| Setting name |
Default value |
Description |
| action-on-reboot |
just-reboot |
Specifies what happens when a worker reboots. The just-reboot value indicates the worker should simply reboot. The value erase-master-boot-record indicates the worker's master boot record is removed and the hard drive is re-imaged. Make sure the PXE server is available and correctly configured.
To re-image the hard-drive, the fabric_system_service (CentOS, RHEL, and Suse) or Appistry System Service (Windows) must be running and have administrative rights. |
| admin-server-port |
0 |
Specifies the port Fabric Control uses to submit calls to the fabric. A value of 0 indicates the operating system selects any available port over 1500. |
| auto-start-worker |
true |
Specifies if the fabric worker manager service automatically starts when the worker boots. True indicates fabric worker manager service starts. False indicates fabric worker manager service does not start. When the fabric manager service is running, the worker processes requests. If you are running on Windows, and the fabric worker manager service settings are set to automatically start up, this option is overridden. |
| concurrent-shutdown-threshold |
2 |
The maximum number of workers that can reboot or restart at the same time. This applies to all updates: configuration files, applications, Appistry EAF, and OS. |
| exclusive-task-count |
1 |
Specifies maximum number of exclusive tasks that may run concurrently per worker or per CPU depending on the exclusive-task-type. Note that this setting is across all exclusive tasks deployed on a worker. |
| exclusive-task-type |
per-machine |
Specifies if the exclusive-task-count is based on a worker or the number of CPUs in a worker. The per-cpu value indicates a worker can run the configured number of exclusive tasks on each CPU. The per-machine value indicates a worker can run the configured number of exclusive tasks. Note that this setting is across all exclusive tasks deployed on a worker. |
| guaranteed-non-fam-memory |
0 |
Specifies the amount of physical addressable memory, in kilobytes, on a worker that is not allocated for FAM spaces. |
| java-vm-max-heap-size-xmx-option |
|
A Java Virtual Machine startup option. This setting specifies the maximum Java memory heap size. |
| max-request-handler-count |
1000 |
The maximum number of request handlers. This number can exceed the number of workers, however, the number of request handlers running in the fabric will not exceed the number of workers. |
| mcast-ttl |
1 |
The number of routers fabric multicast packets pass through before expiring. |
| message-logger-group |
239.255.0.1 |
The MCAST address the log monitor uses to receive fabric messages.
You may delete this parameter. If this parameter is not in the fabric.cfg file, the messages are sent to the multicast address set by the fabric-address option in addr.cfg. |
| message-logger-port |
4000 |
The port the log monitor uses to receive fabric messages. You may delete this parameter. If this parameter is not in the fabric.cfg file, the messages are sent to port 4000. |
| min-request-handler-count |
100 |
The minimum number of request handlers. Setting this to 0 prevents the fabric from processing requests. |
| recovery-store-purge-interval |
30 |
When completed fabric requests cannot be returned to their client, they are placed in the fabric recovery store. When a request has been stored in recovery over a specified amount of time, it is purged from the store. The recovery-store-purge-interval defines the amount of time in minutes the fabric waits to check for requests that exceed the recovery-store-request-ttl parameter. The minimum amount of time is 60 minutes. |
| recovery-store-request-ttl |
24 |
When completed fabric requests cannot be returned to their client, they are placed in the fabric recovery store. When a request has been stored in recovery over a specified amount of time, it is purged from the store. The recovery-store-request-ttl specifies the amount of time, in hours, a completed request is stored in the recovery service before purging. The minimum amount of time is 1 hour. |
| request-handler-percentage |
100 |
The percentage of workers that should be request handlers. The min-request-handler-count and max-request-handler-count can override this setting.
For example, if a fabric consists of 100 workers and the minimum number of request handlers is 1 and the maximum 10, the fabric has only 10 active request handlers. |
| request-service-ssl-enabled |
false |
Specifies if SSL encryption is enabled. True indicates SSL encryption is enabled. False indicates SSL encryption is disabled. |
| request-service-port |
16000 |
The port workers use to receive TCP requests. Zero indicates any port may be used. |
| request-service-ssl-port |
16001 |
The port workers use to receive SSL TCP requests. Zero indicates any port may be used. |
| split-region-threshold |
20 |
Maximum region size. If the number of workers in a region exceed this number, the region will automatically split and reorganize into two regions. |
| task-service-monitor-auto-restart |
true |
Specifies if the fabric worker manager service automatically restarts a crashed task application service. False indicates the crashed task application service is not restarted. True indicates the task application service is restarted. In either event, a message is sent to the log monitor. |
| task-service-port |
16002 |
The port the task service uses to receive task requests. Do not set this value to 0. |
| version |
1.0 |
The fabric.cfg file version. |
