Printable View
I was really hoping they'd gotten this straightened out. *sigh*
I started noticing that several of my machines running the DF client were starting to drag. Upon closer examination, I found this:
That's right, my two folding clients eating up nearly 1gb of RAM. Ugh. I've since changed every one of my clients to use less RAM (-rf). This is going to hurt my production, but I just don't have the time or interest to micro manage a bunch of machines because the client software can't behave. :(
I expect that is the reason that my diskless nodes keep falling over. At this point, I'm just replacing DF with something else when it happens. Still have a couple running, but it doesn't appear they will last through the weekend.
yes, i am restarting my windows clients a couple times a week to clear out the memory
stopping and restarting the service works for me
And what is Howard's response?
**NM..I see it.
I don't think that this ever stopped happening. It was just not as critical on the smaller proteins. The OSX client used to do this really bad, I had a cron job that stopped the client in the middle of the night and then restarted it a minute later.
Someone should really DO something about all the people falling victim to cron jobs! :p
Man no shiznit. Been running DF for 3.5 days and it had eaten 45MB extra RAM till I restarted just now.
Hope the new client released today fixes this.
From what people are saying it the new one seems to be working well.