress, and scheduled File Scans.
The high-level procedure for discovering resources and adding them to the inventory is:
- Initiate an auto-discovery scan of a platform yourself or let HQ do it automatically.
- When the scan finishes, view the results in either of two places: on the Dashboard or the "Auto-Discovery Results" screen (which shows the results of the most recently performed Auto Scan on any platform, not necessarily the last scan you initiated).
- On either of those two screens, add discovered resources (or changes to already discovered resources) to the inventory.
Sometimes HQ does not automatically discover resources that you want to monitor. This happens most commonly when a resource is:
- A technology that is not monitored by default by HQ.
- A custom plugin.
- Installed in a nonstandard location that HQ does not look in by default, for instance a remote file system.
- A nonstandard name in the process list.
If this happens, you can manually add a resource (platform, server, or platform service) to the HQ inventory that HQ will then monitor.
To auto-discover a WebLogic Server instance on Linux or Solaris, the HQ Agent must be able to read its current working directory /proc/$pid/cwd. To ensure that the agent has sufficient privileges, you can:
- Run the HQ Agent as the same user that runs the WebLogic Server Administration Server,
- Run the HQ Agent as root, or
- Under Solaris 10 only, grant the agent account permission to read /proc/$pid/ files within the Solaris 10 Least Privilege Model (LPM). For instructions, see Configure Agent Account Privileges under Solaris 10.
For additional information about solving problems with auto-discovery of WebLogic Server, see [Configure HQ Agent to Monitor WebLogic Server].
An HQ Agent can fail to auto-discover some types of servers running under Solaris for a combination of reasons.
Solaris limits the length of the publicly-viewable process arguments struct member to 80 bytes, which can result in truncation of process arguments that the agent needs to read. To obtain all arguments, the agent will attempt to access the process address space file /proc/$pid/as and fail, if it lacks permission.
If the HQ Agent does not auto-discover servers under Solaris, run the agent as root, or grant the agent account permission to read /proc/$pid/ files within the Solaris 10 Least Privilege Model (LPM). For instructions, see Configure Agent Account Privileges under Solaris 10.