From nobody Fri Jul 21 03:26:07 2023 X-Original-To: scsi@mlmmj.nyi.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2610:1c1:1:606c::19:1]) by mlmmj.nyi.freebsd.org (Postfix) with ESMTP id 4R6Zhs0hNqz4nwLH for ; Fri, 21 Jul 2023 03:26:21 +0000 (UTC) (envelope-from wlosh@bsdimp.com) Received: from mail-ej1-x62f.google.com (mail-ej1-x62f.google.com [IPv6:2a00:1450:4864:20::62f]) (using TLSv1.3 with cipher TLS_AES_128_GCM_SHA256 (128/128 bits) key-exchange X25519 server-signature RSA-PSS (4096 bits) server-digest SHA256 client-signature RSA-PSS (2048 bits) client-digest SHA256) (Client CN "smtp.gmail.com", Issuer "GTS CA 1D4" (verified OK)) by mx1.freebsd.org (Postfix) with ESMTPS id 4R6Zhr5ZWgz3DBV for ; Fri, 21 Jul 2023 03:26:20 +0000 (UTC) (envelope-from wlosh@bsdimp.com) Authentication-Results: mx1.freebsd.org; none Received: by mail-ej1-x62f.google.com with SMTP id a640c23a62f3a-9922d6f003cso237769866b.0 for ; Thu, 20 Jul 2023 20:26:20 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=bsdimp-com.20221208.gappssmtp.com; s=20221208; t=1689909979; x=1690514779; h=cc:to:subject:message-id:date:from:in-reply-to:references :mime-version:from:to:cc:subject:date:message-id:reply-to; bh=lBPJQQLUPIAujvywawulGetLsBEuZ+/GDH3xHRepNYU=; b=28sSZa5jwc1eLWJxDuAgu1vGy1svseVYTKYxA/qnSdTdTzvqUP6sTFgp0ar4lgJPEF BZfh1ykxhtBQIIirtkz+D2DHS5/6/s9AZISGRoTVtjXdIenDL5y7HlRk/rSANKU2nfnP ozjaI7lEbymfeM0uP5tUX2AC695xCs/xTdOXT0/+bF6S5uZTzLRGz94Vl4qmpuXZzQqq DDPBkh/kx2YPafAS+W3C3aAntrPZX5I4UAO8p6cJqjkSAD1UWwRHitqTz3WUObdveqjc bO+SKEBeKApDTK8+hqUFr/IibzQD8e7YKDw344bJKIb1SSnYQ8zLRv6iC8jO/yyUi4dD D6pA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20221208; t=1689909979; x=1690514779; h=cc:to:subject:message-id:date:from:in-reply-to:references :mime-version:x-gm-message-state:from:to:cc:subject:date:message-id :reply-to; bh=lBPJQQLUPIAujvywawulGetLsBEuZ+/GDH3xHRepNYU=; b=WR2UB7V4RZmJyFh+ErCmArImzpuJ9twRjWfQ8zHRf9SNC8X2rll1IuCo+A6figTEMt YZHPT5YVqImF+OIsy6XORvi46FiSGWlJfQwN2ybOabLIifZsuI7JYcB3JJb5229dxME3 NIunmGyp35WRVS/hyQZUw0C8EtvNxs4a/MWxxs0199ca7KRl+/48IWkug5Yv3JTZReg5 yDXDs2bKI99AZJPW8xk6H8ceXYO6ifWjb4B7D7uz8DnO57A5MVmqkQqDfSyvinQ7hqp7 gueMNVHw34iHEcmCGibtDDgsVUyHUNqGLTt7QMKocaGr+VrZHMn+AWiW+zV5UCaS34tn kuiw== X-Gm-Message-State: ABy/qLbKHRjgyKBhhi8sHwJHfDu3WIgFj7ckings//Ab9UgPDpWeV+gj j55xJbQapdLwUyhhceV5O8a0j6YcHt8ozPOZYh7EsQ== X-Google-Smtp-Source: APBJJlEr4sZPZKt1pSvRmqKfaAP9GYJSKGJ8QkUVQZLj5hdhgIZselRTINa8olnaY5hG5PtRxVQS8bHbanBl3K9H9PY= X-Received: by 2002:a17:906:8451:b0:994:1fd2:cf96 with SMTP id e17-20020a170906845100b009941fd2cf96mr588912ejy.0.1689909978614; Thu, 20 Jul 2023 20:26:18 -0700 (PDT) List-Id: SCSI subsystem List-Archive: https://lists.freebsd.org/archives/freebsd-scsi List-Help: List-Post: List-Subscribe: List-Unsubscribe: Sender: owner-freebsd-scsi@freebsd.org X-BeenThere: freebsd-scsi@freebsd.org MIME-Version: 1.0 References: <7df852e4-5df5-de51-70a6-08bcbcb2f757@interlog.com> In-Reply-To: <7df852e4-5df5-de51-70a6-08bcbcb2f757@interlog.com> From: Warner Losh Date: Thu, 20 Jul 2023 21:26:07 -0600 Message-ID: Subject: Re: ASC/ASCQ Review To: dgilbert@interlog.com Cc: Alan Somers , scsi@freebsd.org Content-Type: multipart/alternative; boundary="000000000000fa60eb0600f6d3dc" X-Rspamd-Queue-Id: 4R6Zhr5ZWgz3DBV X-Spamd-Bar: ---- X-Spamd-Result: default: False [-4.00 / 15.00]; REPLY(-4.00)[]; ASN(0.00)[asn:15169, ipnet:2a00:1450::/32, country:US] X-Rspamd-Pre-Result: action=no action; module=replies; Message is reply to one we originated --000000000000fa60eb0600f6d3dc Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable On Thu, Jul 20, 2023, 9:18 PM Douglas Gilbert wrote= : > On 2023-07-19 11:41, Warner Losh wrote: > > btw, it also occurs to me that if I do add a 'secondary' table, then yo= u > could > > use it to generate a unique errno and experiment > > with that w/o affecting the main code until that stuff was mature. > > > > I'm not sure I'll do that now, since I've found maybe 10 asc/ascq pairs > that I'd > > like to tag as 'if trying harder, retry, otherwise fail' since re-retry > needs > > have changed a lot since cam was written in the late 90s and at least > some of > > the asc/ascq pairs I'm looking at haven't changed since the initial > import, but > > that's based on a tiny sampling of the data I have and is preliminary a= t > best. I > > may just change it to reflect modern usage. > > Hi, > If you are looking for up-to-date [20230325] asc/ascq tables in C you cou= ld > borrow mine at https://github.com/doug-gilbert/sg3_utils in > lib/sg_lib_data.c > starting at line 745 . > In testing/sg_chk_asc.c is a small test program for checking that the > table in > sg_lib_data.c agrees with the file that T10 supplies: > https://www.t10.org/lists/asc-num.txt Thanks for the pointer. I'd already updated CAM's tables for that... what I'm doing now is to make sure CAM's reactions to the asc/ascq is good for the modern drives... it's a good idea though to create a program for our table to match... Warner > Doug Gilbert > > > On Fri, Jul 14, 2023 at 5:34=E2=80=AFPM Warner Losh > > wrote: > > > > > > > > On Fri, Jul 14, 2023 at 12:31=E2=80=AFPM Alan Somers > > wrote: > > > > On Fri, Jul 14, 2023 at 11:05=E2=80=AFAM Warner Losh > > wrote: > > > > > > > > > > > > On Fri, Jul 14, 2023, 11:12 AM Alan Somers < > asomers@freebsd.org > > > wrote: > > >> > > >> On Thu, Jul 13, 2023 at 12:14=E2=80=AFPM Warner Losh > > wrote: > > >> > > > >> > Greetings, > > >> > > > >> > i've been looking closely at failed drives for $WORK > lately. I've > > noticed that a lot of errors that kinda sound like fatal errors > have > > SS_RDEF set on them. > > >> > > > >> > What's the process for evaluating whether those error > codes are > > worth retrying. There are several errors that we seem to be > seeing > > (preliminary read of the data) before the drive gives up the > ghost > > altogether. For those cases, I'd like to post more specific > lists. > > Should I do that here? > > >> > > > >> > Independent of that, I may want to have a more aggressive > 'fail > > fast' policy than is appropriate for my work load (we have a lo= t > of data > > that's a copy of a copy of a copy, so if we lose it, we don't > care: > > we'll just delete any files we can't read and get on with life, > though I > > know others will have a more conservative attitude towards data > that > > might be precious and unique). I can set the number of retries > lower, I > > can do some other hacks for disks that tell the disk to fail > faster, but > > I think part of the solution is going to have to be failing for > some > > sense-code/ASC/ASCQ tuples that we don't want to fail in > upstream or the > > general case. I was thinking of identifying those and creating = a > 'global > > quirk table' that gets applied after the drive-specific quirk > table that > > would let $WORK override the defaults, while letting others kee= p > the > > current behavior. IMHO, it would be better to have these > separate rather > > than in the global data for tracking upstream... > > >> > > > >> > Is that clear, or should I give concrete examples? > > >> > > > >> > Comments? > > >> > > > >> > Warner > > >> > > >> Basically, you want to change the retry counts for certain > ASC/ASCQ > > >> codes only, on a site-by-site basis? That sounds > reasonable. Would > > >> it be configurable at runtime or only at build time? > > > > > > > > > I'd like to change the default actions. But maybe we just do > that for > > everyone and assume modern drives... > > > > > >> Also, I've been thinking lately that it would be real nice > if READ > > >> UNRECOVERABLE could be translated to EINTEGRITY instead of > EIO. That > > >> would let consumers know that retries are pointless, but > that the data > > >> is probably healable. > > > > > > > > > Unlikely, unless you've tuned things to not try for long at > recovery... > > > > > > But regardless... do you have a concrete example of a use > case? > > There's a number of places that map any error to EIO. And I'd > like a use > > case before we expand the errors the lower layers return... > > > > > > Warner > > > > My first use-case is a user-space FUSE file system. It only ha= s > > access to errnos, not ASC/ASCQ codes. If we do as I suggest, > then it > > could heal a READ UNRECOVERABLE by rewriting the sector, wherea= s > other > > EIO errors aren't likely to be healed that way. > > > > > > Yea... but READ UNRECOVERABLE is kinda hit or miss... > > > > My second use-case is ZFS. zfsd treats checksum errors > differently > > from I/O errors. A checksum error normally means that a read > returned > > wrong data. But I think that READ UNRECOVERABLE should also > count. > > After all, that means that the disk's media returned wrong data > which > > was detected by the disk's own EDC/ECC. I've noticed that zfsd > seems > > to fault disks too eagerly when their only problem is READ > > UNRECOVERABLE errors. Mapping it to EINTEGRITY, or even a new > error > > code, would let zfsd be tuned better. > > > > > > EINTEGRITY would then mean two different things. UFS returns in whe= n > > checksums fail for critical filesystem errors. I'm not saying no, > per se, > > just that it conflates two different errors. > > > > I think both of these use cases would be better served by CAM's > publishing > > of the errors to devctl today. Here's some example data from a > system I'm > > looking at: > > > > system=3DCAM subsystem=3Dperiph type=3Dtimeout device=3Dda36 serial= =3D"12345" > > cam_status=3D"0x44b" timeout=3D30000 CDB=3D"28 00 4e b7 cb a3 00 04= cc 00 " > > timestamp=3D1634739729.312068 > > system=3DCAM subsystem=3Dperiph type=3Dtimeout device=3Dda36 serial= =3D"12345" > > cam_status=3D"0x44b" timeout=3D30000 CDB=3D"28 00 20 6b d5 56 00 00= c0 00 " > > timestamp=3D1634739729.585541 > > system=3DCAM subsystem=3Dperiph type=3Derror device=3Dda36 serial= =3D"12345" > > cam_status=3D"0x4cc" scsi_status=3D2 scsi_sense=3D"72 03 11 00" CDB= =3D"28 00 > ad 1a > > 35 96 00 00 56 00 " timestamp=3D1641979267.469064 > > system=3DCAM subsystem=3Dperiph type=3Derror device=3Dda36 serial= =3D"12345" > > cam_status=3D"0x4cc" scsi_status=3D2 scsi_sense=3D"72 03 11 00" CDB= =3D"28 00 > ad 1a > > 35 96 00 01 5e 00 " timestamp=3D1642252539.693699 > > system=3DCAM subsystem=3Dperiph type=3Derror device=3Dda39 serial= =3D"12346" > > cam_status=3D"0x4cc" scsi_status=3D2 scsi_sense=3D"72 04 02 00" CDB= =3D"2a 00 > 01 2b > > c8 f6 00 07 81 00 " timestamp=3D1669603144.090835 > > > > Here we get the sense key, the asc and the ascq in the scsi_sense > data (I'm > > currently looking at expanding this to the entire sense buffer, > since it > > includes how hard the drive tried to read the data on media and > hardware > > errors). It doesn't include nvme data, but does include ata data > (I'll have > > to add that data, now that I've noticed it is missing). With the > sense data > > and the CDB you know what kind of error you got, plus what block > didn't > > read/write correctly. With the extended sense data, you can find ou= t > even > > more details that are sense-key dependent... > > > > So I'm unsure that trying to shoehorn our imperfect knowledge of > what's > > retriable, fixable, should be written with zeros into the kernel an= d > > converting that to a separate errno would give good results, and > tapping > > into this stream daemons that want to make more nuanced calls about > disks > > might be the better way to go. One of the things I'm planning for > $WORK is > > to enable the retry time limit of one of the mode pages so that we > fail > > faster and can just delete the file with the 'bad' block that we'd > get > > eventually if we allowed the full, default error processing to run, > but that > > 'slow path' processing kills performance for all other users of the > > drive... I'm unsure how well that will work out (and I know I'm > lucky that > > I can always recover any data for my application since it's just a > cache). > > > > I'd be interested to hear what others have to say here thought, > since my > > focus on this data is through the lense of my rather specialized > application... > > > > Warner > > > > P.S. That was generated with this rule if you wanted to play with > it... > > You'd have to translate absolute disk blocks to a partition and an > offset > > into the filesystem, then give the filesystem a chance to tell you > what of > > its data/metadata that block is used for... > > > > # Disk errors > > notify 10 { > > match "system" "CAM"; > > match "subsystem" "periph"; > > match "device" "[an]?da[0-9]+"; > > action "logger -t diskerr -p daemon.info < > http://daemon.info> $_ > > timestamp=3D$timestamp"; > > }; > > > > --000000000000fa60eb0600f6d3dc Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable


On Thu, Jul 20, 2023, 9:18 PM Douglas Gilbert <d= gilbert@interlog.com> wrote:
On 2023-07-19 11:41, Warner Losh wrote:
> btw, it also occurs to me that if I do add a 'secondary' table= , then you could
> use it to generate a unique errno and experiment
> with that w/o affecting the main code until that stuff was mature.
>
> I'm not sure I'll do that now, since I've found maybe 10 a= sc/ascq pairs that I'd
> like to tag as 'if trying harder, retry, otherwise fail' since= re-retry needs
> have changed a lot since cam was written in the late 90s and at least = some of
> the asc/ascq pairs I'm looking at haven't changed since the in= itial import, but
> that's based on a tiny sampling of the data I have and is prelimin= ary at best. I
> may just change it to reflect modern usage.

Hi,
If you are looking for up-to-date [20230325] asc/ascq tables in C you could=
borrow mine at https://github.com/dou= g-gilbert/sg3_utils in lib/sg_lib_data.c
starting at line 745 .
In testing/sg_chk_asc.c is a small test program for checking that the table= in
sg_lib_data.c agrees with the file that T10 supplies:
=C2=A0 =C2=A0 =C2=A0 https://www.t10.org= /lists/asc-num.txt

=
Thanks for the pointer. I'd already updated CAM's= tables for that...

what= I'm doing now is to make sure CAM's reactions to the asc/ascq is g= ood for the modern drives... it's a good idea though to create a progra= m for our table to match...

Warner


Doug Gilbert

> On Fri, Jul 14, 2023 at 5:34=E2=80=AFPM Warner Losh <imp@bsd= imp.com
> <mailto:imp@bsdimp.com>> wrote:
>
>
>
>=C2=A0 =C2=A0 =C2=A0On Fri, Jul 14, 2023 at 12:31=E2=80=AFPM Alan Somer= s <asomers@freebsd.org
>=C2=A0 =C2=A0 =C2=A0<mailto:asomers@freebsd.org>&= gt; wrote:
>
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0On Fri, Jul 14, 2023 at 11:05=E2=80= =AFAM Warner Losh <imp@bsdimp.com
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0<mailto:imp@bsdimp.com&= gt;> wrote:
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 >
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 >
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 >
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 > On Fri, Jul 14, 2023, 11:12 AM = Alan Somers <asomers@freebsd.org
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0<mailto:asomers@freebs= d.org>> wrote:
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 >>
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 >> On Thu, Jul 13, 2023 at 12:= 14=E2=80=AFPM Warner Losh <imp@bsdimp.com
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0<mailto:imp@bsdimp.com&= gt;> wrote:
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 >> >
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 >> > Greetings,
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 >> >
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 >> > i've been looking = closely at failed drives for $WORK lately. I've
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0noticed that a lot of errors that kin= da sound like fatal errors have
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0SS_RDEF set on them.
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 >> >
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 >> > What's the process= for evaluating whether those error codes are
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0worth retrying. There are several err= ors that we seem to be seeing
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0(preliminary read of the data) before= the drive gives up the ghost
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0altogether. For those cases, I'd = like to post more specific lists.
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0Should I do that here?
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 >> >
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 >> > Independent of that, I= may want to have a more aggressive 'fail
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0fast' policy than is appropriate = for my work load (we have a lot of data
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0that's a copy of a copy of a copy= , so if we lose it, we don't care:
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0we'll just delete any files we ca= n't read and get on with life, though I
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0know others will have a more conserva= tive attitude towards data that
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0might be precious and unique). I can = set the number of retries lower, I
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0can do some other hacks for disks tha= t tell the disk to fail faster, but
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0I think part of the solution is going= to have to be failing for some
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0sense-code/ASC/ASCQ tuples that we do= n't want to fail in upstream or the
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0general case. I was thinking of ident= ifying those and creating a 'global
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0quirk table' that gets applied af= ter the drive-specific quirk table that
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0would let $WORK override the defaults= , while letting others keep the
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0current behavior. IMHO, it would be b= etter to have these separate rather
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0than in the global data for tracking = upstream...
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 >> >
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 >> > Is that clear, or shou= ld I give concrete examples?
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 >> >
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 >> > Comments?
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 >> >
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 >> > Warner
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 >>
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 >> Basically, you want to chan= ge the retry counts for certain ASC/ASCQ
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 >> codes only, on a site-by-si= te basis?=C2=A0 That sounds reasonable.=C2=A0 Would
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 >> it be configurable at runti= me or only at build time?
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 >
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 >
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 > I'd like to change the defa= ult actions. But maybe we just do that for
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0everyone and assume modern drives...<= br> >=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 >
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 >> Also, I've been thinkin= g lately that it would be real nice if READ
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 >> UNRECOVERABLE could be tran= slated to EINTEGRITY instead of EIO.=C2=A0 That
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 >> would let consumers know th= at retries are pointless, but that the data
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 >> is probably healable.
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 >
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 >
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 > Unlikely, unless you've tun= ed things to not try for long at recovery...
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 >
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 > But regardless... do you have a= concrete example of a use case?
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0There's a number of places that m= ap any error to EIO. And I'd like a use
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0case before we expand the errors the = lower layers return...
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 >
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 > Warner
>
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0My first use-case is a user-space FUS= E file system.=C2=A0 It only has
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0access to errnos, not ASC/ASCQ codes.= =C2=A0 If we do as I suggest, then it
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0could heal a READ UNRECOVERABLE by re= writing the sector, whereas other
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0EIO errors aren't likely to be he= aled that way.
>
>
>=C2=A0 =C2=A0 =C2=A0Yea... but READ UNRECOVERABLE is kinda hit or miss.= ..
>
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0My second use-case is ZFS.=C2=A0 zfsd= treats checksum errors differently
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0from I/O errors.=C2=A0 A checksum err= or normally means that a read returned
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0wrong data.=C2=A0 But I think that RE= AD UNRECOVERABLE should also count.
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0After all, that means that the disk&#= 39;s media returned wrong data which
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0was detected by the disk's own ED= C/ECC.=C2=A0 I've noticed that zfsd seems
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0to fault disks too eagerly when their= only problem is READ
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0UNRECOVERABLE errors.=C2=A0 Mapping i= t to EINTEGRITY, or even a new error
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0code, would let zfsd be tuned better.=
>
>
>=C2=A0 =C2=A0 =C2=A0EINTEGRITY would then mean two different things. UF= S returns in when
>=C2=A0 =C2=A0 =C2=A0checksums fail for critical=C2=A0filesystem errors.= I'm not saying no, per se,
>=C2=A0 =C2=A0 =C2=A0just that it conflates two different errors.
>
>=C2=A0 =C2=A0 =C2=A0I think both of these use cases would be better ser= ved by CAM's publishing
>=C2=A0 =C2=A0 =C2=A0of the errors to devctl today. Here's some exam= ple data from a system I'm
>=C2=A0 =C2=A0 =C2=A0looking at:
>
>=C2=A0 =C2=A0 =C2=A0system=3DCAM subsystem=3Dperiph type=3Dtimeout devi= ce=3Dda36 serial=3D"12345"
>=C2=A0 =C2=A0 =C2=A0cam_status=3D"0x44b" timeout=3D30000 CDB= =3D"28 00 4e b7 cb a3 00 04 cc 00 "
>=C2=A0 =C2=A0 =C2=A0 =C2=A0timestamp=3D1634739729.312068
>=C2=A0 =C2=A0 =C2=A0system=3DCAM subsystem=3Dperiph type=3Dtimeout devi= ce=3Dda36 serial=3D"12345"
>=C2=A0 =C2=A0 =C2=A0cam_status=3D"0x44b" timeout=3D30000 CDB= =3D"28 00 20 6b d5 56 00 00 c0 00 "
>=C2=A0 =C2=A0 =C2=A0 =C2=A0timestamp=3D1634739729.585541
>=C2=A0 =C2=A0 =C2=A0system=3DCAM subsystem=3Dperiph type=3Derror device= =3Dda36 serial=3D"12345"
>=C2=A0 =C2=A0 =C2=A0cam_status=3D"0x4cc" scsi_status=3D2 scsi= _sense=3D"72 03 11 00" CDB=3D"28 00 ad 1a
>=C2=A0 =C2=A0 =C2=A035 96 00 00 56 00 " timestamp=3D1641979267.469= 064
>=C2=A0 =C2=A0 =C2=A0system=3DCAM subsystem=3Dperiph type=3Derror device= =3Dda36 serial=3D"12345"
>=C2=A0 =C2=A0 =C2=A0cam_status=3D"0x4cc" scsi_status=3D2 scsi= _sense=3D"72 03 11 00" CDB=3D"28 00 ad 1a
>=C2=A0 =C2=A0 =C2=A035 96 00 01 5e 00 " =C2=A0timestamp=3D16422525= 39.693699
>=C2=A0 =C2=A0 =C2=A0system=3DCAM subsystem=3Dperiph type=3Derror device= =3Dda39 serial=3D"12346"
>=C2=A0 =C2=A0 =C2=A0cam_status=3D"0x4cc" scsi_status=3D2 scsi= _sense=3D"72 04 02 00" CDB=3D"2a 00 01 2b
>=C2=A0 =C2=A0 =C2=A0c8 f6 00 07 81 00 " =C2=A0timestamp=3D16696031= 44.090835
>
>=C2=A0 =C2=A0 =C2=A0Here we get the sense key, the asc and the ascq in = the scsi_sense data (I'm
>=C2=A0 =C2=A0 =C2=A0currently looking at expanding this to the entire s= ense buffer, since it
>=C2=A0 =C2=A0 =C2=A0includes how hard the drive tried to read the data = on media and hardware
>=C2=A0 =C2=A0 =C2=A0errors).=C2=A0 It doesn't include nvme data, bu= t does include ata data (I'll have
>=C2=A0 =C2=A0 =C2=A0to add that data, now that I've noticed it is m= issing).=C2=A0 With the sense data
>=C2=A0 =C2=A0 =C2=A0and the CDB you know what kind of error you got, pl= us what block didn't
>=C2=A0 =C2=A0 =C2=A0read/write correctly. With the extended sense data,= you can find out even
>=C2=A0 =C2=A0 =C2=A0more details that are sense-key dependent...
>
>=C2=A0 =C2=A0 =C2=A0So I'm unsure that trying to shoehorn our imper= fect knowledge of what's
>=C2=A0 =C2=A0 =C2=A0retriable, fixable, should be written with zeros in= to the kernel and
>=C2=A0 =C2=A0 =C2=A0converting that to a separate errno would give good= results, and tapping
>=C2=A0 =C2=A0 =C2=A0into this stream daemons that want to make more nua= nced calls about disks
>=C2=A0 =C2=A0 =C2=A0might be the better way to go. One of the things I&= #39;m planning for $WORK is
>=C2=A0 =C2=A0 =C2=A0to enable the retry time limit of one of the mode p= ages so that we fail
>=C2=A0 =C2=A0 =C2=A0faster and can just delete the file with the 'b= ad' block that we'd get
>=C2=A0 =C2=A0 =C2=A0eventually if we allowed the full, default error pr= ocessing to run, but that
>=C2=A0 =C2=A0 =C2=A0'slow path' processing kills performance fo= r all other users of the
>=C2=A0 =C2=A0 =C2=A0drive...=C2=A0 I'm unsure how well that will wo= rk out (and I know I'm lucky that
>=C2=A0 =C2=A0 =C2=A0I can always recover any data for my application si= nce it's just a cache).
>
>=C2=A0 =C2=A0 =C2=A0I'd be interested to hear what others have to s= ay here thought, since my
>=C2=A0 =C2=A0 =C2=A0focus on this data is through the lense of my rathe= r specialized application...
>
>=C2=A0 =C2=A0 =C2=A0Warner
>
>=C2=A0 =C2=A0 =C2=A0P.S. That was generated with this rule if you wante= d to play with it...
>=C2=A0 =C2=A0 =C2=A0You'd have to translate absolute disk blocks to= a partition and an offset
>=C2=A0 =C2=A0 =C2=A0into the filesystem, then give the filesystem a cha= nce to tell you what of
>=C2=A0 =C2=A0 =C2=A0its data/metadata that block is used for...
>
>=C2=A0 =C2=A0 =C2=A0# Disk errors
>=C2=A0 =C2=A0 =C2=A0notify 10 {
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 match "system&quo= t; =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0"CAM";
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 match "subsystem&= quot; =C2=A0 =C2=A0 =C2=A0 "periph";
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 match "device&quo= t; =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0"[an]?da[0-9]+";
>=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 action "logger -t= diskerr -p daemon.info <http://daemon= .info> $_
>=C2=A0 =C2=A0 =C2=A0timestamp=3D$timestamp";
>=C2=A0 =C2=A0 =C2=A0};
>

--000000000000fa60eb0600f6d3dc--