Hi Guys,

Thanks for taking the time to look at my post!

I have a centos server (version 5.3) and I ran out of space on one of my partitions so I have been deleting files to free up space. When running df -h I get this:

Filesystem Size Used Avail Use% Mounted on
/dev/sda3 901G 859G 0 100% /
/dev/sda1 190M 17M 165M 9% /boot
tmpfs 505M 0 505M 0% /dev/shm

As you can see I have about 40gb free but it shows 100% usage. I thought it might be a locked file or something but I have tried lsof to look for anything that my be locked, I have also rebooted the server but still have had no luck!

If anyone could help me out that would be great! If you need anymore info please let me know.

Matt

Hi,

you probably have 5% of blocks reserved. You can check by running the following as root.

If this is the case, given the size of the disk you could probably reduce this to 1% or 2%. IIRC, you could do this by:
Please check the tune2fs man page for details.

Please note however that it is usually considered good practice to keep your disks below about 90% full.

Evo2.

Hi Evo2,

Thank you very much for your response, If you wouldn't mind taking a look at the result of the command: tune2fs -l /dev/sda3 and helping me go through it.

From the below results it does look like its 5%, i will do a bit of google'in but if you wouldn't mind, is there any risk with having it at 1-2%?
Does having it at 5% mean that I can only use 95% of the drive?


Filesystem OS type: Linux
Inode count: 243564544
Block count: 243563473
Reserved block count: 12178173
Free blocks: 13624880
Free inodes: 243433054
First block: 0
Block size: 4096
Fragment size: 4096
Reserved GDT blocks: 965
Blocks per group: 32768
Fragments per group: 32768
Inodes per group: 32768
Inode blocks per group: 1024
Filesystem created: Sun Apr 26 16:01:52 2009
Last mount time: Mon Aug 6 19:05:51 2012
Last write time: Mon Aug 6 19:05:51 2012
Mount count: 3
Maximum mount count: -1
Last checked: Mon Aug 6 18:16:26 2012
Check interval: 0 (<none>)
Reserved blocks uid: 0 (user root)
Reserved blocks gid: 0 (group root)
First inode: 11
Inode size: 128
Journal inode: 8
Default directory hash: tea
Directory Hash Seed: 659ce280-899b-4b47-a361-98971385ecac
Journal backup: inode blocks


Once again I really appreciate your help!

Matt

If the system filled up completely, everything would come crashing down in a hurry, very very bad things would happen. The reserved blocks are used so that root still has somewhere to work if the filesystem fills up. 5% is an antiquated number from way back in the day when hard drives were much smaller, in my opinion it should have never been a percentage, but a fixed value of maybe 100MB-1GB.

These days, 5% is so ridiculously overkill that it's not even funny anymore, I always resize it to around 1GB after setting up a filesystem. You can do that with:
Or you can keep it a percentage and just lower it to 1% using evo2's suggestion.

1 members found this post helpful.

Hi suicidaleggroll,

That's great thank you! My problem is solved

Matt

In any case, you should figure out why it 'filled up' and fix it, otherwise its only a matter of time before it fills the new space you have created.
I'd start by checking /var/log and the logrotate tool (/etc/logrotate.d dir) to see if its working properly (see /etc/cron.daily dir)

Hi Chris,

Thanks for the advice! This is just a backup server so just squeezing out as much space as I can until I get an answer on what I can delete

Matt

Absolutely agree (with Chris) - but 850 Gig of logs ?. I don't think so.
Seems like a good candidate for spinning off big consumers to separate partitions. That way the root can be left with some reserved allocation, and the data partitions can be set to zero to recover the space.
/home would be my guess as a good start.

in addition to the above, running df-h as root shows the real available space on your drives.