After 20 hours with the same usage as before, I still haven't seen any swapping (well, 516 bytes, but that's ok. Odd, but ok). I tried mozilla, 3D gaming with q3a, hg2 and others, vmware, kwintv, everything I could think of that had 'stolen' memory from me before, or seemed to, but it all seems to work now. Of course, Murphy's law says that the system will crash 2 seconds after I send this, but take it for what it's worth :) This is with 2.4.13-3 btw. I didn't see any -5 in mantel's directory. //Anders On Thursday 01 November 2001 09.09, Ben Rosenberg wrote:
* Konstantin (Kastus) Shchuka (kastus@tsoft.com) [011031 18:25]: ->On Thu, Nov 01, 2001 at 03:17:32AM +0100, Anders Johansson wrote: ->> It may be too early to say for certain, but I just upgraded to 2.4.13-3 from ->> mantel/next and I've been running for an hour now without a single swap-out, ->> and memory usage actually looks reasonable. For me this is a record, since ->> these problems started. Maybe... ->> -> ->Just looked at Mantel's directory, it's 2.4.13-5 already. And it's been -1 ->just on Sunday... ->
I need to do some checking..but as I have read other places ..the fix for this bug is present in the 2.4.14pre6 release. Hubert may have taken the code from this pre6 and put it into his .13 kernel. We shall see. Right now I'm running .10 from the 7.3 update directory on my 7.1 and 7.2 boxes..it seems to not freak out and dump to swap as often as .11, .12 and .13. Before I switched to .10 today I was running .13-3 and I was humming along with 0M of swap used and then *BAM* 158M's eaten in like 3 minutes...everything came to a stand still..which is why I backed down to .10 ..
-----=====-----=====-----=====-----=====----- Ben Rosenberg mailto:ben@whack.org -----=====-----=====-----=====-----=====----- "Technological progress is like an axe in the hands of a pathological criminal" -AE