java.lang.IllegalArgumentException: Too large (805306401 expected elements with load factor 0.75 in VMware GemFire
Article ID: 294335
Updated On:
Products
VMware Tanzu Gemfire
Issue/Introduction
This issue is applicable to all released versions of VMware GemFire.
The cache server fails to restart with:
The cache server fails to restart with:
[error 2020/05/15 10:38:50.993 EDT psin8p021_cache2 tid=0x1] Cache initialization failed because: java.lang.IllegalArgumentException: Too large (805306401 expected elements with load factor 0.75)
Environment
Product Version: 9.5
Resolution
This is a known VMware GemFire Product issue (GEODE-8029) and the fix will be available in the future VMware GemFire releases.
Here are the suggested workarounds, before the official fix is released.
To prevent the issue from happening, periodically run offline compaction of the disk-stores. Please make sure to run compaction before hitting the issue, otherwise the same exception will be thrown because we need to load all disk-stores files into memory to determine what's needed and what's not.
As a workaround and until the official fix is available, try to start the failing member with an empty disk-store so it can recover the entire data set from other running members. In order to do that, move the failing server's entire disk-store directory to another directory (as a backup) and try to start the server from scratch.
IMPORTANT NOTE: Before executing the above step, it's extremely important to double-check that other members hosting the data for all of the implicated regions are fully up and running.
Here are the suggested workarounds, before the official fix is released.
To prevent the issue from happening, periodically run offline compaction of the disk-stores. Please make sure to run compaction before hitting the issue, otherwise the same exception will be thrown because we need to load all disk-stores files into memory to determine what's needed and what's not.
As a workaround and until the official fix is available, try to start the failing member with an empty disk-store so it can recover the entire data set from other running members. In order to do that, move the failing server's entire disk-store directory to another directory (as a backup) and try to start the server from scratch.
IMPORTANT NOTE: Before executing the above step, it's extremely important to double-check that other members hosting the data for all of the implicated regions are fully up and running.
Feedback
Yes
No
Powered by
