Hi all
We had a situation with a bad FICON Director that resulted in a huge
performance impact due to the default way that z/OS handles error recovery on
FICON paths.
The symptoms were many (thousands) of IOS050I or IOS051I messages in the
task's joblog, followed by IOS001E, IOS2001I, IOS080I, and IOS450E messages.
We want to know how long will wait to vary the bad path offline
In our shop HIM DASD=00:15,IOTDASD=2,MSGONLY=YES
From IBM's manual,
Bad paths can encounter elongated recovery times
I/O timeout is 15 seconds + 2 retries are performed = 45 seconds
Q1. It seems to be 45 seconds. is it correct?
Q2 Is there any message which could tell us an I/O started or timeout or
interrupt or retry?
Q3 How to know whether it is 45 seconds correct or not in operlog or in
logrec?
Any suggestions are greatly appreciated! !!
Thanks a lot!
Janson
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
