Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
9 changes: 7 additions & 2 deletions de/15.9/config/admin-index-backup.rst
Original file line number Diff line number Diff line change
Expand Up @@ -177,12 +177,17 @@ Nur ``fess*`` wiederherzustellen genügt ebenfalls nicht: Ohne die Wörterbuchda
"include_global_state": false
}'

2. Schreiben Sie die wiederhergestellten Wörterbücher nach ``config/dictionary`` von OpenSearch aus.
2. Starten Sie OpenSearch neu (in einem Cluster mit mehreren Knoten alle Knoten). Sobald der Index ``configsync`` verfügbar ist, schreiben Sie die wiederhergestellten Wörterbücher nach ``config/dictionary`` von OpenSearch aus.

::

curl -X GET "localhost:9200/_cluster/health/configsync?wait_for_status=yellow&timeout=60s&pretty"

curl -X POST "localhost:9200/_configsync/flush"

.. note::
Mit dem configsync-Plugin 3.8.0 und älter schreibt ``_configsync/flush`` nur die Dateien aus, die registriert wurden, nachdem der Knoten zuletzt einen Schreibvorgang (planmäßig oder ``_configsync/flush``) gestartet hat. Die wiederhergestellten ``configsync``-Dateien tragen den Zeitpunkt, zu dem sie auf dem Cluster registriert wurden, von dem der Snapshot stammt. Sobald der erste planmäßige Schreibvorgang etwa eine Minute nach dem Start von OpenSearch gelaufen ist, schreibt ``_configsync/flush`` ohne Neustart daher keine Wörterbuchdateien aus, und der Cluster wird nach Schritt 3 red. Ein Neustart setzt diesen Zeitpunkt zurück, sodass ``_configsync/flush`` nach dem Neustart alle Wörterbuchdateien ausschreibt. Für jede ausgeschriebene Datei protokolliert OpenSearch ``Updated {Dateipfad}`` von ``ConfigSyncService``.

3. Stellen Sie die |Fess|-Indizes wieder her.

::
Expand Down Expand Up @@ -398,7 +403,7 @@ Wenn der Clusterstatus nach der Wiederherstellung von ``fess*`` ``red`` ist und

Weder ``_cluster/reroute?retry_failed=true`` noch das Bereitstellen der Wörterbuchdateien behebt das: Ein Shard, dessen Wiederherstellung fehlgeschlagen ist, wird erst wieder zugewiesen, wenn sein Index geschlossen oder gelöscht und erneut wiederhergestellt wird. Gehen Sie wie folgt vor.

1. Folgen Sie den Schritten 1 und 2 unter „Wiederherstellung aller Indizes“, um ``configsync`` wiederherzustellen und die Wörterbuchdateien auszuschreiben.
1. Folgen Sie den Schritten 1 und 2 unter „Wiederherstellung aller Indizes“, um ``configsync`` wiederherzustellen, OpenSearch neu zu starten und die Wörterbuchdateien auszuschreiben.
2. Listen Sie die Indizes mit Status ``red`` auf.

::
Expand Down
9 changes: 7 additions & 2 deletions en/15.9/config/admin-index-backup.rst
Original file line number Diff line number Diff line change
Expand Up @@ -177,12 +177,17 @@ Restoring only ``fess*`` does not work either: without the dictionary files, ``f
"include_global_state": false
}'

2. Write the restored dictionaries out to OpenSearch's ``config/dictionary``.
2. Restart OpenSearch (every node in a multi-node cluster). Once the ``configsync`` index is available, write the restored dictionaries out to OpenSearch's ``config/dictionary``.

::

curl -X GET "localhost:9200/_cluster/health/configsync?wait_for_status=yellow&timeout=60s&pretty"

curl -X POST "localhost:9200/_configsync/flush"

.. note::
With configsync plugin 3.8.0 and earlier, ``_configsync/flush`` writes only the files registered after the node last started a write run (a scheduled run or ``_configsync/flush``). The restored ``configsync`` files carry the times they were registered on the cluster the snapshot was taken from, so once the first scheduled write has run, about one minute after OpenSearch starts, running ``_configsync/flush`` without a restart writes no dictionary files and the cluster turns red after step 3. A restart resets that point, so ``_configsync/flush`` after the restart writes every dictionary file. For each file it writes, OpenSearch logs ``Updated {file path}`` from ``ConfigSyncService``.

3. Restore the |Fess| indexes.

::
Expand Down Expand Up @@ -398,7 +403,7 @@ If the cluster status is ``red`` after restoring ``fess*`` and you see the follo

Neither ``_cluster/reroute?retry_failed=true`` nor putting the dictionary files in place recovers it: a shard that failed to restore is not allocated again until its index is closed or deleted and restored once more. Recover as follows.

1. Follow steps 1 and 2 of "Restoring All Indexes" to restore ``configsync`` and write out the dictionary files.
1. Follow steps 1 and 2 of "Restoring All Indexes" to restore ``configsync``, restart OpenSearch, and write out the dictionary files.
2. List the ``red`` indexes.

::
Expand Down
9 changes: 7 additions & 2 deletions es/15.9/config/admin-index-backup.rst
Original file line number Diff line number Diff line change
Expand Up @@ -177,12 +177,17 @@ Tampoco basta con restaurar solo ``fess*``: sin los archivos de diccionario, ``f
"include_global_state": false
}'

2. Escriba los diccionarios restaurados en ``config/dictionary`` de OpenSearch.
2. Reinicie OpenSearch (todos los nodos en un clúster de varios nodos). Cuando el índice ``configsync`` esté disponible, escriba los diccionarios restaurados en ``config/dictionary`` de OpenSearch.

::

curl -X GET "localhost:9200/_cluster/health/configsync?wait_for_status=yellow&timeout=60s&pretty"

curl -X POST "localhost:9200/_configsync/flush"

.. note::
Con el plugin configsync 3.8.0 y anteriores, ``_configsync/flush`` solo escribe los archivos registrados después de la última vez que el nodo inició una escritura (programada o ``_configsync/flush``). Los archivos de ``configsync`` restaurados conservan la hora en que se registraron en el clúster del que se tomó la instantánea, por lo que, una vez que se ha ejecutado la primera escritura programada, aproximadamente un minuto después de iniciar OpenSearch, ejecutar ``_configsync/flush`` sin reiniciar no escribe ningún archivo de diccionario y el clúster pasa a red después del paso 3. El reinicio restablece ese punto, de modo que ``_configsync/flush`` tras el reinicio escribe todos los archivos de diccionario. Por cada archivo que escribe, OpenSearch registra ``Updated {ruta del archivo}`` desde ``ConfigSyncService``.

3. Restaure los índices de |Fess|.

::
Expand Down Expand Up @@ -398,7 +403,7 @@ Si el estado del clúster es ``red`` después de restaurar ``fess*`` y observa l

Ni ``_cluster/reroute?retry_failed=true`` ni colocar los archivos de diccionario lo resuelven: un shard cuya restauración falló no se vuelve a asignar hasta que su índice se cierra o se elimina y se restaura de nuevo. Recupérelo de la siguiente manera.

1. Siga los pasos 1 y 2 de «Restauración de Todos los Índices» para restaurar ``configsync`` y escribir los archivos de diccionario.
1. Siga los pasos 1 y 2 de «Restauración de Todos los Índices» para restaurar ``configsync``, reiniciar OpenSearch y escribir los archivos de diccionario.
2. Liste los índices en estado ``red``.

::
Expand Down
9 changes: 7 additions & 2 deletions fr/15.9/config/admin-index-backup.rst
Original file line number Diff line number Diff line change
Expand Up @@ -177,12 +177,17 @@ Restaurer uniquement ``fess*`` ne suffit pas non plus : sans les fichiers de dic
"include_global_state": false
}'

2. Écrivez les dictionnaires restaurés dans ``config/dictionary`` d'OpenSearch.
2. Redémarrez OpenSearch (tous les nœuds dans un cluster à plusieurs nœuds). Une fois l'index ``configsync`` disponible, écrivez les dictionnaires restaurés dans ``config/dictionary`` d'OpenSearch.

::

curl -X GET "localhost:9200/_cluster/health/configsync?wait_for_status=yellow&timeout=60s&pretty"

curl -X POST "localhost:9200/_configsync/flush"

.. note::
Avec le plugin configsync 3.8.0 et antérieur, ``_configsync/flush`` n'écrit que les fichiers enregistrés après le dernier démarrage d'une écriture par le nœud (planifiée ou ``_configsync/flush``). Les fichiers ``configsync`` restaurés conservent l'heure à laquelle ils ont été enregistrés sur le cluster d'où provient le snapshot : une fois la première écriture planifiée exécutée, environ une minute après le démarrage d'OpenSearch, ``_configsync/flush`` sans redémarrage n'écrit donc aucun fichier de dictionnaire et le cluster passe à red après l'étape 3. Le redémarrage réinitialise ce point, de sorte que ``_configsync/flush`` après le redémarrage écrit tous les fichiers de dictionnaire. Pour chaque fichier écrit, OpenSearch journalise ``Updated {chemin du fichier}`` depuis ``ConfigSyncService``.

3. Restaurez les index de |Fess|.

::
Expand Down Expand Up @@ -398,7 +403,7 @@ Si l'état du cluster est ``red`` après la restauration de ``fess*`` et que vou

Ni ``_cluster/reroute?retry_failed=true`` ni la mise en place des fichiers de dictionnaire ne suffisent : un shard dont la restauration a échoué n'est plus alloué tant que son index n'a pas été fermé ou supprimé puis restauré à nouveau. Procédez comme suit.

1. Suivez les étapes 1 et 2 de « Restauration de tous les index » pour restaurer ``configsync`` et écrire les fichiers de dictionnaire.
1. Suivez les étapes 1 et 2 de « Restauration de tous les index » pour restaurer ``configsync``, redémarrer OpenSearch et écrire les fichiers de dictionnaire.
2. Listez les index à l'état ``red``.

::
Expand Down
9 changes: 7 additions & 2 deletions ja/15.9/config/admin-index-backup.rst
Original file line number Diff line number Diff line change
Expand Up @@ -177,12 +177,17 @@ cronなどを使用して、定期的にバックアップを実行できます
"include_global_state": false
}'

2. リストアした辞書の内容を OpenSearch の ``config/dictionary`` 配下へ書き出します。
2. OpenSearch を再起動し(複数ノードのクラスタではすべてのノード)、 ``configsync`` インデックスが利用可能になったら、リストアした辞書の内容を OpenSearch の ``config/dictionary`` 配下へ書き出します。

::

curl -X GET "localhost:9200/_cluster/health/configsync?wait_for_status=yellow&timeout=60s&pretty"

curl -X POST "localhost:9200/_configsync/flush"

.. note::
configsync プラグイン 3.8.0 以前では、 ``_configsync/flush`` は、そのノードが前回書き出し処理(定期実行または ``_configsync/flush`` )を開始した時点より後に登録されたファイルだけを書き出します。リストアした ``configsync`` のファイルはスナップショットを取得したクラスタで登録された時刻を持つため、OpenSearch の起動から約 1 分後に最初の定期書き出しが済んだ後では、再起動せずに ``_configsync/flush`` を実行しても辞書ファイルは書き出されず、手順 3 の後でクラスタが red になります。再起動するとこの時点が初期化されるため、再起動後の ``_configsync/flush`` ですべての辞書ファイルが書き出されます。書き出したファイルごとに、OpenSearch のログに ``ConfigSyncService`` の ``Updated {ファイルのパス}`` が出力されます。

3. |Fess| のインデックスをリストアします。

::
Expand Down Expand Up @@ -398,7 +403,7 @@ OpenSearch のインデックスとは別に、以下の設定ファイルもバ

``_cluster/reroute?retry_failed=true`` を実行しても、また辞書ファイルを配置しただけでも回復しません。リストアに失敗したシャードは、そのインデックスをクローズまたは削除して、もう一度リストアするまで割り当てられないためです。次の手順で回復します。

1. 「全インデックスのリストア」の手順 1 と 2 に従い、 ``configsync`` をリストアして辞書ファイルを書き出します。
1. 「全インデックスのリストア」の手順 1 と 2 に従い、 ``configsync`` をリストアし、OpenSearch を再起動して辞書ファイルを書き出します。
2. ``red`` のインデックスを確認します。

::
Expand Down
9 changes: 7 additions & 2 deletions ko/15.9/config/admin-index-backup.rst
Original file line number Diff line number Diff line change
Expand Up @@ -177,12 +177,17 @@ cron 등을 사용하여 정기적으로 백업을 실행할 수 있습니다.
"include_global_state": false
}'

2. 복원한 사전 내용을 OpenSearch의 ``config/dictionary`` 아래에 기록합니다.
2. OpenSearch를 재시작하고(여러 노드로 구성된 클러스터에서는 모든 노드), ``configsync`` 인덱스를 사용할 수 있게 되면 복원한 사전 내용을 OpenSearch의 ``config/dictionary`` 아래에 기록합니다.

::

curl -X GET "localhost:9200/_cluster/health/configsync?wait_for_status=yellow&timeout=60s&pretty"

curl -X POST "localhost:9200/_configsync/flush"

.. note::
configsync 플러그인 3.8.0 이하에서는 ``_configsync/flush`` 가 해당 노드가 마지막으로 기록 처리(정기 실행 또는 ``_configsync/flush`` )를 시작한 시점 이후에 등록된 파일만 기록합니다. 복원한 ``configsync`` 의 파일은 스냅샷을 생성한 클러스터에서 등록된 시각을 가지고 있으므로, OpenSearch 시작 후 약 1분 뒤에 첫 번째 정기 기록이 실행된 후에는 재시작하지 않고 ``_configsync/flush`` 를 실행해도 사전 파일이 기록되지 않으며, 3단계 후에 클러스터가 red가 됩니다. 재시작하면 이 시점이 초기화되므로, 재시작 후의 ``_configsync/flush`` 로 모든 사전 파일이 기록됩니다. 기록된 파일마다 OpenSearch 로그에 ``ConfigSyncService`` 의 ``Updated {파일 경로}`` 가 출력됩니다.

3. |Fess| 인덱스를 복원합니다.

::
Expand Down Expand Up @@ -398,7 +403,7 @@ OpenSearch의 인덱스와는 별도로 다음 설정 파일도 백업하십시

``_cluster/reroute?retry_failed=true`` 를 실행하거나 사전 파일을 배치하는 것만으로는 복구되지 않습니다. 복원에 실패한 샤드는 해당 인덱스를 닫거나 삭제한 뒤 다시 복원할 때까지 할당되지 않기 때문입니다. 다음 절차로 복구합니다.

1. “모든 인덱스 복원”의 1단계와 2단계에 따라 ``configsync`` 를 복원하여 사전 파일을 기록합니다.
1. “모든 인덱스 복원”의 1단계와 2단계에 따라 ``configsync`` 를 복원하고 OpenSearch를 재시작하여 사전 파일을 기록합니다.
2. ``red`` 인덱스를 확인합니다.

::
Expand Down
9 changes: 7 additions & 2 deletions zh-cn/15.9/config/admin-index-backup.rst
Original file line number Diff line number Diff line change
Expand Up @@ -177,12 +177,17 @@
"include_global_state": false
}'

2. 将恢复的词典内容写出到 OpenSearch 的 ``config/dictionary`` 下。
2. 重启 OpenSearch(多节点集群中为所有节点),待 ``configsync`` 索引可用后,将恢复的词典内容写出到 OpenSearch 的 ``config/dictionary`` 下。

::

curl -X GET "localhost:9200/_cluster/health/configsync?wait_for_status=yellow&timeout=60s&pretty"

curl -X POST "localhost:9200/_configsync/flush"

.. note::
在 configsync 插件 3.8.0 及更早版本中, ``_configsync/flush`` 只写出在该节点上次开始写出处理(定期执行或 ``_configsync/flush`` )之后登记的文件。恢复的 ``configsync`` 文件带有在拍摄快照的集群上登记时的时间,因此在 OpenSearch 启动约 1 分钟后首次定期写出完成之后,不重启而执行 ``_configsync/flush`` 不会写出任何词典文件,步骤 3 之后集群会变为 red。重启会重置该时间点,因此重启后执行 ``_configsync/flush`` 即可写出所有词典文件。每写出一个文件,OpenSearch 日志中都会输出 ``ConfigSyncService`` 的 ``Updated {文件路径}``。

3. 恢复 |Fess| 的索引。

::
Expand Down Expand Up @@ -398,7 +403,7 @@

执行 ``_cluster/reroute?retry_failed=true`` 或仅放置词典文件都无法恢复。恢复失败的分片在其索引被关闭或删除并重新恢复之前不会被分配。请按以下步骤恢复。

1. 按照“恢复所有索引”的步骤 1 和 2,恢复 ``configsync`` 并写出词典文件。
1. 按照“恢复所有索引”的步骤 1 和 2,恢复 ``configsync``,重启 OpenSearch 并写出词典文件。
2. 确认 ``red`` 状态的索引。

::
Expand Down
Loading