Date: Thu, 07 Apr 2022 10:49:15 +0200
From: egoitz@ramattack.net
To: Eugene Grosbein <eugen@grosbein.net>
Cc: freebsd-fs@freebsd.org, freebsd-hackers@freebsd.org, Freebsd performance <freebsd-performance@freebsd.org>
Subject: Re: {* 05.00 *}Re: Re: Desperate with 870 QVO and ZFS
Message-ID: <55000f00fb64510e8ef6b8ad858d8855@ramattack.net>
In-Reply-To: <cdad7375-21c9-5fc7-1562-b10f2ff8713a@grosbein.net>
References: <4e98275152e23141eae40dbe7ba5571f@ramattack.net> <665236B1-8F61-4B0E-BD9B-7B501B8BD617@ultra-secure.de> <0ef282aee34b441f1991334e2edbcaec@ramattack.net> <28e11d7ec0ac5dbea45f9f271fc28f06@ramattack.net> <ca3f86f2-94a1-be94-ad55-7bd1c9bc50ab@grosbein.net> <7aa95cb4bf1fd38b3fce93bc26826042@ramattack.net> <cdad7375-21c9-5fc7-1562-b10f2ff8713a@grosbein.net>
next in thread | previous in thread | raw e-mail | index | archive | help
--=_85ed0b4a49488ec1a940cb1f7fed0376 Content-Transfer-Encoding: 8bit Content-Type: text/plain; charset=UTF-8 Good morning Eugene!! Thank you so much for your help mate :) :) really :) :) Ok I take good notes of all you have replied me below :) :) Very very thankful for your help really :) Cheers, El 2022-04-06 20:10, Eugene Grosbein escribió: > ATENCION > ATENCION > ATENCION!!! Este correo se ha enviado desde fuera de la organizacion. No pinche en los enlaces ni abra los adjuntos a no ser que reconozca el remitente y sepa que el contenido es seguro. > > 06.04.2022 23:51, egoitz@ramattack.net wrote: > >> About your recommendations... Eugene, if some of them wouldn't be working as expected, >> could we revert some or all of them > > Yes, it all can be reverted. > Just write down original sysctl values if you are going to change it. > >> 1) Make sure the pool has enough free space because ZFS can became crawling slow otherwise. >> >> *This is just an example... but you can see all similarly....* >> >> *zpool list* >> *NAME SIZE ALLOC FREE CKPOINT EXPANDSZ FRAG CAP DEDUP HEALTH ALTROOT* >> *zroot 448G 2.27G 446G - - 1% 0% 1.00x ONLINE -* >> *mail_dataset 58.2T 19.4T 38.8T - - 32% 33% 1.00x ONLINE -* > > It's all right. > >> 2) Increase recordsize upto 1MB for file systems located in the pool >> so ZFS is allowed to use bigger request sizes for read/write operations >> >> *We have the default... so 128K...* > > It will not hurt increasing it upto 1MB. > >> 5) If you have good power supply and stable (non-crashing) OS, try increasing >> sysctl vfs.zfs.txg.timeout from defaule 5sec, but do not be extreme (f.e. upto 10sec). >> Maybe it will increase amount of long writes and decrease amount of short writes, that is good. >> >> *Well I have sync in disabled in the datasets... do you still think it's good to change it? > > Yes, try it. Disabling sync makes sense if you have lots of fsync() operations > but other small writes are not affected unless you raise vfs.zfs.txg.timeout > >> *What about the vfs.zfs.dirty_data_max and the vfs.zfs.dirty_data_max_max, would you increase them from 4GB it's set now?.* > > Never tried that and cannot tell. --=_85ed0b4a49488ec1a940cb1f7fed0376 Content-Transfer-Encoding: quoted-printable Content-Type: text/html; charset=UTF-8 <html><head><meta http-equiv=3D"Content-Type" content=3D"text/html; charset= =3DUTF-8" /></head><body style=3D'font-size: 10pt; font-family: Verdana,Gen= eva,sans-serif'> <p>Good morning Eugene!!</p> <p><br /></p> <p>Thank you so much for your help mate :) :) really :) :)</p> <p><br /></p> <p>Ok I take good notes of all you have replied me below :) :)</p> <p><br /></p> <p>Very very thankful for your help really :)</p> <p><br /></p> <p>Cheers,</p> <div> </div> <p><br /></p> <p>El 2022-04-06 20:10, Eugene Grosbein escribió:</p> <blockquote type=3D"cite" style=3D"padding: 0 0.4em; border-left: #1010ff 2= px solid; margin: 0"><!-- html ignored --><!-- head ignored --><!-- meta ig= nored --> <div class=3D"pre" style=3D"margin: 0; padding: 0; font-family: monospace">= ATENCION<br /> ATENCION<br /> ATENCION!!! Este correo se ha enviado desde f= uera de la organizacion. No pinche en los enlaces ni abra los adjuntos a no= ser que reconozca el remitente y sepa que el contenido es seguro.<br /> <b= r /> 06.04.2022 23:51, <a href=3D"mailto:egoitz@ramattack.net">egoitz@ramat= tack.net</a> wrote:<br /> <br /> <blockquote type=3D"cite" style=3D"padding: 0 0.4em; border-left: #1010ff 2= px solid; margin: 0">About your recommendations... Eugene, if some of them = wouldn't be working as expected,<br /> could we revert some or all of them<= /blockquote> <br /> Yes, it all can be reverted.<br /> Just write down original sysctl v= alues if you are going to change it.<br /> <br /> <blockquote type=3D"cite" style=3D"padding: 0 0.4em; border-left: #1010ff 2= px solid; margin: 0">1) Make sure the pool has enough free space because ZF= S can became crawling slow otherwise.<br /> <br /> *This is just an e= xample... but you can see all similarly....*<br /> <br /> *zpool list= *<br /> *NAME &= nbsp; SIZE ALLOC FREE CKPOINT EXPANDSZ &= nbsp; FRAG CAP DEDUP HEALTH ALTROO= T*<br /> *zroot  = ; 448G 2.27G 446G &nbs= p; - - &nb= sp; 1% 0% 1.00x ONLINE  = ;-*<br /> *mail_dataset 58.2T 19.4T 38.8T &nb= sp; - &nbs= p;- 32% 33% 1.00x ONLINE &n= bsp;-*</blockquote> <br /> It's all right.<br /> <br /> <blockquote type=3D"cite" style=3D"padding: 0 0.4em; border-left: #1010ff 2= px solid; margin: 0">2) Increase recordsize upto 1MB for file systems locat= ed in the pool<br /> so ZFS is allowed to use bigger request sizes for read= /write operations<br /> <br /> *We have the default... so 128K...*</b= lockquote> <br /> It will not hurt increasing it upto 1MB.<br /> <br /> <blockquote type=3D"cite" style=3D"padding: 0 0.4em; border-left: #1010ff 2= px solid; margin: 0">5) If you have good power supply and stable (non-crash= ing) OS, try increasing<br /> sysctl vfs.zfs.txg.timeout from defaule 5sec,= but do not be extreme (f.e. upto 10sec).<br /> Maybe it will increase amou= nt of long writes and decrease amount of short writes, that is good.<br /> = <br /> *Well I have sync in disabled in the datasets... do you still = think it's good to change it?</blockquote> <br /> Yes, try it. Disabling sync makes sense if you have lots of fsync() = operations<br /> but other small writes are not affected unless you raise v= fs.zfs.txg.timeout<br /> <br /> <blockquote type=3D"cite" style=3D"padding: 0 0.4em; border-left: #1010ff 2= px solid; margin: 0">*What about the vfs.zfs.dirty_data_max and the vfs.zfs= =2Edirty_data_max_max, would you increase them from 4GB it's set now?.*</bl= ockquote> <br /> Never tried that and cannot tell.<br /> </div> </blockquote> </body></html> --=_85ed0b4a49488ec1a940cb1f7fed0376--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?55000f00fb64510e8ef6b8ad858d8855>