Find all posts by h foorsa. AIX MPIO stops using any disabled or Defined paths, and therefore, no error detection or recovery will be done as a result of the scheduled maintenance. If scheduled maintenance is planned for a SAN or for a storage device, it is best to identify the disk paths that will be impacted by that maintenance and use the rmpath command to manually disable those paths before starting the maintenance. The new AIX technology levels added the -e flag for errors to the lsmpio command, to be used with the -a and -r flags. Connecting to one or more ports on each of the controllers protects against the failure of an entire storage device controller. This can lead to much quicker recovery from a command time out and also much quicker detection of situations where all paths to a device have failed.
|Date Added:||22 December 2017|
|File Size:||47.86 Mb|
|Operating Systems:||Windows NT/2000/XP/2003/2003/7/8/10 MacOS 10/X|
|Price:||Free* [*Free Regsitration Required]|
View Public Profile for zxmaus. A path that is failed due to timeout policy can later be recovered by the AIX health check commands.
The time now is Gary Domrow Published on February 19, This flag cannot be used with the -P or -T mutlipath. For use on versions prior to AIX 6. Similar to the first case, this indicates an issue with one particular FC port, but this time the problematic port is on the storage device. This attribute indicates the action that the PCM should take when a command timeout occurs.
Another feature mentioned in the IBM article was the new —U flag for the chdev command.
In order to upgrade to the latest version of the ODM you will need to first remove the old one. These attributes have dependencies on settings and features that are provided by the storage subsystem. Another feature of the recent improvements to the AIX MPIO software is the expansion of the output of the lsmpio command to include error counts per adapter and per remote port, allowing the SAN administrator to better deduce the source of intermittent, recurring errors.
IBM AIX MPIO: Best practices and considerations
A path-control module PCM provides the path management functions. In an MPIO configuration, more is not necessarily better. This can lead to much quicker recovery from a command time out and also much quicker detection of mulyipath where all paths to a device have failed. The AIX default PCMs consist of two path control modules, one to manage disk devices and another to mange tape devices.
That includes paths with a state of failed. Attributes for other SCSI tapes type ost. If outstanding commands sent on healthy paths exist on that device, those commands are canceled by the LUN reset operation. Those path states make it obvious that there is a failure associated with the link for that particular port, and therefore, it is easy for the user to determine the source of the problem.
Multiple Path I/O
The health checker never probes paths that are in a Disabled or Missing state. If the error count is zero, then the problem is likely resolved. Multkpath instance, the —S flag provides some interesting statistics and counters for hdisk devices. In fact, an excessive number of paths in an MPIO configuration can actually contribute to system and application performance degradation in the event of SAN, storage, or Fibre Channel fabric issues or aox.
IBM AIX multipath I/O (MPIO) resiliency and problem determination
For example, if the fiber connected to a particular AIX FC port is pulled out, all currently used multipatth originating at that port move to the Failed state. Please repeat this for all devices. Reboot, as prompted by the tool. Thus the issue must be within the SAN fabric. In this mode, the PCM only sends path health checks down paths that are marked as failed. When disabling multiple paths for multiple LUNs, rmpath is simpler than chpathas it does not have to be run on a per-disk basis.
There are other possible configurations beyond what is described here that can be considered.
System reliability and availability are increased by a careful consideration of the user-modifiable options in each system configuration. Sometimes, it might be difficult for the SAN administrator to identify the source of intermittent, recurring errors. Other PCMs might not implement this grace period. Carefully consider the impacts of extra, unnecessary redundancy before using more paths.
What you set it to will need to be unique, you can set this value to “0x1” for example. Note that if the error count for the remote ports is zero or very small, it may indicate that the port is working fine or it may indicate that the port is not currently being used.