Slow Time Machine backup to Time Capsule

What do you mean by "shutdown"?
There's no power switch, so shutdown, to me, means removing power.
I don't see anywhere that suggests anything more than:

Pull the power cord.

You could save wear and tear on the power connector by using a power strip with a switch.
You would want to make sure that a backup is not in progress, of course.
And, if the router is active, you would lose any active network connections, but if you are simply doing a power reset, then the network would come right back after a few seconds.
 
Sure, that will work... The access to a power strip is likely not necessary, if you are just moving the TC to a temporary location.
As I said, there's no real shutdown needed, as long as you know that no backups are in progress. Just unplug, and move. :D
 
DeltaMac said:
Sure, that will work... The access to a power strip is likely not necessary, if you are just moving the TC to a temporary location.
As I said, there's no real shutdown needed, as long as you know that no backups are in progress. Just unplug, and move. :D
Click to expand...

I might add just to be safe, 'unmount the drive' then unplug it and move.
 
OK, back to the struggle. I applied the combo update and let TM start up again, without relocating the TC. It took 3.5 hours to backup 844mb. Before it was struggling to backup 150mb in 2 hours, so I guess it's progress. It still is doing the backup in one big chunk and a second little chunk. Note the line at 11:38 where it has completed copying 86763 files but only 2 mb, and then sits there for nearly an hour before starting the thinning step. After 40 minutes of thinning, it's done.

And during this time, MDS crashed three times.

9:48 Starting standard backup
9:48 Attempting to mount network destination using URL: afp://XXX@YYYYs-Time-Capsule.local/Data
9:48 Mounted network destination using URL: afp://XXX@YYYYs-Time-Capsule.local/Data
9:48 QUICKCHECK ONLY; FILESYSTEM CLEAN
9:48 Disk image /Volumes/Data/imac.sparsebundle mounted at: /Volumes/Time Machine Backups
9:48 Backing up to: /Volumes/Time Machine Backups/Backups.backupdb
10:01 No pre-backup thinning needed: 6.36 GB requested (including padding), 65.22 GB available
10:48 Copied 123668 files (842.2 MB) from volume Macintosh HD.
11:04 No pre-backup thinning needed: 1.00 GB requested (including padding), 65.22 GB available
11:38 Copied 86763 files (2.0 MB) from volume Macintosh HD.
12:33 Starting post-backup thinning
1:13 Deleted backup /Volumes/Time Machine Backups/Backups.backupdb/imac/2013-09-16-045419: 65.22 GB now available
1:13 Post-back up thinning complete: 1 expired backups removed
1:13 Backup completed successfully.
1:13 Ejected Time Machine disk image.
1:13 Ejected Time Machine network volume.
 
I let it go overnight, but the incrementals are now taking longer than ever. So I waited impatiently for one to finish and turned off TM. I have started the clean install process.

No, I have never used DU on anything on the Time Capsule. I tried this morning, from the 10.6.8 machine via wifi, but killed the verify after it told me the estimated time was 11 hours. I then did the same from 10.8.4 via ethernet, but killed it again. I don't have the patience to wait 4 hours.

I did the same thing for my laptop's TC bundle. It finished in 4 minutes.
 
Run the Terminal command (I hope this one your older OS X):
Code: 
tmutil destinationinfo
and post the info of the readout (if your are able to get the command to work and are comfortable posting it).

 
I wasted Thu and Fri trying mightily to find a magic solution, but I only made things worse. Sat AM, I found a backup in progress and waited (im)patiently. It finally finished in a total time of just under four hours. So, I performed one more SuperDuper backup and ...

With the Time Capsule still wired directly, I booted from the install DVD, erased the internal disk, and began a clean install. I asked it to migrate from a TM backup and it started transferring. After about 90 seconds it said it was finished and presented the login screen with my usual list, but it wouldn't accept my password.

Swearing loudly, I started all over again at 9am. At 10am it started migrating but this time kept going. At 11am it had started transferring with an estimate of 3.5 hours to go. At 4pm (so much for its estimate) it finished, rebooted, and graciously allowed me to log in. Things looked "normal", so I reconfigured TM to use a freshly erased firewire drive, started it up, and called it a day.

Sun AM. The console showed mdworker32 crashing about 20 times per second and the backup taking forever. The crash logs now (perhaps always had) showed which files it was working on. All were font files in a folder that was a copy of the last System 9 disk from just before moving to OSX, most of them dated in 1998! Rather than doing something rational with them, mdworker32 just crumped over and over.

I chucked the folder, rebooted into safe mode, re-erased the TM drive, and started a backup. It finished 162gb in 2.5 hours. But, the console was full of permission errors trying to write to the TM drive. Then I recalled that a TM drive must have UNCHECKED the "ignore ownership" option. However, Get Info doesn't present that option at all for my (any?) firewire drive. So ...

I rebooted to safe again, switched TM to a USB drive, with that option unchecked, and started a backup. Done in two hours and ... NO CONSOLE PROBLEMS!

I rebooted normally and started a backup. Done in 3 minutes!

I performed another SuperDuper backup, for the sake of sanity, and declared victory.

Mon AM. Backups are now taking about 10 minutes, still no problems.
 
You must log in or register to reply here.
Back
Top