EHC ❯ Use threadpool for disk-expire-tasks

• Type: Bug
• Status: Open
• Priority: 2 Major
• Resolution:
• Component/s: ehcache-core
• Labels:

• Assignee: cdennis
• Reporter: sourceforgetracker
• Created: September 21, 2009
• Votes: 1
• Watchers: 3
• Updated:

  October 11, 2011
• Resolved:

Description

I’m using ehcache as hibernate 2nd level cache. Currently for all of the entities >50 a disk-expire-thread is acquired, which effectivly do nothing the whole time, but consuming valuable system-resources. Sourceforge Ticket ID: 2044930 - Opened By: mcbain4711 - 10 Aug 2008 09:43 UTC

Comments

gluck 2009-09-25

Yes this is a good idea.

Steve Harris 2009-10-15

is this also an issue in Terracotta cache

Manuel Dominguez Sarmiento 2010-05-24

This issue has Fix version marked as 2.0.0 but I can’t find the relevant change anywhere in the codebase (checking version 2.1.0) - has this been incorporated?

Fiona OShea 2010-06-18

This was originally targeted for 2.0 but has been rescheduled to a future release.

Manuel Dominguez Sarmiento 2011-08-01

Any plans regarding this? We use a huge number of DiskStore-backed caches and EhCache spawns a thread per cache that does basically nothing most of the time. This consumes resources (OS threads and stack memory for each thread, which defaults to 1 MB in 64 bit JVMs, though can be tuned with Xss JVM parameter).