On Thu, 2018-12-06 at 10:02 -0800, Bart Van Assche wrote: +AD4 On Thu, 2018-12-06 at 10:48 -0700, Jens Axboe wrote: +AD4 +AD4 which would result in a non-zero exit, which should be expected for this +AD4 +AD4 test? +AD4 +AD4 Test srp/002 simulates network failures while running fio on top of dm-mpath. +AD4 Since queue+AF8-if+AF8-no+AF8-path is enabled in multipath.conf dm-mpath will keep retrying +AD4 the block layer requests it receives if these requests fail due to path removal. +AD4 If fio reports +ACI-io+AF8-u error+ACI for this test then that means that the test failed. +AD4 I haven't seen fio reporting any I/O errors for this test with any upstream +AD4 kernel that I tested in the past year or so. Hi Jens, Please also verify that the version of multipath-tools that you are running includes the following patch: >From a2675025ae9f652b005345b9082f5042b32c992c Mon Sep 17 00:00:00 2001 From: Bart Van Assche +ADw-bart.vanassche+AEA-sandisk.com+AD4 Date: Fri, 18 Nov 2016 13:33:44 -0800 Subject: +AFs-PATCH+AF0 Avoid that reloading a map sporadically triggers I/O errors Avoid that reloading a map while there are no paths triggers a flush and hence unwanted I/O errors if 'queue+AF8-if+AF8-no+AF8-path' is enabled. Fixes: commit d569988e7528 (+ACI-libmultipath: Fixup 'DM+AF8-DEVICE+AF8-RELOAD' handling+ACI) Signed-off-by: Bart Van Assche +ADw-bart.vanassche+AEA-sandisk.com+AD4 --- libmultipath/devmapper.c +AHw 2 +-- 1 file changed, 1 insertion(+-), 1 deletion(-) diff --git a/libmultipath/devmapper.c b/libmultipath/devmapper.c index 9b6b0537ba6d..1576dd017d6a 100644 --- a/libmultipath/devmapper.c +-+-+- b/libmultipath/devmapper.c +AEAAQA -392,7 +-392,7 +AEAAQA int dm+AF8-addmap+AF8-reload(struct multipath +ACo-mpp, char +ACo-params, int flush) params, ADDMAP+AF8-RO, SKIP+AF8-KPARTX+AF8-OFF)+ADs +AH0 if (r) - r +AD0 dm+AF8-simplecmd(DM+AF8-DEVICE+AF8-RESUME, mpp-+AD4-alias, flush, +- r +AD0 dm+AF8-simplecmd(DM+AF8-DEVICE+AF8-RESUME, mpp-+AD4-alias, +ACE-flush, 1, udev+AF8-flags, 0)+ADs return r+ADs +AH0
