From: Richard Weinberger <richard.weinberger@gmail.com>
To: KY Srinivasan <kys@microsoft.com>
Cc: Christoph Hellwig <hch@infradead.org>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
"devel@linuxdriverproject.org" <devel@linuxdriverproject.org>,
"ohering@suse.com" <ohering@suse.com>,
"jbottomley@parallels.com" <jbottomley@parallels.com>,
"jasowang@redhat.com" <jasowang@redhat.com>,
"apw@canonical.com" <apw@canonical.com>,
"linux-scsi@vger.kernel.org" <linux-scsi@vger.kernel.org>
Subject: Re: [PATCH 6/8] Drivers: scsi: storvsc: Implement an abort handler
Date: Sat, 12 Jul 2014 18:16:36 +0200 [thread overview]
Message-ID: <CAFLxGvyJRr4dXEqv2KfCf30Q-kiA6=mPcKB+m4i1y79Xeoqe3A@mail.gmail.com> (raw)
In-Reply-To: <CAFLxGvzFTeE9J-4KFUn9q1fggB_pGBHEMEPcBtDhje+uRm7yMw@mail.gmail.com>
On Thu, Jul 10, 2014 at 12:33 PM, Richard Weinberger
<richard.weinberger@gmail.com> wrote:
> On Wed, Jul 9, 2014 at 8:51 PM, KY Srinivasan <kys@microsoft.com> wrote:
>>
>>
>>> -----Original Message-----
>>> From: Christoph Hellwig [mailto:hch@infradead.org]
>>> Sent: Wednesday, July 9, 2014 1:44 AM
>>> To: KY Srinivasan
>>> Cc: linux-kernel@vger.kernel.org; devel@linuxdriverproject.org;
>>> ohering@suse.com; jbottomley@parallels.com; jasowang@redhat.com;
>>> apw@canonical.com; linux-scsi@vger.kernel.org
>>> Subject: Re: [PATCH 6/8] Drivers: scsi: storvsc: Implement an abort handler
>>>
>>> On Tue, Jul 08, 2014 at 05:46:50PM -0700, K. Y. Srinivasan wrote:
>>> > Implement a simple abort handler. The host does not support "Abort";
>>> > just ensure that all inflight I/Os have been accounted for.
>>>
>>> The abort handler should abort a single command, not wait for all of them.
>>> What issue do you see that this tries to address?
>>
>> On Azure, we sometimes have unbounded I/O latencies and some distributions (such as SLES12) based on recent kernels are invoking
>> the "Abort Handler". Unfortunately, our scsi emulation on the host does not support aborting a command.
>> The issue I have seen is that the upper level scsi code attempts error recovery when the command times out and finally frees up the command.
>> The host subsequently responds to the command that has timed out and since the memory has been freed up, we end up touching freed memory
>> in this driver. Since the host is also doing error recovery, by just delaying the error handler in the guest until we can account for all the in-flight commands,
>> we can get around the problem.
>
> I see strange issues in Azure and maybe they are related to this.
> Some Linux machines crash in a way that no disk IO is possible (thus,
> no SSH for me) but they still respond to
> ping. It happens rather seldom (every few weeks).
>
> Do you see similar symptoms?
ping?
--
Thanks,
//richard
WARNING: multiple messages have this Message-ID (diff)
From: Richard Weinberger <richard.weinberger@gmail.com>
To: KY Srinivasan <kys@microsoft.com>
Cc: "linux-scsi@vger.kernel.org" <linux-scsi@vger.kernel.org>,
"jasowang@redhat.com" <jasowang@redhat.com>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
"jbottomley@parallels.com" <jbottomley@parallels.com>,
"ohering@suse.com" <ohering@suse.com>,
Christoph Hellwig <hch@infradead.org>,
"apw@canonical.com" <apw@canonical.com>,
"devel@linuxdriverproject.org" <devel@linuxdriverproject.org>
Subject: Re: [PATCH 6/8] Drivers: scsi: storvsc: Implement an abort handler
Date: Sat, 12 Jul 2014 18:16:36 +0200 [thread overview]
Message-ID: <CAFLxGvyJRr4dXEqv2KfCf30Q-kiA6=mPcKB+m4i1y79Xeoqe3A@mail.gmail.com> (raw)
In-Reply-To: <CAFLxGvzFTeE9J-4KFUn9q1fggB_pGBHEMEPcBtDhje+uRm7yMw@mail.gmail.com>
On Thu, Jul 10, 2014 at 12:33 PM, Richard Weinberger
<richard.weinberger@gmail.com> wrote:
> On Wed, Jul 9, 2014 at 8:51 PM, KY Srinivasan <kys@microsoft.com> wrote:
>>
>>
>>> -----Original Message-----
>>> From: Christoph Hellwig [mailto:hch@infradead.org]
>>> Sent: Wednesday, July 9, 2014 1:44 AM
>>> To: KY Srinivasan
>>> Cc: linux-kernel@vger.kernel.org; devel@linuxdriverproject.org;
>>> ohering@suse.com; jbottomley@parallels.com; jasowang@redhat.com;
>>> apw@canonical.com; linux-scsi@vger.kernel.org
>>> Subject: Re: [PATCH 6/8] Drivers: scsi: storvsc: Implement an abort handler
>>>
>>> On Tue, Jul 08, 2014 at 05:46:50PM -0700, K. Y. Srinivasan wrote:
>>> > Implement a simple abort handler. The host does not support "Abort";
>>> > just ensure that all inflight I/Os have been accounted for.
>>>
>>> The abort handler should abort a single command, not wait for all of them.
>>> What issue do you see that this tries to address?
>>
>> On Azure, we sometimes have unbounded I/O latencies and some distributions (such as SLES12) based on recent kernels are invoking
>> the "Abort Handler". Unfortunately, our scsi emulation on the host does not support aborting a command.
>> The issue I have seen is that the upper level scsi code attempts error recovery when the command times out and finally frees up the command.
>> The host subsequently responds to the command that has timed out and since the memory has been freed up, we end up touching freed memory
>> in this driver. Since the host is also doing error recovery, by just delaying the error handler in the guest until we can account for all the in-flight commands,
>> we can get around the problem.
>
> I see strange issues in Azure and maybe they are related to this.
> Some Linux machines crash in a way that no disk IO is possible (thus,
> no SSH for me) but they still respond to
> ping. It happens rather seldom (every few weeks).
>
> Do you see similar symptoms?
ping?
--
Thanks,
//richard
next prev parent reply other threads:[~2014-07-12 16:16 UTC|newest]
Thread overview: 117+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-07-09 0:46 [PATCH 0/8] Drivers: scsi: storvsc: Bug fixes and improvements K. Y. Srinivasan
2014-07-09 0:46 ` K. Y. Srinivasan
2014-07-09 0:46 ` [PATCH 1/8] Drivers: scsi: storvsc: Change the limits to reflect the values on the host K. Y. Srinivasan
2014-07-09 0:46 ` K. Y. Srinivasan
2014-07-09 0:46 ` [PATCH 2/8] Drivers: scsi: storvsc: Filter commands based on the storage protocol version K. Y. Srinivasan
2014-07-09 0:46 ` K. Y. Srinivasan
2014-07-09 8:40 ` Christoph Hellwig
2014-07-09 20:02 ` KY Srinivasan
2014-07-09 20:02 ` KY Srinivasan
2014-07-09 0:46 ` [PATCH 3/8] Drivers: scsi: storvsc: Fix a bug in handling VMBUS " K. Y. Srinivasan
2014-07-09 0:46 ` K. Y. Srinivasan
2014-07-09 8:41 ` Christoph Hellwig
2014-07-09 20:01 ` KY Srinivasan
2014-07-09 0:46 ` [PATCH 4/8] Drivers: scsi: storvsc: Filter WRITE_SAME_16 K. Y. Srinivasan
2014-07-09 0:46 ` K. Y. Srinivasan
2014-07-09 8:43 ` Christoph Hellwig
2014-07-09 8:43 ` Christoph Hellwig
2014-07-09 19:52 ` KY Srinivasan
2014-07-09 19:52 ` KY Srinivasan
2014-07-09 19:56 ` James Bottomley
2014-07-09 19:56 ` James Bottomley
2014-07-09 21:14 ` KY Srinivasan
2014-07-09 21:14 ` KY Srinivasan
2014-07-09 22:27 ` James Bottomley
2014-07-09 22:27 ` James Bottomley
2014-07-09 22:36 ` KY Srinivasan
2014-07-10 10:07 ` hch
2014-07-10 21:02 ` KY Srinivasan
2014-07-10 21:02 ` KY Srinivasan
2014-07-10 22:12 ` James Bottomley
2014-07-10 22:12 ` James Bottomley
2014-07-10 22:12 ` James Bottomley
2014-07-11 6:32 ` hch
2014-07-11 6:32 ` hch
2014-07-11 12:54 ` Martin K. Petersen
2014-07-11 12:54 ` Martin K. Petersen
2014-07-12 2:53 ` KY Srinivasan
2014-07-12 2:50 ` KY Srinivasan
2014-07-13 12:58 ` Martin K. Petersen
2014-07-13 12:58 ` Martin K. Petersen
2014-07-13 18:49 ` KY Srinivasan
2014-07-13 18:49 ` KY Srinivasan
2014-07-14 2:37 ` KY Srinivasan
2014-07-14 2:37 ` KY Srinivasan
2014-07-14 3:19 ` KY Srinivasan
2014-07-16 11:01 ` hch
2014-07-16 11:01 ` hch
2014-07-16 14:53 ` James Bottomley
2014-07-16 14:53 ` James Bottomley
2014-07-16 15:44 ` Martin K. Petersen
2014-07-16 15:44 ` Martin K. Petersen
2014-07-16 17:38 ` hch
2014-07-16 17:38 ` hch
2014-07-16 17:47 ` Martin K. Petersen
2014-07-16 17:47 ` Martin K. Petersen
2014-07-16 17:55 ` hch
2014-07-16 17:55 ` hch
2014-07-16 18:02 ` James Bottomley
2014-07-16 18:43 ` Elliott, Robert (Server Storage)
2014-07-16 19:20 ` Martin K. Petersen
2014-07-16 19:20 ` Martin K. Petersen
2014-07-17 7:42 ` hch
2014-07-17 12:41 ` Martin K. Petersen
2014-07-17 12:41 ` Martin K. Petersen
2014-07-16 18:45 ` Martin K. Petersen
2014-07-16 18:45 ` Martin K. Petersen
2014-07-16 18:50 ` James Bottomley
2014-07-16 18:50 ` James Bottomley
2014-07-16 19:08 ` Martin K. Petersen
2014-07-16 19:08 ` Martin K. Petersen
2014-07-16 19:14 ` James Bottomley
2014-07-16 19:14 ` James Bottomley
2014-07-16 19:44 ` Martin K. Petersen
2014-07-16 19:44 ` Martin K. Petersen
2014-07-09 0:46 ` [PATCH 5/8] Drivers: scsi: storvsc: Fix a bug in the handling of SRB status flags K. Y. Srinivasan
2014-07-09 0:46 ` K. Y. Srinivasan
2014-07-09 8:43 ` Christoph Hellwig
2014-07-09 8:43 ` Christoph Hellwig
2014-07-09 0:46 ` [PATCH 6/8] Drivers: scsi: storvsc: Implement an abort handler K. Y. Srinivasan
2014-07-09 0:46 ` K. Y. Srinivasan
2014-07-09 8:44 ` Christoph Hellwig
2014-07-09 8:44 ` Christoph Hellwig
2014-07-09 18:51 ` KY Srinivasan
2014-07-10 10:13 ` Christoph Hellwig
2014-07-10 10:13 ` Christoph Hellwig
2014-07-10 22:26 ` KY Srinivasan
2014-07-11 9:52 ` Hannes Reinecke
2014-07-11 9:52 ` Hannes Reinecke
2014-07-11 9:54 ` Christoph Hellwig
2014-07-11 9:54 ` Christoph Hellwig
2014-07-11 14:49 ` KY Srinivasan
2014-07-11 14:49 ` KY Srinivasan
2014-07-10 10:33 ` Richard Weinberger
2014-07-12 16:16 ` Richard Weinberger [this message]
2014-07-12 16:16 ` Richard Weinberger
2014-07-12 16:35 ` KY Srinivasan
2014-07-12 16:35 ` KY Srinivasan
2014-07-09 0:46 ` [PATCH 7/8] drivers: scsi: storvsc: Set srb_flags in all cases K. Y. Srinivasan
2014-07-09 0:46 ` K. Y. Srinivasan
2014-07-09 8:44 ` Christoph Hellwig
2014-07-09 18:40 ` KY Srinivasan
2014-07-09 18:40 ` KY Srinivasan
2014-07-10 10:17 ` Christoph Hellwig
2014-07-10 22:22 ` KY Srinivasan
2014-07-10 10:18 ` Christoph Hellwig
2014-07-10 10:18 ` Christoph Hellwig
2014-07-11 21:03 ` KY Srinivasan
2014-07-09 0:46 ` [PATCH 8/8] drivers: scsi: storvsc: Correctly handle TEST_UNIT_READY failure K. Y. Srinivasan
2014-07-09 0:46 ` K. Y. Srinivasan
2014-07-09 8:46 ` Christoph Hellwig
2014-07-09 18:35 ` KY Srinivasan
2014-07-09 18:35 ` KY Srinivasan
2014-07-09 8:39 ` [PATCH 1/8] Drivers: scsi: storvsc: Change the limits to reflect the values on the host Christoph Hellwig
2014-07-09 20:07 ` KY Srinivasan
2014-07-09 20:07 ` KY Srinivasan
2014-07-09 23:39 ` KY Srinivasan
2014-07-09 23:39 ` KY Srinivasan
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to='CAFLxGvyJRr4dXEqv2KfCf30Q-kiA6=mPcKB+m4i1y79Xeoqe3A@mail.gmail.com' \
--to=richard.weinberger@gmail.com \
--cc=apw@canonical.com \
--cc=devel@linuxdriverproject.org \
--cc=hch@infradead.org \
--cc=jasowang@redhat.com \
--cc=jbottomley@parallels.com \
--cc=kys@microsoft.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-scsi@vger.kernel.org \
--cc=ohering@suse.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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.