extern/zrepl
1
0
Fork 0
mirror of https://github.com/zrepl/zrepl.git synced 2025-06-24 11:41:27 +02:00
Code Issues Actions Packages Projects Releases Wiki Activity

docs: document bookmarking + remove warning about replication lag

Browse Source 
refs #34
This commit is contained in:
Christian Schwarz 2017-12-28 13:24:25 +01:00
parent 03ba2bb7c8
commit 9a19615fd4
2 changed files with 14 additions and 9 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
docs/configuration/jobs.rst
View File

@ -54,6 +54,7 @@ Example: :sampleconf:`pullbackup/productionhost.yml`.
- Snapshotting Task (every ``interval``, |patient|) - Snapshotting Task (every ``interval``, |patient|)
- A snapshot of filesystems matched by ``filesystems`` is taken every ``interval`` with prefix ``snapshot_prefix``. - A snapshot of filesystems matched by ``filesystems`` is taken every ``interval`` with prefix ``snapshot_prefix``.
- A bookmark of that snapshot is created with the same name.
- The ``prune`` policy is triggered on filesystems matched by ``filesystems`` with snapshots matched by ``snapshot_prefix``. - The ``prune`` policy is triggered on filesystems matched by ``filesystems`` with snapshots matched by ``snapshot_prefix``.
- Serve Task - Serve Task
@ -62,11 +63,13 @@ Example: :sampleconf:`pullbackup/productionhost.yml`.
A source job is the counterpart to a :ref:`job-pull`. A source job is the counterpart to a :ref:`job-pull`.
Note that the prune policy determines the maximum replication lag:
a pull job may stop replication due to link failure, misconfiguration or administrative action.
The source prune policy will eventually destroy the last common snapshot between source and pull job, requiring full replication.
Make sure you read the |prune| policy documentation. Make sure you read the |prune| policy documentation.
Note that zrepl does not prune bookmarks due to the following reason:
a pull job may stop replication due to link failure, misconfiguration or administrative action.
The source prune policy will eventually destroy the last common snapshot between source and pull job.
Without bookmarks, the prune policy would need to perform full replication again.
With bookmarks, we can resume incremental replication, only losing the snapshots pruned since the outage.
.. _job-pull: .. _job-pull:
@ -148,6 +151,7 @@ Example: :sampleconf:`localbackup/host1.yml`
#. Evaluate ``mapping`` for local filesystems, those with a *target filesystem* are called *mapped filesystems*. #. Evaluate ``mapping`` for local filesystems, those with a *target filesystem* are called *mapped filesystems*.
#. Snapshot *mapped filesystems* with ``snapshot_prefix``. #. Snapshot *mapped filesystems* with ``snapshot_prefix``.
#. Bookmark the snapshot created above.
#. Replicate *mapped filesystems* to their respective *target filesystems*: #. Replicate *mapped filesystems* to their respective *target filesystems*:
#. Only snapshots with prefix ``snapshot_prefix`` are replicated. #. Only snapshots with prefix ``snapshot_prefix`` are replicated.
@ -159,6 +163,8 @@ Example: :sampleconf:`localbackup/host1.yml`
#. The ``prune_rhs`` policy is triggered for all *target filesystems* #. The ``prune_rhs`` policy is triggered for all *target filesystems*
A local job is combination of source & pull job executed on the same machine. A local job is combination of source & pull job executed on the same machine.
Note that while snapshots are pruned, bookmarks are not pruned and kept around forever.
Refer to the comments on :ref:`source job <job-source>` for the reasoning behind this.
Terminology Terminology
----------- -----------

11
docs/configuration/prune.rst
View File

@ -53,12 +53,11 @@ The following procedure happens during pruning:
.. ATTENTION:: .. ATTENTION::
.. TODO this is obsolete as soon as bookmarks are created during snapshotting The configuration of the first interval (``1x1h(keep=all)`` in the example) determines the **maximum allowable replication lag** because the source and destination pruning policies do not coordinate:
if replication does not work for whatever reason, source will continue to execute the prune policy.
Eventually, source destroys a snapshot that has never been replicated to destination, degrading the temporal resolution of your backup.
The configuration of the first interval (``1x1h(keep=all)`` in the example) determines the **maximum allowable replication lag** between source and destination. Thus, **always** configure the first interval to ``1x?(keep=all)``, substituting ``?`` with the maximum time replication may fail due to downtimes, maintenance, connectivity issues, etc.
After the first interval, source and destination likely have different retention settings.
This means source and destination may prune different snapshots, prohibiting incremental replication froms snapshots that are not in the first interval.
**Always** configure the first interval to ``1x?(keep=all)``, substituting ``?`` with the maximum time replication may fail due to downtimes, maintenance, connectivity issues, etc. .. We intentionally do not mention that bookmarks are used to bridge the gap between source and dest that are out of sync snapshot-wise. This is an implementation detail.
After outages longer than ``?`` you may be required to perform **full replication** again.