Friday, June 09, 2006

Too many cooks spoil the broth

I have three working directories one my laptop, other my office desktop and the university server where I have a home directory. Laptop and my office desktop have Fedora core 4 and the university server is a SunOS. But that is not really important here. Three systems and three home directories are all just confusing me; where I have the latest version of the file?; which is the final version of my paper?; did I get the backup of the correct files?. Many such questions arose. For many months I have not been coordinating systematically all these system usages. Hell...now I got into the problem which paper I want, which directory I want and where it is?, is it in my lappy, is it in my office desktop, or university server (ofcourse we can see the timestamp of the file and see which is the latest; but it is a tedious job). If we have only one home there will be no problem at all; get an external HDD and get a periodic backup. But having three homes in three separate systems will be of no use if we do not coordinate the usage; it will be utter confusion ultimately. Regarding my usage, most of the time I use my laptop for my official and personal works but once I come to office I use my office desktop because I do not want to carry my lappy everytime I come to office. After getting into lots of problems, I have decided to do this (not at all an efficient way as you will see why when you read through):

Have my laptop home as the primary home; my office desktop as my secondary home; and, my university home as my backup. Firstly I copy my laptop home to both my office desktop and my university home. Hence I have two backups. Secondly, whenever I need to edit/write some file do it in a new specific directory in my Laptop/office dektop (normally I won't work directly in university home) then write back the changes in the corresponding home directories. Well, this is really manual work. But at least I will not have much confusion as before. Just searching if we can do this in a more automated way? When I (re)write the files in one home we can get (back) the (new) version of the file in the other home; but what if, we delete a file in a home? Hmmmm.....good question...write a readme file of which files deleted? Man!! do we really need all these stuffs can I get a more efficient way of doing all these...

This is with systems now coming to external backups. I have a 160GB external hardrive where I would like to keep my lappy linux home and user directory of WinXP (which I use very minimally). I have to find a good way of getting backup there. I do not need the default way of just copying all the contents to the External HDD periodically I just need to copy the changes made. We have good backup commands in linux have to see through those.....

2 comments:

PS said...

Use rsync for your backups. Its a very handy tool.
And in case you have network connectivity all the time, then a neat solution is to run a NFS mount on one location, say your university server, and mount it remotely in the other two locations, so that you are working on one filesystem all the time. Of course, if you are more tech-savvy, you could use Coda!

Sakthi Balan said...

Thank you. rsych is a good one; I think I can try that.

NFS is a good solution - to have only one filesystem mounted on ur working system is cool. But I purposefully did not want NFS; for some reasons I wanted my office desktop to be an independent system and having NFS with my lappy and university server is not that good. If both of them are in same network then it should be good.