VM Debian - disk size 100% - strange problem

[michadar5]

Hallo,

I have strange problem. Sometimes my virtual machine (production ) report 100% disk size , and have performance problem.

I have debian squeeze 6.0.4, Thick Provision Eager Zeroed disk.

df -h report 100% size, but  du -h --max-depth=1 / show i have free space

When i restart VM problem disappears.

What cause this? How can i create alarm to automatically restart VM when disk size is 100% ?

Thanks

Michadar5

[Jackobli]

Hello and welcome to the VMware Community.

This seems not really related to VMware and ESXi.

df lists the free space per filesystem. du lists the disk used space.

A reboot could clean some temp files and/or swapspace and that may have the effect you describe.

If your filesystem is full, you should consider resize the VM and afterwards resize the partition and filesystem inside the vm.

[michadar5]

Yes reboot could clean some temporary files but i don't see any temporary files (not so many files).

[Jackobli]

Please post your disk layout.

Like fdisk -l /dev/sda (assuming, sda is your disk).

[michadar5]

Dysk /dev/sda: 75.2 GB, bajtów: 75161927680

głowic: 255, sektorów/ścieżkę: 63, cylindrów: 9137

Jednostka = cylindrów, czyli 16065 * 512 = 8225280 bajtów

Rozmiar sektora (logiczny/fizyczny) w bajtach: 512 / 512

Rozmiar we/wy (minimalny/optymalny) w bajtach: 512 / 512

Identyfikator dysku: 0x0001355e

Urządzenie Rozruch   Początek      Koniec   Bloków   ID  System

/dev/sda1   *           1        7319    58787840   83  Linux

/dev/sda2            7319        8778    11718656   83  Linux

/dev/sda3            8778        9138     2890753    5  Rozszerzona (Extended)

/dev/sda5            8778        9138     2890752   83  Linux

[michadar5]

I close this topic. Problem was caused by routing problems in ISP network (It caused growing huge apache and mail queue temp files (something symilar to Ddos attack)

Thanks!!