105500 Views Previous 1 … 15 16 17 18 19 Next 279 Replies Latest reply: Feb 5, 2010 12:11 PM by gmillerpenns Go to original post
I had the same problem with Mac Server 10.5.8 and 70 network users logged at the same time.
I resolved the problem quickly by transferring the network users cache to the local Imac drive. Your server needs to be configured as advanced mode so you can use Work Group Manager. For instructions, go to Page 10"Reducing afp load on your servers" of this PDF http://www.afp548.com/filemgmt_data/files/Leopard%20Server%20Quickstart%20Guide. pdf.
Cons: It takes a bit longer for the user to log in...don't know why!
Pros: Average CPU load passed from 100% to 10%
are there some experience with 10.6.x server and RAID Level 5/6 ?
still causing cpu over load for AppleFileServer ?
since we sold our Tiburon RAID, we still working with 3 internal striped 1.5TB HD's ( server 10.5.6 is running on a macpro )
it isn't the best solution, but it works much better and saver thanit it did !
never again we had any crash of AppleFileServer !!
now the internal disc are running for nearly 18 months, it's time too change and besides we need more space too !
the question is; should we replace them with 3x2TB internal discs
ore decrease up to a raid system again ?
using 10.5x I don't have the courage but perhaps it's solved under 10.6x?
thanks for any inputs
Message was edited by: mici
I've yet to have high cpu loads again on my server since upgrading to 10.5.8 so it seems to have resolved my issue since school started. When my network users cache was redirected to local drive before 10.5.8 upgrades things worked fine except of course for the MSWord saving issues. After the upgrade other programs started exhibiting the same saving issues MSWord had, even couldn't save textedit files to network homes without going to "save as". I ended up disabling the cache redirection to resolve the issue. Are you experiencing anything similar?
To take care of the saving issue after folder redirection, please read the info at the following website, this will take care of your issue:
We noticed that a lot of the users that were having this issue with AFP causing high CPU loads, were using RAIDS. We checked our settings in our RAID boxed and unchecked "Allow Host Cache Flushing" and our AFP server dropped from 338% to 4.5% immediately. Needless to say we are very happy and thought we should share.
Has anyone found a reason why setting afp on server to disconnect idle users after 60 mins results in 300 users being listed after much more than 60 mins? This has been commented on before but CPU usage shoots up and new users are rejected at login.
The setting just does not seem to work as it should!
I found that after doing what you did worked as-well. Unfortunately a week later our RAID Volume became corrupt and I had to destroy and recreate. Along with what you tried I adjusted the afp threshold and buffer settings for the clients and have had low cpu since.
Now I am experiencing something different. Small freezes- beach-ball for 15-30secs sometimes every 10mins or so. Any ideas people?
I'm still having this issue after trying all these suggestions. I'm not at 10.6, but that will be coming soon. The best luck I've had so far was to make sure all my Mac's were shut down and not communicating with the server in any way (150+/- clients). Then I rebooted the server a couple times to make sure it was 'happy'. This cleared the pegged CPU usage for well over a month, but it has come back. One thing I also noticed - if I'm in Activity Monitor watching AFP peg the CPU, when I hit the 'Sample Process' button the CPU usage drops to almost nothing while it samples. As soon as the sample is done, it jumps back up...