Slow to shutdown

[naoliv]

Sometimes I see some slowness while closing/shutting down JOSM:

[INFO] IndexedDiskCache - Region [TMS] Beginning Optimization #1
[INFO] IndexedDiskCache - Region [TMS] Defragmentation took 4872ms.. File Size (before=388067685) (after=388067685) (truncating to 384117881)
[INFO] IndexedDisk - Truncating file [/home/naoliv/.josm/cache/tms/TMS.data] to 384117881
[INFO] IndexedDiskCache - Region [TMS] Finished #1 Optimization took 4923ms.
[INFO] IndexedDiskCache - Region [TMS] Saving keys to: TMS, key count: 25000
[INFO] IndexedDiskCache - Region [TMS] Finished saving keys.
[INFO] IndexedDiskCache - Region [TMS] Shutdown complete.
[INFO] CompositeCache - In DISPOSE, [TMS] disposing of memory cache.
[INFO] AbstractMemoryCache - Memory Cache dispose called.

Couldn't we have some kind of message saying that JOSM is shutting down, with maybe a progress bar or other element displaying something like "Doing a task 1/3"? (where 1/3 is the current step)

JOSM

Repository Root: http://josm.openstreetmap.de/svn
Build-Date: 2015-06-13 01:31:55
Last Changed Author: wiktorn
Revision: 8488
Repository UUID: 0c6e7542-c601-0410-84e7-c038aed88b3b
Relative URL: ^/trunk
URL: http://josm.openstreetmap.de/svn/trunk
Last Changed Date: 2015-06-12 22:51:43 +0200 (Fri, 12 Jun 2015)
Last Changed Rev: 8488

Identification: JOSM/1.5 (8488 pt_BR) Linux Debian GNU/Linux unstable (sid)
Memory Usage: 247 MB / 4029 MB (131 MB allocated, but free)
Java version: 1.7.0_79, Oracle Corporation, OpenJDK 64-Bit Server VM
Java package: openjdk-7-jre:amd64-7u79-2.5.5-1
VM arguments: [-Dawt.useSystemAAFontSettings=on]

Plugins:
- AddrInterpolation (31241)
- Create_grid_of_ways (31241)
- FastDraw (31241)
- FixAddresses (31241)
- OpeningHoursEditor (31241)
- PicLayer (31241)
- SimplifyArea (31241)
- buildings_tools (31241)
- download_along (31241)
- editgpx (31241)
- geotools (31126)
- graphview (31241)
- jts (31126)
- measurement (31241)
- merge-overlap (31241)
- opendata (31241)
- pdfimport (31241)
- poly (31241)
- reverter (31241)
- tagging-preset-tester (31241)
- todo (29154)
- turnrestrictions (31241)
- undelete (31241)

[Don-vip]

likely a problem I introduced in r8412.
But I assume it is a good thing to call JCSCacheManager.shutdown(); no?

[wiktorn]

Yes it's to both.

I plan just to move to BlockDiskCache, where there is no need to defragment the cache file. Though it will crash the existing cache file, so I plan to group this change with other with such effect.

[Don-vip]

skip milestone 15.06

[Don-vip]

Milestone renamed

[simon04]

Postpone enhancements to next release.

[simon04]

Unschedule tickets without progress

[wiktorn]

Resolution blocked till ​https://issues.apache.org/jira/browse/JCS-153 is solved

[wiktorn]

In 9064/josm:

Move from IndexedDiskCache to BlockDiskCache.

HEADSUP: This change trashes contents of the cache files silently (and there's nothing easy we can do about it).

Thanks to Thomans Vandahl at Apache Commons (​https://issues.apache.org/jira/browse/JCS-153) the problem with restoring cache file corruptions is finally solved.

This means, that ot no time cache file optimization will be needed, thus reducing the shutdown time of JCS Caches.

Closes: #11566

[wiktorn]

Reopened due to [9142]

[Don-vip]

Are we blocked by a JCS issue for this ticket? If yes, which one?

[wiktorn]

Right now there is none that I'm aware. Latest was ​https://issues.apache.org/jira/browse/JCS-159

I plan to do some checks, and introduce the change at next release cycle.

[Don-vip]

Milestone renamed

[Don-vip]

Any news on this subject?

[wiktorn]

I did some tests, and I didn't find any problems, but I forgot to introduce the change at the beginning of release cycle, my bad.

I'll commit the change just after current release.

[wiktorn]

In 10169/josm:

Restore IndexDiskCache that was switched off in [9142]

​https://issues.apache.org/jira/browse/JCS-159 is already fixed

Closes: #11566 (hopefully)