Red Hat Bugzilla – Bug 115427
Massive workload combined with ESS flashcopy leading to I/O hang
Last modified: 2007-11-30 17:07:00 EST
Description of problem:
Symptom: FlashCopy IO hang
Problem: State change interrupt handler checks only first request
in cqr-queue, but in case the fastpath for start_IO was successful,
it might happen that the second request is long busy.
Solution: Check the whole cqr-queue for long busy.
LTC bugzilla: 4676
fix is necessary for SCSI device qualification. In last status
meeting Bob Johnson and Hans-Georg Markgraf (IBM) agreed that it need
to be fixed in RHEL 3 Update 2.
True, Bob ?
Ingolf, please attach the patch.
Pete, I will look for the patch.
This is required for DevQual with ESCON, FICON
Why is IBM removing our internal QA contact ?
Adding Brock back in.
Created attachment 97761 [details]
path for device qualification problem
The fix for this problem has just been committed to the RHEL3 U3
patch pool this evening (in kernel version 2.4.21-15.4.EL).
An errata has been issued which should help the problem
described in this bug report. This report is therefore being
closed with a resolution of ERRATA. For more information
on the solution and/or where to find the updated files,
please follow the link below. You may reopen this bug report
if the solution does not work for you.