Apache Server
 

Hi Friends!
I was asked this question in an interview,
What would be the answer of this question?

question was-
There are two partitions A and B both of 50 GB Apache Server is running on Partition A which is now getting full i.e. 48 GB only 2GB remaining.
I have B partition which is of 50GB completely free.
LVM concept is not used here.What should I do in this case?

Pretty simple. Assuming A and B are contiguous partitions:

1. Make sure you have a backup
2. Boot from CD/DVD with gparted
3. Delete B
4. Grow A into the now free space

parted will take care of resizing the ext2 or ext3 filesystem. If you have some weird filesystem, you may have to resize and check it with its own tools after rebooting.

Well there is no definite "right" answer. smallponds answer is fine, and in the longest terms probably the best without LVM - although there is no details about the partitions being on the same device, let alone contiguous. Alternatively though you could easily look to split directories across the two partitions, mounting one under the other, and there should be no blockers preventing that being possible at all, other than the directory layout and usage.

NO, We are not supposed to reboot the system it is production server!
i want answer of this interview question!

Barking orders at us will NOT get you an answer. You want an answer? Then figure it out yourself....if you can't you're not qualified for the job.

Sorry to sound harsh, but you were given two good answers, then you complain about them?

Wow, you really do appear to be a total tool. You certainly don't deserve to pass any intervirews with an awful attitude like that.

For posterity's sake, I'd mention that the question on its own doesn't lend itself to a proper answer. More investigation would be needed to determine the root cause of the problem. (i.e. Is a log file growing out of control? Is something creating large core dump files? Is a shell user doing something that's chewing up space? Is this just a normal condition of a long-used system?)

Trying to tack on bandaids without first understanding the problem is only a temporary "fix".

Mate, there are some of us here that can understand where this type of question leads. These show the journey that the candidate takes and shows the desire to understand the problem and then offer solutions. Yes there is more than one correct answer, depending on the roll and type of company you interviewing for, and this then causes you to have to think about it.

Rebooting a production server is a problem... Then I would suggest that, if they are offering a critical service, their DR plan's are woefully inadequate.

What do you want, quick hack for an internal web server or full blue chip answer with bells and a risk study attached.

Have fun