LogDoor tasks are how you specify the details of how LogDoor will process a log file. Every time you create a new task, LogDoor creates a task file that defines that task. Each task file specifies:
The summary information is displayed in the task status window and written, on a periodic basis, to the summary reports. Summary data accumulates until reset with "Reset Summary Data" from the Task menu. Resetting Summary data erases all data previously stored in the task file.
LogDoor writes a task's data to disk periodically. If you wish to ensure that a task's data is written to disk at a specific time, you can choose "Checkpoint Now" from the Task menu.
Tasks are created using "New Task" from the File menu (also using "Create Task for Site Item..." from the Items command in the Task menu). You can name the task and specify where the task file will be stored. Tasks are opened and closed using "Open Task" and "Close Task", respectively. When a task is opened or created, its status window is displayed. Note that closing a status window does not close the associated task. Tasks are closed using "Close Task" or "Close All Tasks" from the File menu. When a task is closed it stops all processing. Closing a task's status window does not affect the task's processing in any way.
By default, a task does not begin processing until the Processing button in the status window is clicked, or "Run" is chosen from the Task menu. To have a task automatically begin processing when opened, choose "Scheduling..." from the Task menu. In the Scheduling dialog, check "Start processing upon task open?".
LogDoor can process more than one task file at a time. In many cases you will need only one LogDoor task. If however, you use multiple log files (to produce detailed site information, or one for your Web server and one for HomeDoor), or want to process the log file in different ways, you will need to set up several LogDoor tasks. You can create a new task, or choose one that's been created previously. To launch LogDoor and have it open several tasks at once, select the desired tasks and double-click them.
To remove an old site or root-level file from the state information associated with the task, and from the task status window, select the item in the task status window, and from the Task menu choose "Items" and then "Remove Item". All state information about the item will be deleted, and no future log files or reports for the item will be written, unless a valid access to the item is encountered during subsequent processing.
It is possible to create a site within a task's state information before any data has been logged to that site. It is not even necessary that a corresponding real site exist in the Web server's hierarchy. To create a task site, choose Item from the Task menu, and then choose "Create New Site...". The "Create New Site" dialog appears as shown in Figure 1 below.
Figure 1. Create New Site Item Dialog
Enter the name of the task site in the text box and choose the site's output options.
Creating a site item in a task is used to preset a new site's output options. For example, suppose you are about to bring a new site online for Company X, and you know that you don't want log files to be produced for that site. Say Company X's site is represented in the Web server's hierarchy by a folder named "companyX." You would choose Item from the Task menu, then choose "Create New Site...". Name the site item "companyX." Select "Use these options", deselect "Log Lines" and click OK. Now, when this task processes accesses made to companyX, it will not write a log file to companyX's folder within the task's output folder.
Virtual Root - By default, a task's root is the same as the Web server's root. Another folder to be used by LogDoor as root can be specified by choosing "Processing Options..." from the Task menu. The "Task Processing Options" dialog appears as shown in Figure 2 below.
Figure 2. Task Processing Options Dialog
To specify another folder to be used by LogDoor as root, enter the folder's pathname, relative to the root folder of the server itself, into the "Process items with this URL prefix as Root" text box. Root pathnames can be entered into the "Task Processing Options" dialog using colons or slashes as delimiters. For example, to set the virtual root for companyX, you could enter either ":companyX:" or "/companyX/".
If ":companyX:" is specified as root, then as LogDoor processes lines from the input log file, it processes only those lines beginning with ":companyX:", but as if ":companyX:" were removed from each line. Thus, the task's root is the folder companyX, and folders directly under companyX will be displayed as root-level sites. This mechanism is at the heart of detailed subsite reports.
Data for Root Files - By default, LogDoor logs access data for files at a task's root level to a special item called "Root". If so desired, such data can be logged to individual items representing the individual files. In the "Task Processing Options" dialog (see Figure 2 above), in the "Keep data for files at root in" section, choose either "'Root' item" or "Individual items".
"Old" Data Period - Detailed day-by-day information is maintained only for a certain number (90 by default) of days. Information prior to that period (or from log entries that don't include a valid date) is grouped together in a category called "Old". The number of days to store detailed day-by-day information for can be set by entering the number of days into the "Record data for up to" text box.
If the period is changed to a smaller value after a task has accumulated "Old" data, a certain amount of day-by-day information will be thrown away and lumped in with "Old" data. This cannot be reversed. That is, changing the "Old" period back to its original value will not retrieve the original day-by-day data.
If the "Old" period is increased from, say, 30 days to 90 days, note that when the task begins accumulating data with the new "Old" period, it will have data from 30 days ago lumped in with its "Old" data. In other words, increasing the "Old" period does not adjust the data stored in the "Old" category.
NOTE: Increasing the "Old" period can cause task files to become very large. Subsequently decreasing the "Old" period will not cause the size of a task file to shrink. Use caution if increasing the "Old" data period.
Visits - In addition to processing hits, LogDoor also processes visits. All hits to a given site from the same client count as one visit, provided the interval between hits does not exceed a specified amount.
For example, suppose the visit period is configured to be thirty minutes. The first access a given client makes to a given site is a visit to that site. If the client makes another access to the same site within thirty minutes, that access is part of the same visit. Thus, a visit could go on for hours as long as no interval between accesses to the same site by the same client exceeds thirty minutes. The first access the client makes to the same site that is more than thirty minutes after the last access is counted as part of a new visit.
What time period to use to define visits is a matter of interpretation, and so it is configurable in the "Task Processing Options" dialog. The default value is thirty minutes. To change this, bring up the "Task Processing Options" dialog (see Figure 2 above) and enter the new value in the "new visit starts after" text box.
Determining what constitutes a unique visit is a resource-intensive task. Due to the need to maintain and search a list of all "visitors" to a particular site for the indicated "visit" period, it is recommended that you set the time period which defines a visit to a low value (under an hour). Doing so will minimize the CPU overhead and RAM requirements of this feature. Additionally, since the visitor list for each site can be large, LogDoor only maintains this unique visitor list in RAM and does not write it to disk. Thus, any time you restart LogDoor or close a LogDoor task, its unique visitor list will be reset. You can disable visit processing entirely (visits always equal zero) for a task by setting the visit time to zero.
Processing Mode - How a LogDoor task processes its input log file is selected in the Scheduling dialog. Choose "Scheduling..." from the Task menu. The Scheduling dialog appears as shown in Figure 3 below.
Figure 3. Scheduling Dialog
In the "Processing Mode" section, choose the desired processing mode.
Start Processing on Task Open - By default, a task opens (or is created) in the suspended state, and must be manually started using the processing button in the status window, or the Run command in the Task menu. To have a task start processing automatically when it is opened, check the "Start processing upon task open?" checkbox in the Scheduling dialog (see Figure 3 above).
Output Options - Using the "Output Files" dialog it is possible to specify a task's output options. From the Task menu, choose "Output Files...". In the "Output Folder" section you can choose the location of the task's output folder. The default is a folder named "Output" within the task's folder. Other output options are covered in Input & Output Log Files and Summary Reports.
LogDoor 2.0 is backward-compatible with LogDoor 1.2. That is, LogDoor 2.0 can open and run LogDoor 1.2 (and older) tasks. Because LogDoor 2.0 uses task data not present in LogDoor 1.2 tasks, LogDoor will fill in those missing values with reasonable assumptions.
Tasks created or modified by LogDoor versions 2.0 or later will not be readable by previous versions. If you are going to open a pre-2.0 task with LogDoor 2.0 or later, and you will want to open it later with a pre-2.0 version, you should save a copy of the task file before opening it with version 2.0.
Back to Table of Contents
Back to LogDoor Basics
Forward to Input & Output Log
Files