I just upgraded to 2.4.1 today, and found the same behavior you did. I know that I can just set the "zFileSystemMapIgnoreNames" or "zFileSystemMapIgnoreTypes" zProperties to tell Zenoss to ignore these new "filesystems", but let's be honest--is that really a solution? Doesn't that just disable the new functionality that the Zenoss team has built in? Don't you think that Zenoss should have seen this kind of problem in their test environments? I came back from lunch and had 200-some-odd events from just about every Linux box in my environment stating that they were low on "Real Memory". Just like HummerBoy states, Linux will use as much memory as possible for caching purposes (and Windows is doing better on this front as well), so OF COURSE those pseudo-filesystems will be more than 90% utilized on any system that is doing anything useful.
I believe this to be unacceptable behavior, and I should not have to implement my own workarounds for something that should have been tested before being released. However, that's exactly what I'll be doing, which kind of defeats the purpose of this new functionality, doesn't it? --seth P.S. I'm sorry if I sound angry, but I have yet to have a Zenoss upgrade go smoothly. Either the upgrade re-enables monitoring globally for a bunch of services that I have to keep disabling, or something like this happens. -------------------- m2f -------------------- Read this topic online here: http://forums.zenoss.com/viewtopic.php?p=36275#36275 -------------------- m2f -------------------- _______________________________________________ zenoss-users mailing list [email protected] http://lists.zenoss.org/mailman/listinfo/zenoss-users
