On Thu, Mar 25, 2010 at 11:22:44AM -0400, Benjamin Long wrote: > > are you at least able to reproduce this memory consumption > > problem? > > I had a problem like this with 3.0.2 and the Quickread translator. That is a > known issue. Do you have the same problem with the latest version of Gluster > and no performance translators? I think your email says you went back to 2.x > before removing the translators. I went back to 2.x after I saw this old problem (without any performance translators) under 3.0.3. Then I made the test case posted to the list and it was with 2.0.10rc1, but I'm pretty sure it would be the same with 3.x. The one difference was that I saw the problem under 3.0.3 with my production system (that does not and did not use performance translators and is a volume unified over six subvolumes, each of which is 2x replicated), not with the test case. While setting up the glusterfs filesystem, I decided that performance translators may be tried out _after_ everything works OK without them. But it is still not the case. If you would find it useful, I may repeat the test with 3.0.3 tomorrow and post the result. But i think it is time to look into the code. I tried even once upon a time, but ended with a segmentation fault, not finding anything :(. Regards Krzysztof