Переглянути джерело

docs: Add a note about tx metadata affecting blockstore size (#6007)

There have been a good number of people asking why their node with
--enable-rpc-transaction-data takes up more space than the 500 GB value
listed in the docs
steviez 6 місяців тому
батько
коміт
d889ceabb4
1 змінених файлів з 10 додано та 5 видалено
  1. 10 5
      docs/src/operations/guides/validator-start.md

+ 10 - 5
docs/src/operations/guides/validator-start.md

@@ -314,11 +314,11 @@ the validator to ports 11000-11020.
 ### Limiting ledger size to conserve disk space
 ### Limiting ledger size to conserve disk space
 
 
 The `--limit-ledger-size` parameter allows you to specify how many ledger
 The `--limit-ledger-size` parameter allows you to specify how many ledger
-[shreds](https://solana.com/docs/terminology#shred) your node retains on disk. If you do not
-include this parameter, the validator will keep all received ledger data
-until it runs out of disk space. Otherwise, the validator will continually
-purge the oldest data once to stay under the specified `--limit-ledger-size`
-value.
+[shreds](https://solana.com/docs/terminology#shred) your node retains on disk.
+If you do not include this parameter, the validator will keep all received
+ledger data until it runs out of disk space. Otherwise, the validator will
+periodically purge the oldest data (FIFO) to remain under the specified
+`--limit-ledger-size` value.
 
 
 The default value attempts to keep the blockstore (data within the rocksdb
 The default value attempts to keep the blockstore (data within the rocksdb
 directory) disk usage under 500 GB. More or less disk usage may be requested
 directory) disk usage under 500 GB. More or less disk usage may be requested
@@ -333,6 +333,11 @@ These items may include (but are not limited to):
 - Persistent accounts index
 - Persistent accounts index
 - Snapshots
 - Snapshots
 
 
+Additionally, specifying `--enable-rpc-transaction-history` will store extra
+block and transaction metadata. The space required to store this data varies
+with cluster activity, and is hard to account for. Thus, using this flag will
+likely cause the 500 GB target to be exceeded.
+
 ### Systemd Unit
 ### Systemd Unit
 
 
 Running the validator as a systemd unit is one easy way to manage running in the
 Running the validator as a systemd unit is one easy way to manage running in the