3.7.0 HP StorageWorks HP Scalable NAS File Serving Software release notes for Linux (AG513-96006, October 2009)

ManualsBrandsHP ManualsStorageHP DL380 - WSS Clustered Gateway

Operating system and environment issues

DescriptionDefect

Parent dentries are not revalidated

During path lookups, the operating system does not revalidate parent dentries when they are

specified as ".." components. The operating system assumes that the parent dentry associated

with ".." is always valid; however, this is not always the case in a distributed environment.

For example, server A may have a process that executes cd /a/b/c. Server B has a process

that renames /a/b/c to a/c and removes directory b. If the process on server A executes

cd .., it will be allowed to change directory into directory b, which no longer exists. Any

operations that attempt to modify the contents of directory b will fail with the appropriate

error. To get out of directory b, you will need to either execute another cd .. command or

change directory using an absolute pathname.

455

Problem with unmounting filesystems on the same mountpoint

Linux allows two filesystems to be mounted on the same mountpoint. The second filesystem

to be mounted hides the first filesystem. If you attempt to unmount the first filesystem, the

second filesystem will be unmounted because the first filesystem is not visible to the operating

system. This problem can occur with any type of filesystem.

To avoid the problem, do not mount two filesystems on the same mountpoint, or unmount the

second filesystem and then unmount the first filesystem.

701

Mount and unmount may fail if run in parallel

If you run several mount or umount commands in parallel, an operation may fail with the

following error message:

Apr 30 10:06:47 qar54s1 kernel: scsi(1:5:12): QUEUE FULL status

detected 0x1c-0x928, pid=128300970.Apr 30 10:06:47 qar54s1 kernel:

scsi(1:5:4): QUEUE FULL status detected 0x1c-0x928, pid=128300892.

Kernel memory exhaustion can result in the failure of applications or ejection of the server

from the cluster. If this condition is encountered, it is recommended that the command tag

queue depth be reduced from the default of 32 to 16 or less.

1041

QLogic HBA driver can disable HBAs

When the qla driver is unable to log into the FC switch for a long period of time (about 50

minutes), the driver can disable the HBA. When the switch is restored, the HBA remains

disabled. You will need to reboot the node to re-enable the HBA.

5024

QLogic driver can exhaust kernel memory

Under certain high-stress I/O conditions, the QLogic FibreChannel driver can exhaust kernel

memory.

This condition typically occurs while a server is running at very high I/O rates, and is possibly

triggered by storage subsystem error recovery. During this period the system administrator

may see many messages of the form: