Time Machine stuck in preparing backup...

vtgill wrote:

. . .

Error writing to backup log. NSFileHandleOperationException:*** -[NSConcreteFileHandle writeData:]: No space left on device

The TC's internal HD is chock-full. There's not even enough room for it to start the (hidden) log it keeps for each backup.

Do you have other data on the TC's internal HD, besides your backups? If so, copy something elsewhere temporarily and delete it (a few dozen MBs should be enough).

If not, try manually "compacting" the sparse bundle, per the pink box in Time Machine - Frequently Asked Question #12. If that doesn't help, delete your oldest backup per the earlier part of that same link, then compact it again.

EDIT: If you do have other data on the TC, see #Q3 in Using Time Machine with a Time Capsule for some possible ways to avoid this happening again.

Message was edited by: Pondini

vtgill wrote:

Thanks, I was able to free up 150 MB or so by deleting some files I'd been storing on the Time Capsule when using it as a network drive. The backup still doesn't seem to be starting, even with this change.

It will start, figure out there isn't enough room, compact the sparse bundle, then restart. Then it will begin deleting your oldest backup(s) to make room as/if necessary. Seems rather slow, but it should do what it needs to. Connect via Ethernet cable if at all possible -- it will still be slow, but 2-3 times faster than via WIFI.

Is there a way I can manually delete some off my older backups (back to 2010) to free up more space?

Yes, but it's tedious and time-consuming. See Time Machine - Frequently Asked Question #12.

vtgill wrote:

Thanks, it seems like the log is capturing a different set of activity. Maybe this is it compacting the sparse bundle as you suggested.

No, not yet -- it hasn't gotten far enough.

Only one comment in the log looks problematic "Event store UUIDs don't match for volume: Macintosh HD."

That's normal, under these circumstances. It just means that TM isn't sure that what's on your last backup matched your system (at the time); so it will have to do a "deep scan," comparing everything on your Mac to the backups, to figure that out. But it can't even do that, yet, because the backups need to be indexed first.

Waiting for index to be ready (100)

Waiting for index to be ready (100)

You'll see one of those every minute until the indexing is done; then a message about the "deep scan," then (a while later) one saying the scan is done, then (a while later) how much space is needed and available. Then if there isn't enough room, it will unmount the sparse bundle and compact it, etc.

If the indexing continues for a very long time, the index itself may be damaged. If so, post back and we'll work on that.

Are you connected via Ethernet all the way? From your Mac to the Airport to the Time Capsule?

Message was edited by: Pondini

vtgill wrote:

Backup has almost completed, status is currently "Cleaning up old backups...",

Maybe not -- it may be doing that to make room; if it doesn't, it will stop and restart itself.

Just keep an eye on it, via the widget; it may go through all sorts of gyrations (incuding another deep scan), but should finish eventually.

so things are looking up, and things are definitely unstuck. Looks like the problem was the fullness of my TC and the fact that part of the reason it was so full was data that had not been placed there by TM.

Yes, exactly. Normally, TM will leave enough room for the next backup to get started -- it creates an "in.Progress" package, a hidden backup log and a hidden exclusion list, before it figures out how much space it needs, and whether it has to delete old backups. So it's a bit of an excruciating process on a network backup (but much improved over Leopard and Snow Leopard).

The trick, obviously, is not to put so much of the other stuff there that the next backup can't get started. Once this is all sorted out, take a look at #Q3 in Using Time Machine with a Time Capsule.

I ran into the same issue after upgrading my macbook pro from Leopard to Lion. It was stuck in the message "preparing the backup..." for hrs. If you dont care about the backups and have all the data on machine, simply removing the whole backup from backup disk was faster for me. I started the backup afresh.I feel its a use case thats missed out by Apple when moving across OSs, Lion wasnt able to honor the last backup info to resume.

GREAT PONDINI - please help!

I'm having a similar issue. I'm currently tracking my progress in the widget, and I have the following so far:

Starting standard backup

Attempting to mount network destination URL: afp://Ross%20@Ross-Time-Capsule.local/Data

Mounted network destination at mountpoint: /Volumes/Data-1 using URL: afp://Ross@Ross-Time-Capsule.local/Data

QUICKCHECK ONLY; FILESYSTEM CLEAN

Disk image /Volumes/Data-1/Ross’s MacBook Pro.sparsebundle mounted at: /Volumes/Time Machine Backups

Backing up to: /Volumes/Time Machine Backups/Backups.backupdb

Event store UUIDs don't match for volume: Macintosh HD

Waiting for index to be ready (100)

Waiting for index to be ready (100)

Waiting for index to be ready (100)

Waiting for index to be ready (100)

Waiting for index to be ready (100)

Waiting for index to be ready (100)

Waiting for index to be ready (100)

Waiting for index to be ready (100)

Waiting for index to be ready (100)

Waiting for index to be ready (100)

Waiting for index to be ready (100)

Waiting for index to be ready (100)

Waiting for index to be ready (100)

Waiting for index to be ready (100)

Waiting for index to be ready (100)

Waiting for index to be ready (101)

Waiting for index to be ready (100)

Waiting for index to be ready (100)

Waiting for index to be ready (100)

Waiting for index to be ready (100)

See how there is a 101 mixed in there? Is that something to be concerned about? I've been having issues with it being "stuck on preparing backup" for weeks now. I've tried almost everything on your website (except for deleting my entire Sparse bundle). Now, I'm following along with the advice you provided above. Should I just let this process keep on going? How long should I let it "wait for index" before I realize there is a problem?

Thanks!

Update: now the following occured:

Deep event scan at path:/ reason:must scan subdirs|new event db|

Finished scan

In the menubar, it is still "preparing backup..."

On the widget, nothing is happening. I guess I should just still wait?

Ross Sheingold wrote:

. . .

Event store UUIDs don't match for volume: Macintosh HD

That's a notice that Time Machine isn't sure the previous backup completed normally, or there have been too many changes, so it can't use the quick method (a hidden log of changes that OSX keeps on each drive) to figure out what's changed and needs to be backed-up. This may or may not indicate a problem; see #A6 in Time Machine - Troubleshooting for details on some of the common messages.

Waiting for index to be ready (100)

Waiting for index to be ready (100)

. . .

See how there is a 101 mixed in there? Is that something to be concerned about?

No; it's more the fact that there are a number of them, which means the backups had to be indexed. If the previous backup failed or was cancelled, that's not too unusual.

If you see many of them on every backup, there may be a problem with the index.

Deep event scan at path:/ reason:must scan subdirs|new event db|

Finished scan

That's the process TM had to use instead of the quick method above. It compared every file on your system to the backups. Now it's preparing it's list of items to be backed-up, and figureing out how much space that will take. Then it will see if there's enough room on your Time Capsule for the backup, and either make space or start backing-up.

What you should see next is a "xxx GB required (including padding); yyy GB available" message. The xxx is about 20% larger than the estimated size of the backup. If xxx is larger than yyy, of course, there isn't enough room, so it will have to dismount and "compact" the sparse bundle to recover space, then mount it again to see if that made enough room. If not it will begin deleting old backups.

If you're not connected via Ethernet cable, do so -- it's still a slow process, but will be 2-3 times faster than via WIFIl.

It's hard to tell what's going on, much less what to do about it, until we see that next message. If it's been a while since the last competed backup, xxx is likely to be fairly sizeable.

Yeah, it's been a while, so the number will likely be VERY large. I've done the manual compression via terminal, but it only saved around 250mb out of a 94gb backup.

Anyway, it's been over an hour and the latest message is still

Deep event scan at path:/ reason:must scan subdirs|new event db|

Finished scan

I will update when something new occurs. Thanks so much for all of your help! Is there somewhere I can donate to help keep your website running - you're such a helpful resource for Time Machine and Time Capsule users!

Ross Sheingold wrote:

. . .

Anyway, it's been over an hour and the latest message is still

Deep event scan at path:/ reason:must scan subdirs|new event db|

Finished scan

That's not a good sign. 😟

Have you tried the things in the green box of #D2 in Time Machine - Troubleshooting? Some of them can be done while a backup is running.

And/or, look back with the TM Buddy widget to a previous backup attempt. You'll probably see most of the same stuff, but if there's anything of interest after the "xxx GB required ..." message, copy and post it here. A clue may be lurking there.