Koozali.org: home of the SME Server

Obsolete Releases => SME 9.x Contribs => Topic started by: psoren on June 06, 2021, 08:59:42 AM

Title: Delete parts of affa archive
Post by: psoren on June 06, 2021, 08:59:42 AM
Hi

I have an affa archive where i need to delete files that i don't need and just take a lot of space that i do need.

Is there a way to delete them through the entire affa archive with just one command instead of going through each and very backup set?

Thanks
Per
Title: Re: Delete parts of affa archive
Post by: mmccarn on June 07, 2021, 12:46:17 PM
(I haven't used affa for quite a while...)

I don't see any option listed in the affa documentation (http://affa.sourceforge.net/AffaPdfMan.pdf) that would remove a folder from all archives. Possibly reconfiguring the backup job to ignore the unwanted folder and running affa --cleanup - but this is for removing entire old archives after making changes to the retention settings, not portions of existing archives.

@google remove all hard links to a file in linux (https://www.google.com/search?q=remove+all+hard+links+to+a+file+in+linux) indicates that your best bet is 'find'. The examples I found work by counting hardlinks or creating an index of inodes, but if the unwanted folder has a unique name (somthing that doesn't occur elsewhere) that might be easier.

Title: Re: Delete parts of affa archive
Post by: psoren on June 07, 2021, 01:48:43 PM
Quote from: mmccarn on June 07, 2021, 12:46:17 PM
(I haven't used affa for quite a while...)

I don't see any option listed in the affa documentation (http://affa.sourceforge.net/AffaPdfMan.pdf) that would remove a folder from all archives. Possibly reconfiguring the backup job to ignore the unwanted folder and running affa --cleanup - but this is for removing entire old archives after making changes to the retention settings, not portions of existing archives.

@google remove all hard links to a file in linux (https://www.google.com/search?q=remove+all+hard+links+to+a+file+in+linux) indicates that your best bet is 'find'. The examples I found work by counting hardlinks or creating an index of inodes, but if the unwanted folder has a unique name (somthing that doesn't occur elsewhere) that might be easier.

Yes i have looked at documentation as well and got to the same as you. I would rather not loose the archive totally and redo it.

Probably the best way is to go by name as you suggest. its folders with thousands of small files.

Thanks
Per