Hard disk corruption or failure

FortiWeb appliances usually have multiple disks. FortiWeb stores its firmware (operating system) and configuration files in a flash disk, but most models of FortiWeb also have an internal hard disk or RAID that is used to store non-configuration/firmware data such as logs, reports, and website backups for anti-defacement. During startup, after FortiWeb loads its boot loader, FortiWeb will attempt to mount its data disk. If this fails due to errors, you will have the opportunity to attempt to recover the disk.

To determine if one of FortiWeb’s internal disks may either:

• Have become corrupted

• Have experienced mechanical failure

view the event log. If the data disk failed to mount, you should see this log message:

date=2012-09-27 time=07:49:07 log_id=00020006 msg_id=000000000002 type=event subtype="system" pri=alert device_id=FV-1KC3R11700136 timezone="(GMT-5:00)Eastern Time(US & Canada)" msg="log disk is not mounted"

Connect to FortiWeb’s CLI via local console, then supply power. After the boot loader starts, you should see this prompt:

Press [enter] key for disk integrity verification.

Pressing the Enter key will cause FortiWeb to check the hard disk’s file system to attempt to resolve any problems discovered with that disk’s file system, and to determine if the disk can be mounted (mounted disks should appear in the internal list of mounted file systems, /etc/mtab). During the check, FortiWeb will describe any problems that it finds, and the results of disk recovery attempts, such as:

ext2fs_check_if_mount: Can’t detect if filesystem is mounted due to missing mtab file while determining where /dev/sda1 is mounted.

/dev/sda1: recovering journal

/dev/sda1: clean, 56/61054976 files, 3885759/244190638 blocks

If the problem occurs while FortiWeb is still running (or after an initial reboot and attempt to repair the file system), in the CLI, enter:

diagnose hardware harddisk list

to display the number and names of mounted file systems.

For example, on a FortiWeb 1000C with a single properly functioning internal hard disk plus its internal flash disk, this command should show two file systems:

name size(M)

sda 1000204.89

sdb 1971.32

where sda, the larger file system, is from the hard disk used to store non-configuration/firmware data.

If that command does not list the data disk’s file system, FortiWeb did not successfully mount it. Try to reboot and run the file system check.

If the data disk’s file system is listed and appears to be the correct size, FortiWeb could mount it. However, there still could be other problems preventing the file system from functioning, such as being mounted in read-only mode, which would prevent new logs and other data from being recorded. To determine this, enter:

diagnose hardware logdisk info

to display the count, capacity, RAID status/level, partition numbers, and read-write/read-only mount status.

For example, on a FortiWeb-1000C with a single properly functioning data disk, this command should show:

disk number: 1

disk[0] size: 976.76GB

raid level: raid1

partition number: 1

mount status: read-write

To prevent file system corruption in the future, and to prevent possible physical damage, always make sure to shut down FortiWeb’s operating system before disconnecting the power.

You can also display the status of each individual disk in the RAID array:

FortiWeb # diag hardware raid list

disk-number size(M) level

0(OK),1(OK), 1877274 raid1

If the file system could not be fixed by the file system check, it may be physically damaged or components may have worn out prematurely. Most commonly, this is caused by either:

Failing to shut down FortiWeb’s operating system before disconnecting the power (e.g. someone pulled the power plug while FortiWeb was running)

Logging misconfiguration (e.g. logging very frequent logs like traffic logs or debug logs for an extended period of time to the local hard drive)

For hardware replacement, contact Fortinet Customer Service & Support: