fapifta commented on code in PR #11434: URL: https://github.com/apache/ozone/pull/11434#discussion_r4324323641
########## hadoop-hdds/docs/content/design/append.md: ########## @@ -0,0 +1,1248 @@ +--- +title: Append support in Ozone +summary: Reopen an existing file for a single appender while preserving its committed data and snapshot references. +date: 2026-10-02 +jira: HDDS-4333 +status: draft +author: Siyao Meng +--- +<!-- + Licensed to the Apache Software Foundation (ASF) under one or more + contributor license agreements. See the NOTICE file distributed with + this work for additional information regarding copyright ownership. + The ASF licenses this file to You under the Apache License, Version 2.0 + (the "License"); you may not use this file except in compliance with + the License. You may obtain a copy of the License at + + http://www.apache.org/licenses/LICENSE-2.0 + + Unless required by applicable law or agreed to in writing, software + distributed under the License is distributed on an "AS IS" BASIS, + WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. + See the License for the specific language governing permissions and + limitations under the License. +--> + +## Status and tracking + +This is a proposal for discussion, not an implemented or accepted feature. The code baseline is Apache Ozone master +`29b9853270342eea4d065b1f14c332ad384f83e5`. The behavioral choices are specified below; remaining implementation details +and activation checks are listed in the [implementation plan](#implementation-plan-and-activation). + +[HDDS-4333](https://issues.apache.org/jira/browse/HDDS-4333), "Ozone supports append operation", is the existing append +umbrella. Its design subtask is [HDDS-4373](https://issues.apache.org/jira/browse/HDDS-4373). The older combined append/truncate +issue is [HDDS-3714](https://issues.apache.org/jira/browse/HDDS-3714); its append subtask, +[HDDS-4240](https://issues.apache.org/jira/browse/HDDS-4240), was resolved as a duplicate with a pointer to HDDS-4333. +The earlier [design PR #1515](https://github.com/apache/ozone/pull/1515) was closed without merging. + +This draft incorporates the useful requirements and review discussion from PR #1515, the new-block approach from +Wei-Chiu Chuang's later append proposal, and the review questions summarized below. It does not imply that the earlier +authors or reviewers endorse these new decisions. + +Draft prepared with assistance from Codex (GPT-6 Astra) and Claude Code (Fable 5.1). + +## Problem statement + +OzoneFS currently rejects `FileSystem.append()`. Applications that reopen transaction logs or resume existing output files +cannot use that operation through `ofs://` or `o3fs://`. Applications may have additional requirements such as `truncate()`; +adding append alone does not establish compatibility with every HDFS application. + +The primary use cases are reopening transaction logs during recovery and resuming existing output files. Log writers can +keep a stream open across many small writes and frequent `hflush()`/`hsync()` calls. In the inspected Solr 8.11.4 path, append Review Comment: SolR10 has removed HDFS support as a file storage backend, and also SIP-20/SOLR-17125 discarded the idea to put the SolR indexes to an Object Store. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
