2TB time capsule not working with 10.6.1 - log included

[wibblyone]

Just bought a 2TB time capsule for my wife's imac, a 3 ghz intel core duo machine with 2 gb ram, recently upgraded to Snow Leopard. Initial TC setup seemed to go fine (including a TC firmware update to 2.4.7) but it's been impossible to complete a backup; the process keeps failing with a variety of error messages. I'm trying to use ethernet to do the backup because she has c.350 gigs of data on her hard drive. She isn't running any AV software (she has some installed but it doesn't work in the background), and doesn't have iDrive. No USB devices (other than a Belkin USB hub) are connected.

Below is a sample logfile cut and pasted from the console which shows the TC being mounted; a process then fails and ultimately an error message (codes 60 and 20) is generated. I've since tried erasing the TC and resetting it but it then hung at 'making disc available' with error code 1. Any suggestions gratefully received as my wife is becoming convinced I've bought her an expensive lemon. Actually, reading some of the other posts, it looks like 10.6.1 itself could be the problem but who knows; I'm a mac n00b who normally uses "another operating system" and am on unfamiliar turf here.

++++++++++++++++++++++++++++++++++++++++++++++++++++++++

PM 09:32:44 com.apple.backupd3328 Attempting to mount network destination using URL: afp://Mr.%20Big@Mr-Bigs-Time-Capsule.local/Data
PM 09:32:56 com.apple.backupd3328 Mounted network destination using URL: afp://Mr.%20Big@Mr-Bigs-Time-Capsule.local/Data
PM 09:32:56 com.apple.backupd3328 Creating disk image /Volumes/Data/Mr_Big.sparsebundle9/28/09
PM 09:32:56 kernel AFP_VFS afpfs_mount: /Volumes/Data, pid 33379
PM 09:33:56 com.apple.launchd[1] (com.apple.netauth.sysagent3337) Exited with exit code: 255
PM 09:34:00 GrowlHelperApp2845 *** __NSAutoreleaseNoPool(): Object 0x315940 of class NSCFNumber autoreleased with no pool in place - just leaking
[similar errors omitted here]
PM 09:35:10 kernel FireWire (OHCI) Lucent ID 5901 built-in: 179 bus resets in last 3 minutes.
PM 09:35:16 kernel ASP_TCP Disconnect: triggering reconnect by bumping reconnTrigger from curr value 0 on so 0x666333
PM 09:35:17 kernel AFP_VFS afpfs_DoReconnect started /Volumes/Data prevTrigger 0 currTrigger
PM 09:35:17 kernel AFP_VFS afpfs_DoReconnect started /Volumes/Data prevTrigger 0 currTrigger
PM 09:35:17 kernel AFP_VFS afpfs_DoReconnect started /Volumes/Data prevTrigger 0 currTrigger 1
PM 09:35:17 kernel AFP_VFS afpfs_DoReconnect started /Volumes/Data prevTrigger 0 currTrigger
PM 09:35:17 kernel AFP_VFS afpfs_DoReconnect started /Volumes/Data prevTrigger 0 currTrigger
PM 09:35:17 kernel AFP_VFS afpfs_DoReconnect started /Volumes/Data prevTrigger 0 currTrigger
PM 09:35:17 kernel AFP_VFS afpfs_DoReconnect started /Volumes/Data prevTrigger 0 currTrigger 1
PM 09:37:49 GrowlHelperApp2845 *** attempt to pop an unknown autorelease pool (0x809200)
PM 09:37:56 com.apple.backupd3328 Error 60 creating backup disk image
PM 09:37:56 com.apple.backupd3328 Failed to create disk image
PM 09:38:01 com.apple.backupd3328 Backup failed with error: 20
PM 09:38:38 com.apple.backupd3328 Failed to eject volume /Volumes/Data (FSVolumeRefNum: -138; status: -47; dissenting pid: 0)
PM 09:38:38 com.apple.backupd3328 Failed to eject volume /Volumes/Data (FSVolumeRefNum: -138; status: -47; dissenting pid: 0)

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++

 

[skorpien]

As I am a Snow Leopard user as well, I can vouch that Time Machine works on my Time Capsule (although I've been using my TC since Leopard and the backups just carried over). Have you tried renaming the Time Capsule's network and hard drive to something short without spaces or apostrophes? I remember when I initially purchased my Time Capsule that it was giving me issues until I did this.

 

[wibblyone]

Renaming Doesn't Work :-(

Thanks for the tip, but I changed the machine name to 'BigMac' and the TC name to 'BMT' (Big Mac Time, of course) without the quotes and that didn't work either. Here's the output from the Time Machine Buddy widget:

++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Starting standard backup
Attempting to mount network destination using URL: afp://BigMac@BMT.local/Data
Mounted network destination using URL: afp://BigMac@BMT.local/Data
Creating disk image /Volumes/Data/BigMac.sparsebundle
GetMasterBlock: Error 5 from read; amount=-1, wanted=512
mount_hfs: Input/output error
Error 1 creating backup disk image
Failed to create disk image
Backup failed with error: 20
[SnapshotUtilities mountPointForVolumeRef] FSGetVolumeInfo returned: -35
Failed to eject volume (null) (FSVolumeRefNum: -105; status: -35; dissenting pid: -1)

+++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Anything else I could try?

 

[skorpien]

Sorry, I'm coming up blank. The only thing I can think of is to make sure you've updated the firmware on your TC. Otherwise I'd call AppleCare as you're probably within your AppleCare window. Or if you'd rather, take it in to an Apple Store. If they can't get it working, it may just be a faulty unit.

 

[wibblyone]

Working via LAN but sloooooow

Thanks. The oddest thing is that I have discovered that it *does* work, only *not* if the Mac is directly connected to the TC via ethernet. It will *only* work if I connect the Mac and the TC separately to a switch on my LAN so that it backs up over the home network. As I write it's done about 84 of the 350 gigs, but it's taken nearly 24 hours so far. Still, better than nothing. We'll see if it completes and I'll post back.

 

[skorpien]

Wait, so you have a network already that you're connecting your TC to? If so, have you set up your TC in Bridge Mode? I'm just trying to glean information about your network setup.

As for the backing up slowly, that's just the way Time Capsule works. Especially for a 350GB initial backup, you're looking at a loooooong time to complete. But once it's completed, incremental backups usually go by considerably faster.

 

[wibblyone]

Yes I already have a home lan. The iMac is connected to the net and the lan over cat5 ethernet cable via the following slightly circuitous route:

ISP <> Motorola cable modem <> Linksys WT54G router <> Belkin 10/100 ethernet switch <> Shiro 200 Mbps power line plugs <> Linksys SD208 switch <> iMac.

We don't use the iMac's wi-fi as it's only a G router and we find the wi-fi is far slower than the ethernet for both uploads and downloads under normal circumstances.

Anyway, the TC and the iMac are both now plugged directly into the Linksys SD208 switch. But writing to the TC is still far slower than writing to other network drives, although I suspect the power line networking section of the route has something to do with it; the bits of my network that avoid that link do generally show faster speeds.

That said, the TC should still work when directly connected to the iMac. Even though it's working after a fashion, it's not fully working, and something is clearly not right.

 

[skorpien]

Unfortunately I can't give any further advice. You're absolutely correct that it should be working, especially by connecting it via ethernet. Maybe AppleCare would be able to provide better suggestions? If it's still backing up now I'd wait for it to finish and try connecting the iMac directly to the TC and see if it works for future incremental backups. Good luck.

PS - I've found from reading online that the TC's read/write speeds to the hard drive are quite a bit slower than other NAS drives. That may be why you're experiencing the slow speeds, but that doesn't account for it not working while directly connected.