More graceful handling of full or almost-full target disk

[Germar]

I have a 500gb disk with 1.9g disk space left
I have set auto-remove to, keep for 1 year, if free space is less than 2gb
smart remove, 2d,7d,4w,24m and don't remove named snapshots
I have snap shoots 1 of each for dec-2010,jan-june-2011 last week and 3 this week, when I run the backup I get insufficient space

os is ubuntu 11.04 64bit

---

Imported from Launchpad using lp2gh.

• date created: 2011-07-08T11:41:38Z
• owner: michael-dacova
• the launchpad url was https://bugs.launchpad.net/bugs/807476

[Germar]

(by michael-dacova)

[E] Error: rsync: writefd_unbuffered failed to write 4 bytes to socket [sender]: Broken pipe (32)
[E] Error: rsync: write failed on "/media/backup/backintime/mdclinux/mdacova/1/new_snapshot/backup/home/mdacova/.VirtualBox/HardDisks/WXP32-disk1.vdi": No space left on device (28)
[E] Failed to take snapshot 08/07/11 12:42:38 !!!

[Germar]

(by danleweb)
You can get the size used by all snapshots by getting the size ("nautilus" or just "du -sh " command) of the folder:
/backintime

I see that you backup virtualbox drives (*.vdi). This file can be real big and change every time you start the virtual machine.
You should check this files sizes.

The problem is that BIT remove old snapshots only after it takes a new one (because it doesn't know how disk space it will use).
If you take big snapshots you should change the rule to "if free space is less than 50gb" (or more).

Regards,
Dan

[Germar]

(by michael-dacova)
On 08/07/11 14:23, Dan wrote:

You can get the size used by all snapshots by getting the size ("nautilus" or just "du -sh" command) of the folder:
/backintime

I see that you backup virtualbox drives (*.vdi). This file can be real big and change every time you start the virtual machine.
You should check this files sizes.
I do a backup of my laptop every day before I start using Virtualbox,
and must say BIT has saved me on a number of times
The problem is that BIT remove old snapshots only after it takes a new one (because it doesn't know how disk space it will use).
If you take big snapshots you should change the rule to "if free space is less than 50gb" (or more).
I removed one of the snaps and all is working again.

Regards,
Dan

Keep up the good work, is there a user forum

[Germar]

(by danleweb)
No, at least not for now.

Even it works now it can stop working next time.
You should evaluate the size of a snapshots and configure "if free space is less than XXX gb" where XXX is the size of a full snapshot + 2-3 GB.

Regards,
Dan

[rsholmes]

Question: How does one recover from this situation? After hitting an insufficient space problem I now have a backup volume with 0 bytes free. Is there some way, either through Backintime or manually, to free up space in the backup volume without completely trashing the entire backup structure?

[buhtz]

Hello rsholmes,

can you explain "recover"? Was there any harm? Did you lose data?

To my understanding the only "harm" is that there is not backup possible anymore.

To solve this just navigate into the backup folder using the shell or your file manager and delete some of the snapshots. Please report back how it goes.

Best regards,

[rsholmes]

Not sure what happened but after a day of failed backups due to the disk being full, last night BIT started working again and is now backing up without errors. This was without my deleting any snapshots although it appears a couple of the oldest ones have been auto deleted.

[buhtz]

Will close the issue soon because it is to broad and covered by meta issue #1945 and its relative issues.

[buhtz]

Closing this ticket based on the comment above. Feel free to reopen
if the problem still exists. Thank you for your efforts. If you have
any further questions, ideas or encounter any other issues, please
don't hesitate to let us know.

Best regards,