Am 19.10.2020 um 16:17 schrieb Julian Reschke:
Am 19.10.2020 um 08:37 schrieb Tanvi Shah:
Hi ,
We have started Se Garbage Collection and S3 size is more than 2 TB.
We are facing memory issues while executing GC even when we have given
11 GB of memory to the application.
Code is:
final MarkSweepGarbageCollector gc =
documentNodeStore.createBlobGarbageCollector(seconds,
repository.toString(), wb, new
DefaultStatisticsProvider(Executors.newScheduledThreadPool(1)));
gc.collectGarbage(markOnly);
final OperationsStatsMBean stats = gc.getOperationStats();
log.info(
"number deleted : " + stats.numDeleted() + " Size deleted : " +
stats.sizeDeleted());
The exceptions are :
...
It might be necessary to set the fetch size (see
<https://stackoverflow.com/questions/3682614/how-to-read-all-rows-from-huge-table>,
around
<https://github.com/apache/jackrabbit-oak/blob/5af43cb29fb6cfc226eac6f89bf4985d7ce6d89f/oak-store-document/src/main/java/org/apache/jackrabbit/oak/plugins/document/rdb/RDBDocumentStoreJDBC.java#L590>).
Sorry, can't test that right now myself.
Best regards, Julian
...or you could try to set the defaultRowFetchSize as parameter in the
JDBC URL (see
<https://jdbc.postgresql.org/documentation/head/connect.html#connection-parameters>).
Best regards, Julian
