We finally took some steps to attack this problem on the processor that showed the highest I/O retry count.
- During the week, the CE removed all unused ESCON cards on one processor. (Nondisruptive) - On the monthly IPL weekend, we brought down all LPARs on the processor. - We did two PORs: first with the previous IOCDS, then with the current IOCDS. The double POR was advised by the CE in order to force redistribution of the SAPs across remaining chpids. Otherwise a minimal POR with the same IOCDS would not necessarily reassign SAPs. The result is heartening but a trifle puzzling. RMF still reports unbalanced SAP usage, with one being used conspicuously less than the others. But the real problem--excessive I/O retries--appears to be fixed. The count is not zero but far lower than it was before the exercise. . . JO.Skip Robinson Southern California Edison Company SHARE MVS Program Co-Manager 626-302-7535 Office 323-715-0595 Mobile [EMAIL PROTECTED] IBM Mainframe Discussion List <IBM-MAIN@BAMA.UA.EDU> wrote on 03/01/2006 09:38:28 AM: > We're getting reports from RMF of excessively high I/O retries on some > systems. In one case, we've seen the number of retries go higher than the > actual number of I/Os! > > Assuming that the numbers are being reported correctly, does anyone have a > notion of what might cause this anomaly? Some facts: > > - Reports do not detail the I/Os being retried by channel or device; just > a total number > - No channel looks excessively busy according to RMF > - We see this on more than one CEC (all z900) > - SAP/IOP processors are 'unbalanced' > > This last point might be an issue. We have a fairly large number of now > unused ESCON channels whose function has moved over to FICON. It appears > that the three SAPs divide up all channels among themselves, but one of > them--which happens to have the most idle channels--is less than half as > busy as the other two. Maybe we need to physically remove the unused > channels in order to redistribute the SAPs more evenly across the active > channels. ---------------------------------------------------------------------- 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
