Patrick M. Hausen
2019-Apr-15 08:51 UTC
NVME aborting outstanding i/o and controller resets
> Am 15.04.2019 um 08:46 schrieb Patrick M. Hausen <hausen at punkt.de>: > So I?ll test RELENG_12 next. If that works, I can probably craft > a FreeNAS 11.2 installation with a 12 kernel. I would be hesitating to run > HEAD in production, though.root at hurz:/var/tmp # uname -a FreeBSD hurz 11.2-RELEASE FreeBSD 11.2-RELEASE #0 r335510: Fri Jun 22 04:32:14 UTC 2018 root at releng2.nyi.freebsd.org:/usr/obj/usr/src/sys/GENERIC amd64 root at hurz:/var/tmp # dd if=/dev/urandom of=hurz bs=10m Result: Apr 15 09:56:07 hurz kernel: nvme4: resetting controller Apr 15 09:56:07 hurz kernel: nvme3: resetting controller Apr 15 09:56:07 hurz kernel: nvme4: aborting outstanding i/o Apr 15 09:56:07 hurz kernel: nvme4: WRITE sqid:5 cid:126 nsid:1 lba:188361216 len:208 Apr 15 09:56:07 hurz kernel: nvme4: ABORTED - BY REQUEST (00/07) sqid:5 cid:126 cdw0:0 Apr 15 09:56:07 hurz kernel: nvme4: aborting outstanding i/o Apr 15 09:56:07 hurz kernel: nvme4: WRITE sqid:5 cid:127 nsid:1 lba:188368784 len:64 Apr 15 09:56:07 hurz kernel: nvme4: ABORTED - BY REQUEST (00/07) sqid:5 cid:127 cdw0:0 Apr 15 09:56:07 hurz kernel: nvme4: aborting outstanding i/o Apr 15 09:56:07 hurz kernel: nvme4: WRITE sqid:5 cid:125 nsid:1 lba:188371408 len:48 Apr 15 09:56:07 hurz kernel: nvme4: ABORTED - BY REQUEST (00/07) sqid:5 cid:125 cdw0:0 Apr 15 09:56:07 hurz kernel: nvme4: aborting outstanding i/o Apr 15 09:56:07 hurz kernel: nvme4: WRITE sqid:5 cid:124 nsid:1 lba:188371456 len:16 Apr 15 09:56:07 hurz kernel: nvme4: ABORTED - BY REQUEST (00/07) sqid:5 cid:124 cdw0:0 [?] Now, RELENG_12 kernel, 11.2-RELEASE userland: root at hurz:/var/tmp # uname -a FreeBSD hurz 12.0-STABLE FreeBSD 12.0-STABLE r346220 GENERIC amd64 root at hurz:/var/tmp # dd if=/dev/urandom of=hurz bs=10m Result: no problems, not with two of these jobs running in parallel, not with a zpool scrub at the same time ? I uploaded a complete dmesg of the system running RELENG_12: https://cloud.hausen.com/s/5dRMsewCtDFHRYA Is there anything else I should send? pciconf, nvmecontrol ?? Kind regards Patrick -- punkt.de GmbH Internet - Dienstleistungen - Beratung Kaiserallee 13a Tel.: 0721 9109-0 Fax: -100 76133 Karlsruhe info at punkt.de http://punkt.de AG Mannheim 108285 Gf: Juergen Egeling
Patrick M. Hausen
2019-Apr-15 09:15 UTC
NVME aborting outstanding i/o and controller resets
Hi!> Am 15.04.2019 um 10:51 schrieb Patrick M. Hausen <hausen at punkt.de>: > Now, RELENG_12 kernel, 11.2-RELEASE userland: > > root at hurz:/var/tmp # uname -a > FreeBSD hurz 12.0-STABLE FreeBSD 12.0-STABLE r346220 GENERIC amd64 > root at hurz:/var/tmp # dd if=/dev/urandom of=hurz bs=10m > > Result: > > no problems, not with two of these jobs running in parallel, not with a zpool scrub at the same time ?After they ran for half an hour I find these in /var/log/messages: Apr 15 11:03:54 hurz kernel: nvme2: Missing interrupt Apr 15 11:07:07 hurz kernel: nvme3: Missing interrupt Apr 15 11:09:47 hurz kernel: nvme4: Missing interrupt They are the only occurrences. The system does not seem to hang or otherwise misbehave ... Kind regards Patrick -- punkt.de GmbH Internet - Dienstleistungen - Beratung Kaiserallee 13a Tel.: 0721 9109-0 Fax: -100 76133 Karlsruhe info at punkt.de http://punkt.de AG Mannheim 108285 Gf: Juergen Egeling