Getting message in logs "Maximum memory usage reached (6251610112), cannot reserve size of 4194304"

Question

question

stephen9834 asked Dec 8, '20 Erick Ramirez commented Dec 13, '20

Getting message in logs "Maximum memory usage reached (6251610112), cannot reserve size of 4194304"

Below Message seen in logs

INFO  [CoreThread-0] 2020-12-08 12:34:30,743  NoSpamLogger.java:95 - Maximum memory usage reached (6251610112), cannot reserve size of 4194304
DEBUG [CoreThread-0] 2020-12-08 12:34:30,744  NoSpamLogger.java:92 - Requested buffer size 4.000KiB has been allocated directly due to buffer pool exhausted
INFO  [CoreThread-2] 2020-12-08 12:35:00,800  NoSpamLogger.java:95 - Maximum memory usage reached (6251610112), cannot reserve size of 4194304
DEBUG [CoreThread-2] 2020-12-08 12:35:00,801  NoSpamLogger.java:92 - Requested buffer size 4.000KiB has been allocated directly due to buffer pool exhausted
INFO  [CoreThread-1] 2020-12-08 12:35:32,442  NoSpamLogger.java:95 - Maximum memory usage reached (6251610112), cannot reserve size of 4194304
DEBUG [CoreThread-1] 2020-12-08 12:35:32,442  NoSpamLogger.java:92 - Requested buffer size 4.000KiB has been allocated directly due to buffer pool exhausted
INFO  [CoreThread-1] 2020-12-08 12:36:03,738  NoSpamLogger.java:95 - Maximum memory usage reached (6251610112), cannot reserve size of 4194304
DEBUG [CoreThread-1] 2020-12-08 12:36:03,738  NoSpamLogger.java:92 - Requested buffer size 4.000KiB has been allocated directly due to buffer pool exhausted
INFO  [CoreThread-1] 2020-12-08 12:36:40,741  NoSpamLogger.java:95 - Maximum memory usage reached (6251610112), cannot reserve size of 4194304
DEBUG [CoreThread-1] 2020-12-08 12:36:40,742  NoSpamLogger.java:92 - Requested buffer size 4.000KiB has been allocated directly due to buffer pool exhausted

Not able to understand this message could you help in this ?

jvm

10 |1000

Attachments: Up to 8 attachments (including images) can be used with a maximum of 1.0 MiB each and 10.0 MiB total.

DataStax Enterprise is powered by the best distribution of Apache Cassandra ™

Apache, Apache Cassandra, Cassandra, Apache Tomcat, Tomcat, Apache Lucene, Lucene, Apache Solr, Apache Hadoop, Hadoop, Apache Spark, Spark, Apache TinkerPop, TinkerPop, Apache Kafka and Kafka are either registered trademarks or trademarks of the Apache Software Foundation or its subsidiaries in Canada, the United States and/or other countries.

Privacy Policy Terms of Use

Answer 1 · 2020-12-09T00:28:08Z

Erick Ramirez answered Dec 9, '20 Erick Ramirez commented Dec 13, '20

The entry gets logged when the memory buffer for the file cache is full so requests to reserve memory from the pool can't be satisfied. This is indicative that read requests are filling up the file cache quicker than the buffer segments can be recycled.

The size available to the file buffers is determined by the file cache size parameter file_cache_size_in_mb. You can increase it by explicitly setting the size in cassandra.yaml or setting the JVM option -XX:MaxDirectMemorySize.

For details, see Tuning the JVM. Cheers!

2 Share

10 |1000

Attachments: Up to 8 attachments (including images) can be used with a maximum of 1.0 MiB each and 10.0 MiB total.

stephen9834 commented · Dec 11, 2020 at 06:58 PM

Thanks for super quick response !! @Erick Ramirez we have enough free ram but still we are continuously observing this message in logs and we haven't defined file_cache_size_in_mb parameter in cassandra.yml . could you guide on how to decide value of this parameter ?

0 ·

Erick Ramirez ♦♦ stephen9834 commented · Dec 13, 2020 at 11:27 PM

I can only provide you general guidance in this Q&A forum. See the document I linked.

As I stated in other posts, if you need a recommendation specific to your environment as a DSE customer then you will need to log a ticket with DataStax Support so one of our engineers can assist you. This forum isn't a replacement for support. Cheers!

0 ·

question