![]() | Sun System Handbook - ISO 4.1 October 2012 Internal/Partner Edition | ||
|
|
![]() |
||||||||||||
Solution Type Problem Resolution Sure Solution 1021032.1 : VSM Channel Detected Errors when using "Short Reads" and SLI /LBI
PreviouslyPublishedAs 269508 This article addresses VSM Channel Detected Errors when using "Short Reads" and SLI /LBI
Applies to:Sun StorageTek VSM System - Version: 4 to 5C - Release: 4.0 to 5.0All Platforms SymptomsApplication might receive an IOS050I CHANNEL DETECTED ERROR when reading data by the use of “Short Reads” along with the Suppress Length Indicator (SLI) when reading a large data block (> 64k). CauseThe source of the channel detected error is a combination of two conditions: The error is in the way the VSM handles reconnecting to the host after staging the rest of the data into cache. In response to a read operation, the VSM always starts transferring data to the host as soon as there is data available in cache. If the data in cache becomes been depleted, the VSM initiates a command retry (CCR) sequence. After the remaining data has been staged into cache, the channel processor has two methods to resume the data transfer. If the channel processor detected simple data transfer (data blocks <64k and not command chained), the CCR process is terminated at the VSM and previously un-transferred data is released to the channel/host. If Large Block Interface (LBI) is detected (> 64k data blocks plus data chaining), the queued CCR is released to the host. This causes the host to discard all previously transferred data and re-drive the entire read request. SolutionThe fix is to have the VSM to process/resume a LBI data transfer correctly. The VSM will now release the CCR to the channel instead of discarding it. Code fix will be included in all future code levels starting with D02.06.00.00. Code level D02.05.05.00 has been released by VSM engineering for sites needing immediate relief. Attachments This solution has no attachment |
||||||||||||
|