This is an automated email from the ASF dual-hosted git repository.
adoroszlai pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/ozone-site.git
The following commit(s) were added to refs/heads/master by this push:
new d3969c9cac HDDS-10685. [Auto] Update configuration documentation (#520)
d3969c9cac is described below
commit d3969c9cac6a5a6d3facbe3fd0b3399f8abd0fcf
Author: asf-ci-deploy <[email protected]>
AuthorDate: Sat Aug 1 17:43:32 2026 +0200
HDDS-10685. [Auto] Update configuration documentation (#520)
---
docs/05-administrator-guide/02-configuration/99-appendix.md | 11 +++++++++--
1 file changed, 9 insertions(+), 2 deletions(-)
diff --git a/docs/05-administrator-guide/02-configuration/99-appendix.md
b/docs/05-administrator-guide/02-configuration/99-appendix.md
index 5fbd681ffd..2ff2ed344f 100644
--- a/docs/05-administrator-guide/02-configuration/99-appendix.md
+++ b/docs/05-administrator-guide/02-configuration/99-appendix.md
@@ -164,9 +164,11 @@ This page provides a comprehensive overview of the
configuration keys available
| `hdds.datanode.read.threadpool` | 10 | `OZONE`, `HDDS`, `PERFORMANCE` | The
number of threads in RPC server reading from the socket for Datanode client
service endpoints. This config overrides Hadoop configuration
"ipc.server.read.threadpool.size" for HddsDatanodeClientProtocolServer. The
default value is 10. |
| `hdds.datanode.recovering.container.scrubbing.service.interval` | 1m |
`SCM`, `DELETION` | Time interval of the stale recovering container scrubbing
service. The recovering container scrubbing service runs on Datanode
periodically and deletes stale recovering container Unit could be defined with
postfix (ns,ms,s,m,h,d). |
| `hdds.datanode.replication.outofservice.limit.factor` | 2.0 | `DATANODE`,
`SCM` | Decommissioning and maintenance nodes can handle morereplication
commands than in-service nodes due to reduced load. This multiplier determines
the increased queue capacity and executor pool size. |
+| `hdds.datanode.replication.per.volume.enabled` | false | `DATANODE` | When
true, push-based container replication uses a separate replication handler
thread pool per data volume so that slow replication on one disk does not block
replication on other disks. Pull replication and other replication tasks
continue to use the global replication handler thread pool. |
+| `hdds.datanode.replication.per.volume.streams.limit` | 2 | `DATANODE` | When
`hdds.datanode.replication.per.volume.enabled` is true, maximum concurrent push
replication commands per data volume (each volume has its own handler thread
pool; effective push parallelism on the datanode is roughly the number of
volumes times this limit, with outofservice.limit.factor applied per pool on
decommissioning or maintenance nodes). Push replication is usually disk-bound,
so one or two concurrent t [...]
| `hdds.datanode.replication.port` | 9886 | `DATANODE`, `MANAGEMENT` | Port
used for the server2server replication server |
| `hdds.datanode.replication.queue.limit` | 4096 | `DATANODE` | The maximum
number of queued requests for container replication |
-| `hdds.datanode.replication.streams.limit` | 10 | `DATANODE` | The maximum
number of replication commands a single datanode can execute simultaneously |
+| `hdds.datanode.replication.streams.limit` | 10 | `DATANODE` | Sets both the
base size of the global replication handler executor and the inbound
replication server executor. The global executor is subject to
outofservice.limit.factor scaling. When
`hdds.datanode.replication.per.volume.enabled` is false (default), all
source-side replication tasks use the global executor. When per.volume.enabled
is true, per-volume executors handle normal source-side push tasks, while this
limit still a [...]
| `hdds.datanode.replication.work.dir` | | `DATANODE` | This configuration is
deprecated. Temporary sub directory under each `hdds.datanode.dir` will be used
during the container replication between datanodes to save the downloaded
container(in compressed format). |
| `hdds.datanode.rocksdb.auto-compaction-small-sst-file` | true | `DATANODE` |
Auto compact small SST files
(rocksdb.auto-compaction-small-sst-file-size-threshold) when count exceeds
(rocksdb.auto-compaction-small-sst-file-num-threshold) |
| `hdds.datanode.rocksdb.auto-compaction-small-sst-file-num-threshold` | 512 |
`DATANODE` | Auto compaction will happen if the number of small SST files
exceeds this threshold. |
@@ -379,7 +381,10 @@ This page provides a comprehensive overview of the
configuration keys available
| `ozone.client.max.retries` | 5 | `CLIENT` | Maximum number of retries by
Ozone Client on encountering exception while writing a key |
| `ozone.client.read.max.retries` | 3 | `CLIENT` | Maximum number of retries
by Ozone Client on encountering connectivity exception when reading a key. |
| `ozone.client.read.retry.interval` | 1 | `CLIENT` | Indicates the time
duration in seconds a client will wait before retrying a read key request on
encountering a connectivity exception from Datanodes. By default the interval
is 1 second |
-| `ozone.client.read.timeout` | 30s | `OZONE`, `CLIENT`, `MANAGEMENT` |
Timeout for ozone grpc client during read. |
+| `ozone.client.read.short-circuit` | false | `CLIENT`, `DATANODE` | Whether
read short-circuit is enabled or not |
+| `ozone.client.read.short-circuit.buffer.size` | 128KB | `CLIENT`, `DATANODE`
| Buffer size of reader/writer. |
+| `ozone.client.read.short-circuit.disable.interval` | 600 | `CLIENT` | If
some unknown IO error happens on Domain socket read, short circuit read will be
disabled temporarily for this period of time(seconds). |
+| `ozone.client.read.timeout` | 30s | `OZONE`, `CLIENT`, `MANAGEMENT` |
Timeout for ozone grpc and short-circuit client during read. |
| `ozone.client.retry.interval` | 0 | `CLIENT` | Indicates the time duration a
client will wait before retrying a write key request on encountering an
exception. By default there is no wait |
| `ozone.client.server-defaults.validity.period.ms` | 3600000 | `OZONE`,
`CLIENT`, `SECURITY` | The amount of milliseconds after which cached server
defaults are updated. By default this parameter is set to 1 hour. Support
multiple time unit suffix(case insensitive). If no time unit is specified then
milliseconds is assumed. |
| `ozone.client.socket.timeout` | 5000ms | `OZONE`, `CLIENT` | Socket timeout
for Ozone client. Unit could be defined with postfix (ns,ms,s,m,h,d) |
@@ -395,6 +400,7 @@ This page provides a comprehensive overview of the
configuration keys available
| `ozone.client.stream.readblock.enable` | false | `CLIENT` | Allow ReadBlock
to stream all the readChunk in one request. |
| `ozone.client.verify.checksum` | true | `CLIENT` | Ozone client to verify
checksum of the checksum blocksize data. |
| `ozone.client.wait.between.retries.millis` | 2000 | | Expert only. The time
to wait, in milliseconds, between retry attempts to contact OM. Wait time
increases linearly if same OM is retried again. If retrying on multiple OMs
proxies in round robin fashion, the wait time is introduced after all the OM
proxies have been attempted once. |
+| `ozone.client.write.timeout` | 30s | `OZONE`, `CLIENT`, `MANAGEMENT` |
Timeout for ozone short-circuit client during write. |
| `ozone.container.cache.lock.stripes` | 1024 | `PERFORMANCE`, `CONTAINER`,
`STORAGE` | Container DB open is an exclusive operation. We use a stripe lock
to guarantee that different threads can open different container DBs
concurrently, while for one container DB, only one thread can open it at the
same time. This setting controls the lock stripes. |
| `ozone.container.cache.size` | 1024 | `PERFORMANCE`, `CONTAINER`, `STORAGE`
| The open container is cached on the data node side. We maintain an LRU cache
for caching the recently used containers. This setting controls the size of
that cache. |
| `ozone.default.bucket.layout` | | `OZONE`, `MANAGEMENT` | Default bucket
layout used by Ozone Manager during bucket creation when a client does not
specify the bucket layout option. Supported values are OBJECT_STORE and
FILE_SYSTEM_OPTIMIZED. OBJECT_STORE: This layout allows the bucket to behave as
a pure object store and will not allow interoperability between S3 and FS APIs.
FILE_SYSTEM_OPTIMIZED: This layout allows the bucket to support atomic
rename/delete operations and also allo [...]
@@ -806,6 +812,7 @@ This page provides a comprehensive overview of the
configuration keys available
| `ozone.snapshot.key.deleting.limit.per.task` | 20000 | `OM`, `PERFORMANCE`,
`DELETION` | The maximum number of deleted keys to be scanned by Snapshot
Deleting Service per snapshot run. |
| `ozone.ssl.enabled.protocols` | | `OZONE`, `SECURITY`, `CRYPTO_COMPLIANCE`
| The supported SSL protocols used to restrict connections towards the WebUI of
different components, and the S3 GateWay. |
| `ozone.sst.filtering.service.timeout` | 300000ms | `OZONE`, `PERFORMANCE`,
`OM`, `DELETION` | A timeout value of sst filtering service. |
+| `ozone.tracing.client.application-aware` | true | `OZONE`, `HDDS` | Only
effective when ozone.tracing.enabled=false. When true, Ozone will continue an
application-supplied trace (via GlobalOpenTelemetry or a wire-propagated
context) as child spans, but will NOT start a new root trace on its own. |
| `ozone.tracing.enabled` | false | `OZONE`, `HDDS` | If true, tracing is
initialized and spans may be exported (subject to sampling). |
| `ozone.tracing.endpoint` | | `OZONE`, `HDDS` | OTLP gRPC receiver endpoint
URL. |
| `ozone.tracing.sampler` | -1 | `OZONE`, `HDDS` | Root trace sampling ratio
(0.0 to 1.0). |
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
