you might expect.
You need to do the lookup in a second txn, that starts unambiguously *after* the updating txn ends.
Michał Mosiewicz wrote:
I've been testing cache. I've noticed there is much less cache hits that I supposed it should be. Then I discovered the following behaviour:
DEBUG [Thread-0] (ReadWriteCache.java:68) - Cache lookup: 40201 DEBUG [Thread-0] (ReadWriteCache.java:78) - Cache hit: 40201 DEBUG [Thread-0] (ReadWriteCache.java:106) - Invalidating: 40201 DEBUG [Thread-0] (ReadWriteCache.java:178) - Releasing: 40201 DEBUG [Thread-0] (ReadWriteCache.java:68) - Cache lookup: 40201 DEBUG [Thread-0] (ReadWriteCache.java:87) - Cached item was locked: 40201
I've grepped only one record. Here is what it reads - first there is a cache hit, than item 40201 is updated, so it's invalidated in cache. Then it's released. At next lookup it appears locked. Why?
When I looked at the code I thought that releasing is to put back the item into cache after it's commited. But it seems that it still remains locked after it's released. Is ot intentional behaviour?
Or - is there any cache that saves the database lookup after record is updated? I mean my configuration is for single machine only - I don't need to bother that my database gets inconsistent.
Another question - why records are flushed out the cache, efectively forcing database reread after they are updated in nonstrict-read-write. Is it also intentional? I think that after hibernate updates (successfully) database it should be assumed to have the most valid row copy. Flushing it out and forcing a reread is hard to understand for me, unless we assume that row is processed at db server causing potential insanity.
I'm hibernated to 2.1.6.
-- Mike
-------------------------------------------------------
This SF.Net email is sponsored by: thawte's Crypto Challenge Vl
Crack the code and win a Sony DCRHC40 MiniDV Digital Handycam
Camcorder. More prizes in the weekly Lunch Hour Challenge.
Sign up NOW http://ad.doubleclick.net/clk;10740251;10262165;m
_______________________________________________
hibernate-devel mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/hibernate-devel
-- Gavin King +61 410 534 454 +1 404 822 8349 callto://gavinking
Hibernate [EMAIL PROTECTED] http://hibernate.org
JBoss Inc [EMAIL PROTECTED] http://jboss.com
------------------------------------------------------- This SF.Net email is sponsored by: YOU BE THE JUDGE. Be one of 170 Project Admins to receive an Apple iPod Mini FREE for your judgement on who ports your project to Linux PPC the best. Sponsored by IBM. Deadline: Sept. 24. Go here: http://sf.net/ppc_contest.php _______________________________________________ hibernate-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/hibernate-devel
