My Email service (Zimbra) has stopped and its crying low disk space via the warning: WARNING: Disk space below threshold for /opt/zimbra/store.
.
On investigation, I've found the weirdest thing I've seen in Ubuntu in forever. When I login via SSH, I get the usual summary of the server including the storage. It states:
=> / is using 90.9% of 12.79GB
=> /mail is using 94.9% of 88.58GB
When I do a df -h
the output shows that on /mail I only have 4kilobytes left and that usage is 100%?
Filesystem Size Used Avail Use% Mounted on
/dev/sda1 13G 12G 532M 96% /
udev 1.7G 4.0K 1.7G 1% /dev
tmpfs 343M 256K 343M 1% /run
none 5.0M 0 5.0M 0% /run/lock
none 1.7G 0 1.7G 0% /run/shm
dev/sdb5 89G 85G 4.0K 100% /mail <============This is what is crazy
I have Nagios on a remote server and it reads that I have 94.9% usage just like in the login summary. So is df - h
giving the wrong results which then makes zimbra die?
@Jakke:
df -ih
output as below:
Filesystem Inodes IUsed IFree IUse% Mounted on
/dev/sda1 832K 235K 598K 29% /
udev 427K 443 426K 1% /dev
tmpfs 429K 323 429K 1% /run
none 429K 6 429K 1% /run/lock
none 429K 1 429K 1% /run/shm
/dev/sdb5 5.7M 378K 5.3M 7% /mail
Additional Info:
A while ago the /mail partition was full and I added 10GB space into the /mail partition using gparted and space moved from 80GB to 90GB. Could that be affecting it?
Using Ubuntu 12.04.2 LTS
df -ih
? This will display inodes as well. Maybe you ran out...df -ih
. what does that output mean?