All of lore.kernel.org
 help / color / mirror / Atom feed
* [NVME-CLI] NVMe Format Issues - HGST Drives
@ 2016-10-24 23:02 Zeta Syanthis
  2016-10-25 12:41 ` Christoph Hellwig
  0 siblings, 1 reply; 4+ messages in thread
From: Zeta Syanthis @ 2016-10-24 23:02 UTC (permalink / raw)


Recently, when running an nvme-format against some HGST NVMe drives, I saw some odd errors crop up in syslog that all looked like this. I've seen it on a few systems now and have become a little more concerned.? I've seen references to this around the mailing list, but am a bit confused. Is this a kernel bug, or is it likely to be a hardware issue?

Oct 14 09:45:45 localhost kernel: nvme 0000:05:00.0: I/O 1 QID 0 timeout, reset controller

For reference, I'm running Centos 7.2 with the 4.4.1 elrepo kernel and nvme-cli 0.8.? I'm running these two commands in sequence:

nvme format -s 1 /dev/nvme0n1
nvme format? /dev/nvme0n1

Thank you,

Zeta Mercy Syanthis
Software Test Engineer
StackVelocity

^ permalink raw reply	[flat|nested] 4+ messages in thread

* [NVME-CLI] NVMe Format Issues - HGST Drives
  2016-10-24 23:02 [NVME-CLI] NVMe Format Issues - HGST Drives Zeta Syanthis
@ 2016-10-25 12:41 ` Christoph Hellwig
  2016-10-26 17:05   ` Zeta Syanthis
  0 siblings, 1 reply; 4+ messages in thread
From: Christoph Hellwig @ 2016-10-25 12:41 UTC (permalink / raw)


Hi Zeta,

a format will probably require a longer timeout than the default 60
seconds, try specfing ? --timeout argument with a longer timeout on the
nvme-cli command line.

^ permalink raw reply	[flat|nested] 4+ messages in thread

* [NVME-CLI] NVMe Format Issues - HGST Drives
  2016-10-25 12:41 ` Christoph Hellwig
@ 2016-10-26 17:05   ` Zeta Syanthis
  2016-10-26 19:24     ` Keith Busch
  0 siblings, 1 reply; 4+ messages in thread
From: Zeta Syanthis @ 2016-10-26 17:05 UTC (permalink / raw)


I tried -t 3600 for an hour, but it hung the couple of systems I tried it  on. What exactly does that timeout do? And is it not in seconds?

Thank you,

Zeta Mercy Syanthis
Software Test Engineer
StackVelocity  

From: Christoph Hellwig <hch@infradead.org>
Sent: Tuesday, October 25, 2016 5:41 AM
To: Zeta Syanthis
Cc: linux-nvme at lists.infradead.org
Subject: Re: [NVME-CLI] NVMe Format Issues - HGST Drives
?   
Hi Zeta,

a format will probably require a longer timeout than the default 60
seconds, try specfing ? --timeout argument with a longer timeout on the
nvme-cli command line.

^ permalink raw reply	[flat|nested] 4+ messages in thread

* [NVME-CLI] NVMe Format Issues - HGST Drives
  2016-10-26 17:05   ` Zeta Syanthis
@ 2016-10-26 19:24     ` Keith Busch
  0 siblings, 0 replies; 4+ messages in thread
From: Keith Busch @ 2016-10-26 19:24 UTC (permalink / raw)


On Wed, Oct 26, 2016@05:05:05PM +0000, Zeta Syanthis wrote:
> I tried -t 3600 for an hour, but it hung the couple of systems I tried it  on. What exactly does that timeout do? And is it not in seconds?

It's in milliseconds. The default when not set uses the kernel's
default, which is 60 seconds, or 60000 milliseconds.

^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2016-10-26 19:24 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2016-10-24 23:02 [NVME-CLI] NVMe Format Issues - HGST Drives Zeta Syanthis
2016-10-25 12:41 ` Christoph Hellwig
2016-10-26 17:05   ` Zeta Syanthis
2016-10-26 19:24     ` Keith Busch

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.