You can make a difference in the Apple Support Community!

When you sign up with your Apple Account, you can provide valuable feedback to other community members by upvoting helpful replies and User Tips.

Looks like no one’s replied in a while. To start the conversation again, simply ask a new question.

AFP connections dropping on 10.8.5 Server

We are experiencing a very odd problem when AFP connections to our server are dropping part way through, which often then makes the client machine crash and force a hard reboot.

AFP logs on the server show nothing, but the /var/log/system.log events below show the disconnection as it happens.

Note I have installed the 10.8.5 Combo Updater, reinstalled the Server.app from the AppStore, experienced the same on various client machines, and seen the same problem.

We have plenty of disk space on both sides and the folder in question are about 40GB.

Disk Warrior on both ends has been used already, nothing interesting found.

Oddly enough connecting via SMB does not cause these problems.

Originally I thought it might be network related, but I tried connecting an Ethernet cable directly between client and server (bypassing the switch) and saw the same problem.

I can only surmise a recent OSX update (not sure if this problem is on the server end or client end) caused this problem.

Here are the log details:


Oct 7 18:28:14 frankie.example.com KernelEventAgent[46]: tid 00000000 received event(s) VQ_NOTRESP (1)

Oct 7 18:28:14 frankie.example.com KernelEventAgent[46]: tid 00000000 type 'afpfs', mounted on '/Volumes/INCOMING', from '//tony@ml.example.com/INCOMING', not responding

Oct 7 18:28:14 frankie kernel[0]: ASP_TCP Disconnect: triggering reconnect by bumping reconnTrigger from curr value 0 on so 0xffffff801e112ef0

Oct 7 18:28:14 frankie kernel[0]: AFP_VFS afpfs_DoReconnect started /Volumes/INCOMING prevTrigger 0 currTrigger 1

Oct 7 18:28:14 frankie kernel[0]: AFP_VFS afpfs_DoReconnect: doing reconnect on /Volumes/INCOMING

Oct 7 18:28:14 frankie kernel[0]: AFP_VFS afpfs_DoReconnect: posting to KEA EINPROGRESS for /Volumes/INCOMING

Oct 7 18:28:14 frankie kernel[0]: AFP_VFS afpfs_DoReconnect: Max reconnect time: 30 secs, Connect timeout: 15 secs for /Volumes/INCOMING

Oct 7 18:28:14 frankie kernel[0]: AFP_VFS afpfs_DoReconnect: connect to the server /Volumes/INCOMING

Oct 7 18:28:14 frankie kernel[0]: AFP_VFS afpfs_DoReconnect: Logging in with uam 10 /Volumes/INCOMING

Oct 7 18:28:14 frankie.example.com KernelEventAgent[46]: tid 00000000 found 1 filesystem(s) with problem(s)

Oct 7 18:28:14 frankie kernel[0]: AFP_VFS afpfs_DoReconnect: Restoring session /Volumes/INCOMING


Any ideas, suggestions appreciated!

OS X Server, 10.8.5

Posted on Oct 8, 2013 8:38 AM

Reply
22 replies

Oct 8, 2013 5:06 PM in response to Doug Niven1

We are having EXACTLY the same issues.

Upgraded to 10.8.5 and upgraded server.

Connection is being dropped opening a photoshop file (CC6) or copying a file to a client local drive.

Client requires a hard boot.

We had ports aggregated on the server and client and have dropped those to a single line.

The clients that have dropped are all desktop clients. We have not seen the problem with powerbooks which lead us to dropping the aggreagation.

Oct 8, 2013 8:38 PM in response to Linc Davis

Hey Linc,


Yes, it's a Mac Pro tower with two Ethernet ports, and I connected to its second interface directly with an Ethernet cable, bypassing the switch entirely, and experienced the same problem.


And yes, I used a different Ethernet cable than the existing one. So I'm pretty sure it's not the network.


I also had a fresh install of OSX Mountain Lion with a fresh install of Server.app on an external hard drive, then booted the Mac Pro tower from that. All the server shares are on drives within the Mac Pro.


Client then connected via AFP to this fresh new server installation and downloaded the same 40GB files/folderst that caused them to crash, without problem.


So it's definitely an OS/AFP problem with our existing server OS/AFP, but one that cannot be fixed with either the 10.8.5 Combo installer or a fresh install of Server.app.


SMB still works like a charm so we'll be using that till I can reinstall OSX Mountain Lion and Server.app fresh.

Oct 8, 2013 10:42 PM in response to Doug Niven1

Please take each of the following steps until the problem is resolved. Back up all data before making any changes.

Step 1

Delete all sharepoints in the File Sharing tab of the Sharing preference pane. Recreate them one at a time and test each one.

Step 2


Boot into Recovery mode, launch Disk Utility, select the startup volume, and run Repair Disk (not Repair Permissions.) If any problems are found, repeat. Then reboot as usual.

If any sharepoints are on a different volume, repair it too.

Step 3


Boot the server in safe mode and test. Note: If FileVault is enabled on some models, or if a firmware password is set, or if the boot volume is a software RAID, you can’t do this. After testing, reboot as usual and verify that you still have the problem.

Step 4

Triple-click anywhere in the line below on this page to select it:

/Library/Preferences/com.apple.ByteRangeLocking.plist


Right-click or control-click the line and select


Services Reveal

from the contextual menu.* A folder should open with an item selected. Move the selected item to the Trash. Repeat with this line:

/var/db/BRLM.db

Reboot.

*If you don't see the contextual menu item, copy the selected text to the Clipboard (command-C). In the Finder, select

Go Go to Folder...

from the menu bar, paste into the box that opens (command-V). You won't see what you pasted because a line break is included. Press return.

Step 5


Reinstall the OS. You don't need to erase the startup volume, and you won't need your backup unless something goes wrong. If your Mac was upgraded from an older version of OS X, you’ll need the Apple ID and password you used to upgrade.

If you installed the Java runtime distributed by Apple and still need it, you'll have to reinstall it.

Oct 15, 2013 8:54 PM in response to Chris_ilines

Similar problems here. We have three Mac Pros sharing drives through AFP between each other as well as 3 Macbook Pros connecting. We have been experiencing random server connection interrupted warnings and also extreme bottleneck in read speed which causes major hang all all connected computers. I began to suspect it was the 10.8.5 update on one of the Mac Pros.

Oct 16, 2013 11:18 PM in response to donaldk7

We've got a similar setup, with 3 Mac Pros talking to a Mac Pro server. Assumed it was the 10.8.5 update, so I rolled back the update on the server but left the update on the other 3 machines, and the issue still persisted. SMB was too slow for us, and also seemed to crash whenever AFP was accessed as well.

We seemed to solve the problem (at least for now) by switching our network to use jumbo frames. AFP seems to be working without errors now. I'll update if anything changes.

Oct 30, 2013 8:27 AM in response to Chris_ilines

I wonder if it's a driver issue with certain models of Mac Pro. As far as I know 10.8.5 Server works fine on Mac Mini i7 Late 2012 and Late 2009. I have one Mac Pro 2008 early, but that's on client, far too power hungry to leave on 24x7. Then have another Macmini i7 Late 2012 using 10,.9 Server, but that's a testing server, AFP so far runs fine.

AFP connections dropping on 10.8.5 Server

Welcome to Apple Support Community
A forum where Apple customers help each other with their products. Get started with your Apple Account.