R. David Murray: Since this is a very heavily used Zope install, I am getting trashed faster than you can say 'leakage'.
On Mon, 22 May 2000, Jason Spisak wrote:
i did it again, and now I'm down to 160 objects total. That's more liek it. But system memory is still dropping. I was at 173632 available (This is only a half our after the Zope restart BTW. It ate 100MB in under that time.) before the second pass and now I'm at 173616.
I'm *very* interested in this.
I have a similar setup (stock 2.1.4 + the Lexicon.py patch, no external database, no non-bundled Products, custom CatalogAware ZClass, large number of objects (~60,000). I'm running under FreeBSD 3.4-STABLE.
I tried the 'minimize' trick and got down to around 200 objects, with *no* decrease in memory used. I'm not leaking memory as fast as you, but over time memory usage does go up until swapping starts and I'm forced to restart Zope.
Also as (apparently) with you, manage_debug does *not* give any clues as to what is taking the memory; everything in the list looks normal and does not add up to the observed memory usage as far as I can guess.
I'm glad to see someone else is observing this memory weirdness, I thought I was the only one.
--RDM
I'm anxious to discover what's going on too. It's will really be a boon for DC to have Zope scalable to this size of ZODB and have the stability I've come to rely on. All my best, Jason Spisak CIO HireTechs.com 6151 West Century Boulevard Suite 900 Los Angeles, CA 90045 P. 310.665.3444 F. 310.665.3544 Under US Code Title 47, Sec.227(b)(1)(C), Sec.227(a)(2)(B) This email address may not be added to any commercial mail list with out my permission. Violation of my privacy with advertising or SPAM will result in a suit for a MINIMUM of $500 damages/incident, $1500 for repeats.