release notes | Book: 2.10, 2.11, 2.12, 2.13, 2.14 | Wiki | Q&A black_bg
Web: Multi-page, Single page | PDF: A4-size, Letter-size | eBook: epub black_bg

Properties of the replica service

replica.cell.name

Default: dcache.enable.replica

Cell name of the replica service

dcache.enable.replica

Default: false

Set this value to true if you want to use the replica service.

replica.poolgroup

Default: ResilientPools

If you want to use another pool group for the resilient pools set this value to the name of the resilient pool group.

replica.db.host

Default: localhost

Set this value to the name of host of the replica service database.

replica.db.name

Default: replica

Name of the replica database table.

replica.db.user

Default: dcache

Change if the replicas database was created with a user other than dcache.

replica.db.password.file

Default: no password

replica.db.driver

Default: org.postgresql.Driver

replica service was tested with PostgreSQL only.

replica.limits.pool-watchdog-period

Default: 600 (10 min)

Pools Watch Dog poll period. Poll the pools with this period to find if some pool went south without sending a notice (messages). Can not be too short because a pool can have a high load and not send pings for some time. Can not be less than pool ping period.

replica.limits.excluded-files-expiration-timeout

Default: 43200 (12 hours)

replica.limits.delay-db-start-timeout

Default: 1200 (20 min)

On first start it might take some time for the pools to get connected. If replication started right away, it would lead to massive replications when not all pools were connected yet. Therefore the database init thread sleeps some time to give a chance to the pools to get connected.

replica.limits.adjust-start-timeout

Default: 1200 (20 min)

Normally Adjuster waits for database init thread to finish. If by some abnormal reason it cannot find a database thread then it will sleep for this delay.

replica.limits.wait-replicate-timeout

Default: 43200 (12 hours)

Timeout for pool-to-pool replica copy transfer.

replica.limits.wait-reduce-timeout

Default: 43200 (12 hours)

Timeout to delete replica from the pool.

replica.limits.workers

Default: 6

Number of worker threads to do the replication. The same number of worker threads is used for reduction. Must be more for larger systems but avoid situation when requests get queued in the pool.

replica.limits.replicas.min

Default: 2

Minimum number of replicas in pools which are online or offline.

replica.limits.replicas.max

Default: 3

Maximum number of replicas in pools which are online or offline.

replica.enable.check-pool-host

Default: true

Checks tag.hostname which can be specified in the layout file for each pool.

Set this property to false if you do not want to perform this check.

replica.enable.same-host-replica

Default: false

If set to true you allow files to be copied to a pool, which has the same tag.hostname as the source pool.

Note

The property replica.enable.check-pool-host needs to be set to true if replica.enable.same-host-replica is set to false.