Received-SPF: pass (google.com: best guess record for domain of linux-nfs-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Content-Type: text/plain;
        charset=us-ascii
Mime-Version: 1.0 (Mac OS X Mail 12.4 \(3445.104.11\))
Subject: Re: [PATCH 3/3] mountstats: Check for RPC iostats version >= 1.1 with
 error counts
From:   Chuck Lever <chuck.lever@oracle.com>
In-Reply-To: <7211D5DF-6923-459D-9B84-2BD264EB9F11@oracle.com>
Date:   Wed, 19 Jun 2019 13:45:43 -0400
Cc:     Steve Dickson <SteveD@redhat.com>,
        Linux NFS Mailing List <linux-nfs@vger.kernel.org>
Content-Transfer-Encoding: quoted-printable
Message-Id: <E3BDBBD8-C75C-48EA-85D5-37657DF3AE14@oracle.com>
References: <20190612190229.31811-1-dwysocha@redhat.com>
 <20190613120314.1864-1-dwysocha@redhat.com>
 <20190613120314.1864-3-dwysocha@redhat.com>
 <FD291454-53BB-46DB-BEBE-9AA2F8DE18DE@oracle.com>
 <CALF+zOkFKXZQsFodJphAg1UBNxKyQq_GfO1wFqfak0TTre=dng@mail.gmail.com>
 <7211D5DF-6923-459D-9B84-2BD264EB9F11@oracle.com>
To:     David Wysochanski <dwysocha@redhat.com>
Sender: linux-nfs-owner@vger.kernel.org
Precedence: bulk


> On Jun 19, 2019, at 1:42 PM, Chuck Lever <chuck.lever@oracle.com> =
wrote:
>=20
>=20
>=20
>> On Jun 19, 2019, at 1:22 PM, David Wysochanski <dwysocha@redhat.com> =
wrote:
>>=20
>>=20
>>=20
>> On Wed, Jun 19, 2019 at 12:35 PM Chuck Lever <chuck.lever@oracle.com> =
wrote:
>>=20
>>=20
>>> On Jun 13, 2019, at 8:03 AM, Dave Wysochanski <dwysocha@redhat.com> =
wrote:
>>>=20
>>> Add explicit check for statsvers instead of array based check.
>>=20
>> Hi Dave,
>>=20
>> I don't understand why this change is necessary. The patch =
description
>> should explain.
>>=20
>>=20
>> Steve had already taken commit 73491ef for mountstats which was an =
array based check.  This just makes this patch consistent with the =
others.  Is that what you mean - you want a statement about consistency =
and refer to the other commit?  How about:
>>=20
>> Commit 73491ef added per-op error counts for mountstats command but =
used an array based check rather than checking statsver. Add explicit =
check for statsver instead of array based check for consistency with =
other tools.
>=20
> This is a better patch description (explains "why" not "what"),
> but I'm not clear why this change is necessary in either place.

In other words, was this change necessary to fix a bug? Or is
this a defensive change to make parsing more robust?


>>> Signed-off-by: Dave Wysochanski <dwysocha@redhat.com>
>>> ---
>>> tools/mountstats/mountstats.py | 2 +-
>>> 1 file changed, 1 insertion(+), 1 deletion(-)
>>>=20
>>> diff --git a/tools/mountstats/mountstats.py =
b/tools/mountstats/mountstats.py
>>> index 5f13bf8e..2ebbf945 100755
>>> --- a/tools/mountstats/mountstats.py
>>> +++ b/tools/mountstats/mountstats.py
>>> @@ -476,7 +476,7 @@ class DeviceData:
>>>                if retrans !=3D 0:
>>>                    print('\t%d retrans (%d%%)' % (retrans, ((retrans =
* 100) / count)), end=3D' ')
>>>                    print('\t%d major timeouts' % stats[3], end=3D'')
>>> -                if len(stats) >=3D 10 and stats[9] !=3D 0:
>>> +                if self.__rpc_data['statsvers'] >=3D 1.1 and =
stats[9] !=3D 0:
>>>                    print('\t%d errors (%d%%)' % (stats[9], =
((stats[9] * 100) / count)))
>>>                else:
>>>                    print('')
>>> --=20
>>> 2.20.1
>>>=20
>>=20
>> --
>> Chuck Lever
>>=20
>>=20
>>=20
>>=20
>>=20
>> --=20
>> Dave Wysochanski
>> Principal Software Maintenance Engineer
>> T: 919-754-4024 =20
>=20
> --
> Chuck Lever

--
Chuck Lever