Date: Fri, 27 Mar 2020 22:32:31 -0500 From: Karl Denninger <karl@denninger.net> To: freebsd-questions@freebsd.org Subject: Re: drive selection for disk arrays Message-ID: <753e87e0-8313-faaa-523e-f630a80cb8a4@denninger.net> In-Reply-To: <ad2d65c8-b3ef-de46-42b6-102794c33a9d@holgerdanske.com> References: <20200325081814.GK35528@mithril.foucry.net> <713db821-8f69-b41a-75b7-a412a0824c43@holgerdanske.com> <20200326124648725158537@bob.proulx.com> <alpine.BSF.2.21.9999.2003261630030.47777@mail2.nber.org> <20200327104555.1d6d7cd9.freebsd@edvax.de> <1bcd7aa2-31e5-91f1-5151-926c9d16e16e@holgerdanske.com> <8e74482f-b951-ee97-50b8-04ea1f0d46a3@denninger.net> <ad2d65c8-b3ef-de46-42b6-102794c33a9d@holgerdanske.com>
next in thread | previous in thread | raw e-mail | index | archive | help
[-- Attachment #1 --] On 3/27/2020 20:46, David Christensen wrote: > On 2020-03-27 17:45, Karl Denninger wrote: >> >> On 3/27/2020 19:39, David Christensen wrote: >>> On 2020-03-27 02:45, Polytropon wrote: >>> >>>> When a drive _reports_ bad sectors, at least in the past >>>> it was an indication that it already _has_ lots of them. >>>> The drive's firmware will remap bad sectors to spare >>>> sectors, so "no error" so far. >>> >>> If a drive detects an error, my guess is that it will report the error >>> to the OS; regardless of the outcome of a particular I/O operation >>> (data read, data written, data lost) or internal actions taken (block >>> marked bad, block remapped, etc.). It is then up to the OS to decide >>> what to do next. RAID and/or ZFS offer the means for shielding the >>> application from I/O and drive failures. >>> >> Yes, but... >> >> Those drives that can do "SMART" will report (if you have a patrol >> daemon for it running) if they do a "silent" sector reassignment. >> Otherwise the OS is none the wiser and neither is ZFS (or anything >> else.) > > I guess I need to RTFM: > > https://www.intel.com/content/dam/www/public/us/en/documents/technical-specifications/serial-ata-ahci-spec-rev1-3-1.pdf > > > >> Needless to say if reassignments increase you might want to >> think about swapping the drive *before* it blows up! > > Agreed. > > >> I have the daemon running on all my machines. It works nicely and has >> warned me a few times over the years. With that said it doesn't ALWAYS >> catch a drive before it pukes. > > Are you referring to periodic, smartd, or something else? > > Smartd. -- Karl Denninger karl@denninger.net <mailto:karl@denninger.net> /The Market Ticker/ /[S/MIME encrypted email preferred]/ [-- Attachment #2 --] 0 *H 010 `He 0 *H 00 H^Ōc!5 H0 *H 01 0 UUS10U Florida10U Niceville10U Cuda Systems LLC10U Cuda Systems CA1!0 U Cuda Systems LLC 2017 CA0 170817164217Z 270815164217Z0{1 0 UUS10U Florida10U Cuda Systems LLC10U Cuda Systems CA1%0#U Cuda Systems LLC 2017 Int CA0"0 *H 0 h-5B>[;olӴ0~͎O9}9Ye*$ g !ukvʶLzN`jL>MD'7U 45CB+kY`bd~b*c3Ny-78ju]9H euέsӬDؽmgwER?&UURj'}9nWD i`XcbGz \gG=u%\Oi13ߝ4 K44pYQr]Ie/r0+eEޝݖ0C15Mݚ@JSZ(zȏ NTa(25DD5.l<g[[Z arQQ%Buȴ~~`IohRbʳڟu2MS8EdFUClCMaѳ !}ș+2k/bųE,n当ꖛ\(8 WV8 d]b yXw ܊:I39 00 U ]^§Q\ӎ0U #0T039N0b01 0 UUS10U Florida10U Niceville10U Cuda Systems LLC10U Cuda Systems CA1!0 U Cuda Systems LLC 2017 CA @Ui0U 0 0U 0 *H :P U!>vJ n io -#ן]Wyu jǑR̀Q nƇ!G ѦF g\ yLx gw=OPycehf[ }ܷ['4ڝ\[p 6\o.B&JF" ZC{;*o*mcCcLY߾` t*S!(` ]DHP5A~/NPp6=mhk밣'doA$86hm5ӚS@ j ެEgl )0JG`%k35PaC?σ ׳HEt}!P㏏%*BxbQwaKG$6 h¦Mve; [o-Iی& I,Tc ߎ#t wPA@l0P+KXBպT zGv;NcI3&J ĬUPNa?/%W6G۟N000 k#Xd\ =0 *H 0{1 0 UUS10U Florida10U Cuda Systems LLC10U Cuda Systems CA1%0#U Cuda Systems LLC 2017 Int CA0 170817212120Z 220816212120Z0W1 0 UUS10U Florida10U Cuda Systems LLC10U karl@denninger.net0"0 *H 0 T[I-ΆϏ dn;Å@שy.us~_ZG%<MYd\gvfnsa1'6Egyjs"C [{~ _K Pn+<*pv#Q+H/7[-vqDV^U>f%GX)H.|l`M(Cr >е͇6#odc"YljҦln8@5SA0&ۖ"OGj?UDWZ5 dDB 7k-)9Izs-JAv J 6L$Ն1SmY.Lqw*SH;EF'DĦH]MOgQQ|Mٙג2Z9y@ y]}6ٽeY9Y2xˆ$T=e CǺǵbn֛{j |@LLt1[Dk5:$= ` M 00<+00.0,+0 http://ocsp.cudasystems.net:88880 U 0 0 `HB0U 0 U %0++03 `HB &$OpenSSL Generated Client Certificate0 U %՞V =;bzQ0U #0]^§Q\ӎϡ01 0 UUS10U Florida10U Niceville10U Cuda Systems LLC10U Cuda Systems CA1!0 U Cuda Systems LLC 2017 CA H^Ōc!5 H0 U 0karl@denninger.net0 *H ۠A0-j%--$% g2#ޡ1^>{K+uGE v1ş7Af&b&O; .;A5* U)ND2bF|\= ]<sˋL!wrw٧> YMÄ3\mWR hSv!_zvl? 3_ xU%\^#O*Gk̍YI_&Fꊛ@&1n } ͬ:{hTP3B.;bU8:Z=^Gw8 !k-@xE@i,+'Iᐚ:f hztX7/(hY` O.1}a`%RW^akǂpCAufgDix UTЩ/7}%=jnVZvcF<M= 2^GKH5魉 _O4ެByʈySkw=5@h.0z> W1000{1 0 UUS10U Florida10U Cuda Systems LLC10U Cuda Systems CA1%0#U Cuda Systems LLC 2017 Int CA k#Xd\ =0 `He E0 *H 1 *H 0 *H 1 200328033231Z0O *H 1B@:]of ?3{ m.2$ RmpH"o[SmQb'B>mXZB0l *H 1_0]0 `He*0 `He0 *H 0*H 0 *H @0+0 *H (0 +7100{1 0 UUS10U Florida10U Cuda Systems LLC10U Cuda Systems CA1%0#U Cuda Systems LLC 2017 Int CA k#Xd\ =0 *H 10{1 0 UUS10U Florida10U Cuda Systems LLC10U Cuda Systems CA1%0#U Cuda Systems LLC 2017 Int CA k#Xd\ =0 *H tp1V>i=(y\@_٧ R,N . }{ i`! n<a2 }NOP$vµ d3䈖z\*ʱcl= t+E*wln1xȤ>2M?*P*/O$jr4I(@Iln4Thqs{gU>WVR< fvI1È+xag#|6 "{Nb(C^QJ~193(v EC}|9 S{rY3 lqOQЯ VnDž:ipLq?ADjkNGyD1Nv>z9z6MC9GA} ܂ 02y"qjM.'&pFɓ5^6T 8}JN$?);]eE}g C &Azܴ V«Y^bD1{
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?753e87e0-8313-faaa-523e-f630a80cb8a4>