WeeboTech Posted June 28, 2013 Share Posted June 28, 2013 take a peek at my google code page and check out a script called rsync_linked_backup. This would allow you to rsync critical directories to a dated version of those directories on another drive or another machine. The only issue is the rsync_linked_backup is a pull operation. It will look at the local repository and find the most recent directory. Then link all the files in that directory to a new directory. It will then go out and rsync the source directory over. So you will have at least 1 whole copy of the directory plus 1 version of any changes per date (or execution). If you name the destination directory by date, it's daily. by week, it's weekly. by hour, it's hourly if you are running it that often. You'll need to learn about shell and rsync. It's an easy way to handle the incremental backups. Frankly, after what went through with Hurricane Sandy, I would suggest you rsync your critical files to a removable drive that you bring somewhere else. Then rotate them. To give you an idea, here is the source home folder from one server. root@rgclws:/local/backups/npgvm3 # du -hs 2013* 6.9G 20130507 108M 20130508 67M 20130509 70M 20130510 70M 20130511 95M 20130512 66M 20130513 101M 20130514 75M 20130515 68M 20130516 69M 20130517 69M 20130518 63M 20130519 63M 20130520 77M 20130521 77M 20130522 123M 20130523 74M 20130524 72M 20130525 67M 20130526 67M 20130527 70M 20130528 70M 20130529 70M 20130530 70M 20130531 70M 20130601 67M 20130602 67M 20130603 70M 20130604 71M 20130605 70M 20130606 70M 20130607 71M 20130608 67M 20130609 67M 20130610 71M 20130611 76M 20130612 83M 20130613 4.0K 20130614 7.0G 20130615 67M 20130616 67M 20130617 70M 20130618 70M 20130619 70M 20130620 70M 20130621 70M 20130622 67M 20130623 67M 20130624 70M 20130625 70M 20130626 98M 20130627 70M 20130628 if I were to prune the 201305 files. You'll see the same 6.9G for the first folder, then the changes per day. root@rgclws:/local/backups/npgvm3 # du -hs 201306* 7.0G 20130601 67M 20130602 67M 20130603 70M 20130604 71M 20130605 70M 20130606 70M 20130607 71M 20130608 67M 20130609 67M 20130610 71M 20130611 76M 20130612 83M 20130613 4.0K 20130614 7.0G 20130615 67M 20130616 67M 20130617 70M 20130618 70M 20130619 70M 20130620 70M 20130621 70M 20130622 67M 20130623 67M 20130624 70M 20130625 70M 20130626 98M 20130627 70M 20130628 it does not keep multiple versions of the same file in the same directory. They are separated by the directory name you make. yet each directory can stand on it's own as a source for accessing a file. root@rgclws:/local/backups/npgvm3 # find 20130601 -type f | wc -l 118835 root@rgclws:/local/backups/npgvm3 # find 20130628 -type f | wc -l 119132 On one backup I named the directories according to the week. 2013-nn i.e. %Y year %U week number of year, with Sunday as first day of week (00..53) So when it was run daily, I would only be keeping 1 weeks worth of changes in the archives. Just some ideas. Quote Link to comment
garycase Posted June 29, 2013 Share Posted June 29, 2013 If I get the disk back I'm happy to lose those 5 files, not worth it, this is a good deal. Waiting on pass 3, showing all file entries within dir structure, is this a good sign? What do you mean "... get the disk back" ?? Earlier you said "... I tried to recover some accidentally deleted files." and that "... There was no HDD corruption, no red ball, there's nothing wrong with the drive prior to following steps in my first post to try recover with reiserfsck and get lost+found folder." So apparently the disk was fine to start with -- except you wanted to recover some files. If that's the case, you definitely should re-run the full rescan to see if you can indeed get those files. Doesn't cost anything to do so except a bit of time (and it's all "PC time" -- not "your time" Except, of course, for the few seconds it takes you to type ... reiserfsck --rebuild-tree --scan-whole-partition /dev/md2 Quote Link to comment
Juzzotec Posted June 29, 2013 Author Share Posted June 29, 2013 If I get the disk back I'm happy to lose those 5 files, not worth it, this is a good deal. Waiting on pass 3, showing all file entries within dir structure, is this a good sign? What do you mean "... get the disk back" ?? Earlier you said "... I tried to recover some accidentally deleted files." and that "... There was no HDD corruption, no red ball, there's nothing wrong with the drive prior to following steps in my first post to try recover with reiserfsck and get lost+found folder." So apparently the disk was fine to start with -- except you wanted to recover some files. If that's the case, you definitely should re-run the full rescan to see if you can indeed get those files. Doesn't cost anything to do so except a bit of time (and it's all "PC time" -- not "your time" Except, of course, for the few seconds it takes you to type ... reiserfsck --rebuild-tree --scan-whole-partition /dev/md2 Sorry, wrong terminology, it was really late... get the files back i meant to say. Hehe, I'm too scared to ever touch a file system again and will never do it. Success friends, everything has recovered and I'll tell u the full story of what happened. My brother in law is a young guy who came over to borrow some movies of me, I have some new ones stored in my server. While he came over, I was playing/versing my cousin in a video game, FIFA soccer to be exact. I was helping him copy over the files and he was I can do it, don't worry just play ur game, it became a tug o war, I just gave in and he advised he knows how to copy and paste. I left him to do it,to my knowledge I noticed taken 2 missing. Freaking out, I googles file recovery and found the steps contained in first post! The session was interrupted and the file system was urn formatted. Lucky enough with all your help, it's back on track. And guess what, the file wasn't deleted. He moved it into another movies folder, I worked it out when I ran xbmc 'update library' and found a second entry for movie taken 2. Moral of the story, don't let anyone have read/write previlidges to ur unraid box and never let ur win8 power options put computer to sleep in 30mins. Thanks for your help guys, hopefully this post helps people with what not to do... Quote Link to comment
WeeboTech Posted June 29, 2013 Share Posted June 29, 2013 Moral of the story, don't let anyone have read/write previlidges to ur unraid box and never let ur win8 power options put computer to sleep in 30mins. Use a UPS. (has nothing to do with windows sleep) When doing things at the command line. Use the screen program, then detach. This way if you get disconnected your session is still active Quote Link to comment
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.