IIRC, some of those STK drives were faster than ESCON could go. Have you looked at the path?
-----Original Message----- From: IBM Mainframe Discussion List [mailto:[email protected]] On Behalf Of John Kelly Sent: Wednesday, February 03, 2010 12:51 PM To: [email protected] Subject: Tape drive disconnect time issue I have an old STK tape system ( ten 9840's in 3490 emulation mode inside a 9740. Each drive ESCON point to point) , that will hopefully go away but .... I have an issue with HSM dump running longer than expected. The problem appears disconnect time. During the HSM dumps, we're using four drives for two DumpClasses and most other stuff is stopped (some FTPs will come in). No other tasks or LPAR is using any tape drive. Cpu is in the low 20% range, DASD service time is around 4 ms and DASD rate is around 50 and tape rate is around 600. One set of tape drives (1 tape for each of the two DumpClasses) has an RMF average disconnect time of .84 and the other set of drives has an average disconnect time of .06. This translates into a difference of 40k SIO in a 15 minute interval. This disparity will last for almost an hour, then the high disconnect time goes away (still using the same four tapes and drives). SMF21 doesn't indicate any Erase Gaps or temporary errors which I know causes high disconnect times. Jack Kelly 202-502-2390 (Office) NOTICE: This electronic mail message and any files transmitted with it are intended exclusively for the individual or entity to which it is addressed. The message, together with any attachment, may contain confidential and/or privileged information. Any unauthorized review, use, printing, saving, copying, disclosure or distribution is strictly prohibited. If you have received this message in error, please immediately advise the sender by reply email and delete all copies. ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: GET IBM-MAIN INFO Search the archives at http://bama.ua.edu/archives/ibm-main.html
