Service Profiler Snapshot Collector

This section describes what is the purpose of the tool, and:

• Which mechanisms it depends on (Scheduled Task) and required configurations (name-ruled Remote Aliases);
• The included functionalities with a small explanation and a link to their page.

This section describes the security concerns that the tool addresses and with which mechanisms they are assured.

The indications provided in this section point to security configurations that must be set in the Integration Server for controlled access to the tool.

This section describes administration aspects, namelly the Scheduled Tasks that are automatically created upon installation and their status:

• Green means that their are active;
• Red means thet they are suspended.

Setting the status of these Scheduled Tasks is the correct way of defining if the tool is active or not.

Attention: Do not change the status of the Triggers included with this package. These triggers are a mechanism of improving the gathering performance when a lot of servers are being monitored. It allows the parallel processing of the gathering requests and their results. Suspending the triggers does not remove the document subscription and if the gathering Scheduled Task is left active with the Triggers suspended the documents will keep accumulating in the server queue... until it is exhausted.

A particular server can be removed from the gathering list by simply deleting their configured Remote Alias. If needed, it can be reativated later by simply recreating its Remote Alias:

• The exact name of the alias is not important, but the server netname and port is.

  If it does not match a deactivated server, it is considered new, otherwise adds the new snapshots to the matched server.

The tool automatically creates a set of Scheduled Tasks driving the invocation of services named wia.isspscoll.scheduled.*:execute:

• A snapshot Gatherer from the configured remote servers;

  Scans the list of active remote servers and requests a Service Profiler snapshot sample to each of them, persisting it to the local bespoke XML repository.

• A repository Purger;

  Scans the repository and deletes all old, outdated, replaced, temporary ard/or leftover snapshots.

  Also deletes any files in the repository structure that are not recognized as a managed item.

  This task as some property settings that rule its behaviour, which are managed in a dedicated page accessible from the homepage.

• A snapshot Compactor.

  Scans all the snashots in the repository and combines old snapshots into a single one, accumulating the counters, included services, etc.

  The rules to decide which snapshots to combine/compact and when are defined through some property settings, in a dedicated page accessible from the homepage.

The above tasks:

• Are created Suspended;

  You have to activate them manually in this server's Scheduled Tasks management page.

• Are configured, by default, passing no arguments to the task's invoked service.

  It assumes the default argument value, which is resetCounters=false, which means that the snapshot counters at the remote server are never reset and the current sample keeps accumulating, with the same start timestamp.

  The collector will only create a new snapshot sample file if the start timestamp changes. If it remains the same, it just replace the one it has with that start timestamp with the new one.

  The start timestamp can change if:

  • Indicate a counters reset, through the scheduled task invoked service with argument resetCounters=true, or manually at the remote Service Profiler.
  • A remote Service Profiler package reload or restart;
  • A remote server restart.
• Are configurable in their behavior through some property settings;

  These are managed in a dedicated page accessible from the homepage.

• May have, to all of them but specially for the Gatherer, to have their scheduled calendar ajusted for your environment particular needs.

  Just take into considerartion that the more snapshot sample files you generate, and depending on their individual size, the more disk space will be used.

  Initially, monitor closely how much space is being used daily, and also weekly, to see if you risk running out of disk space any time soon.

  Then, depending on you environment, to increase the saving of disk space, you may have to change the Purger scheduler to more than once a day. This is because, in order to agilize the Gatherer, it does only minimal housekeeping in the repository and delegates the majority of that work to the Purger.

This page lists all the servers available for drilling down into, to view their snapshots.

The table is sortable by some of its columns.

The <Refresh> link at the top of the page reloads it, refreshing the contents of the table.

Column	Description
Checkbox	You can select one, more or all servers by checking their checkboxes. The checkbox on the column title is a shortcut to toggle select/deselect all.
Host	The name of the server. This name comes from the Remote Alias configuration, stripped of the domain part.
Port	The port of the server. This comes from the Remote Alias configuration.
Active	The servers that have this column with the value "Y" are the ones that are currently active for gathering, i.e. that have a Remote Alias configured according with the established naming rules. The others are servers that were active at some point in time and were deactivated by deleting their corresponding Remote Alias.
Select	You can go directly to the list of snapshots gathered for a server by clicking the button [Go], in alternative to using a single checkbox. If you which to honor the checkbox selection, the access to the snapshots must be done by clicking the button [List Snapshots], at the bottom of the page.

• Adding a server to be monitored (activate)

  Create the Remote Alias to access the intended server.

  This adds the server to the list of servers to be gathered by the Scheduled Task.

  The existence of duplicate definitions is not validated.

• Removing a server from monitoring (deactivate)

  Delete the Remote Alias.

• Reactivating a server to be monitored

  Recreate the Remote Alias to access the intended server.

  If the name and port matches an already inactive server, it is reactivated and the new gathered snapshots are added to it. Otherwise, it is activated as a new server.

The time it takes to load the page with the list of snapshots depends on the number of servers selected and the number of snapshots that are currently archived for them.

This page lists all the snapshots for the selected servers which can be selected for viewing, sorted by start date descending.

The table is not sortable by any of its columns.

The dates are strings formatted with a pattern that allows them to be naturally ordered even by being a string.

The <Refresh> link at the top of the page reloads it, refreshing the contents of the table.

The <Launch a Snapshot Gatherer Now> link at the top of the page overrides the Gatherer schedule and launches its process immediately. The page needs to be refreshed to view the new results.

Selection by date interval	Description
[Select] button	Applies the date selection to the checkboxes in the snapshot list table.
Start Date	Select snapshots starting from this date as part of its start date. If left blank, it means 'any start date' or 'the oldest'. Because the dates are strings, the value is matched as substring, from left to right, considering the remainder value blocks filled with values 0 (zero). For instance: "2016" will select all dates that start with that string; Which in practice means "all of the year 2016". "2016-11" will select all for November 2016; "2016-0" will select all for January to September 2016.
End Date	Select snapshots ending at this date as part of its end date. If left blank, it means 'any end date' or 'the newest'. The string matching comparison indicated for the Start Date is also used here, but considering the remainder value blocks filled with values 9 (nine).

Column	Description
Checkbox	You can select one, more or all snapshots by checking their checkboxes. The checkbox on the column title is a shortcut to toggle select/deselect all.
Host	The name of the server. This name comes from the Remote Alias configuration, stripped of the domain part.
Port	The port of the server. This comes from the Remote Alias configuration.
Sample Start	Start date timestamp of the snapshot.
Sample End	End date timestamp of the snapshot.
Select	You can go directly to the list of snapshots gathered for a server by clicking the button [Go], in alternative to using a single checkbox. If you which to honor the checkbox selection, the access to the snapshots must be done by clicking the button [Merge Selected & Load], at the bottom of the page. When more than one snapshot is selected, which can be of different servers, they are merged together as one. The resulting snapshot is then directly uploaded to the locally installed Service Profiler for viewing, and a new page is open with information about the operation: Load to Service Profiler.

This page shows the status of Merge Selected & Load snapshots.

It shows the context of the composed snapshot loaded into Service Profiler:

• The list of servers that are part of it;
• The time span covered by it, indicating the lowest timestamp as a start date and the highest timestamp as an end date.

This page allows removing the whole information about a server. It is an all-or-nothing operation.

Only inactive servers can be removed.

Item	Description
Forget Server(s)	When a server is added it is assigned a UUID. The same is also done to identify each new sampled snapshot. The use of these UUIDs allows moving the server's identification and all its gathered data from one instance of the Snapshot Collector to another without mixing or clashing with the data at the target repository. If this checkbox is unchecked, the server UUID is not deleted and the server can be reactivated because, if matched by name, it will reuse that identifier. If this checkbox is checked, the server UUID is also deleted and if an attempt to reactivate it happens, it will be given a completely new identifier, meaning that it is considered a new server.
Checkbox	You can select one, more or all servers by checking their checkboxes. The checkbox on the column title is a shortcut to toggle select/deselect all.
Host	The name of the server. This name comes from the Remote Alias configuration, stripped of the domain part.
Port	The port of the server. This comes from the Remote Alias configuration.
Active	The servers that have this column with the value "Y" are the ones that are currently active for gathering, i.e. that have a Remote Alias configured according with the established naming rules. The others are servers that were active at some point in time and were deactivated by deleting their corresponding Remote Alias.
Select	You can go directly to the delete operation of a server by clicking the button [Go], in alternative to using a single checkbox. If you which to honor the checkbox selection, the delete operation must be done by clicking the button [Delete Server(s)], at the bottom of the page.

A dialog box is shown to confirm the delete server operation and upon confirmation the operation is underdone and the page refreshed with the current list of servers.

This page edits the settings that govern the behavior of parts of the Snapshot Collector.

The properties are presented in a table grouped by sub-heading row that identifies to which part of the tool they belong.

Each property has a description column which helps identifying its purpose and expected values.

Heading	Description
Scheduled Purger	The Purger is triggered by an automatically created Scheduled Task. It automatically removes from the repository: Duplicated or outdated snapshots; Temporary or working snapshots created by the gatherer task; Any file that is not recognizably a valid repository item. This set of properties define the period of time (number of valid time units) the sampled snapshots are kept in the repository before being deleted, relative to the snapshot end timestamp. This task helps saving disk space.
Scheduled Compactor (same day)	The Compactor is triggered by an automatically created Scheduled Task. It compacts old snapshots by merging them into a single snapshot. This task helps saving disk space. This set of properties define the period of time (number of valid time units) the sampled snapshots are kept unmerged in the repository before being merged to a single one if they belong to the same day, based on the start and end timestamps.
Scheduled Compactor (same week)	This set of properties define the period of time (number of valid time units) the sampled snapshots are kept unmerged in the repository before being merged to a single one if they belong to the same week, based on the start and end timestamps. This time period must be larger than the one for 'same day'.
Scheduled Compactor (same month)	This set of properties define the period of time (number of valid time units) the sampled snapshots are kept unmerged in the repository before being merged to a single one if they belong to the same month, based on the start and end timestamps. This time period must be larger than the one for 'same week'.
Scheduled Compactor (same quarter)	This set of properties define the period of time (number of valid time units) the sampled snapshots are kept unmerged in the repository before being merged to a single one if they belong to the same quarter, based on the start and end timestamps. This time period must be larger than the one for 'same month'.

Click the [Save] button to persist any changes made to the properties.

When entering this page it may take a while because it starts by scanning all active servers, which is the same list used by the Gatherer task service, to collect from them their installed Service Profiler status:

• Version;

  The version of the installed tool.

• Days Left;

  How many days are left before the tool's License Key expires.

• Expiration Date;

  The license expiration date.

• Status Message.

  A message reported by the tool describing its status, or an error message.

This information is presented in a table with a row per active server. The table is sortable by some of its columns.

If the user logged in belongs to the Administrators ACL of this server, license key setting options will be revealled.

With these options, a New License Key input textbox is available for entering a valid License Key which can be sent for update of the Service Profilers selected by checking their corresponding checkbox.

Clicking the [Set New License Key] button triggers the sequential update of all the selected servers. This may take a while.

Upon completion a new page is shown with the result of each remote update. The License Key is validated before sending. If invalid, an error message is immediatelly shown.

This page is shown as the result of updating the License Key of the Service Profiler at the selected active remote servers.

It presents, in a tabular format, the list of servers where the License Key update was attempted and a message reflecting the result of that operation.

There are no extra actions that can be performed at this point in this page, with exception to navigating out of it.