Date: Wed, 06 Apr 2022 16:26:53 +0200
From: egoitz@ramattack.net
To: Eugene Grosbein <eugen@grosbein.net>
Cc: freebsd-hackers@freebsd.org
Subject: Re: {* 05.00 *}Re: Desperate with 870 QVO and ZFS
Message-ID: <15ce19eff500edfead2b56ea38cace7b@ramattack.net>
In-Reply-To: <3b0451a3-01ec-9f30-4eb4-4cbcd91f78cf@grosbein.net>
References: <6cf6c03c5a4aa8128575ec4e2f70b168@ramattack.net> <15a86fae-90fd-951d-50e0-48f9be8b4bbc@grosbein.net> <109127fb4e43e70cd548fecde2c1f755@ramattack.net> <3b0451a3-01ec-9f30-4eb4-4cbcd91f78cf@grosbein.net>
next in thread | previous in thread | raw e-mail | index | archive | help
--=_03a51938caf864ccc70ca989be3d2b44 Content-Transfer-Encoding: 8bit Content-Type: text/plain; charset=UTF-8 Hi Eugene, Thank you so much really! As those values seem to be related to basically the average service time in the disks and the average service times increases in crisis moments... I assume yes it does... I will obviously keep an eye on them... sure... Thanks again! Regards, El 2022-04-06 14:23, Eugene Grosbein escribió: > 06.04.2022 19:02, egoitz@ramattack.net wrote: > >> Hi Eugene, >> >> No... I normally don't have many delete operations..... in fact the bast majority of them are left for the night... they are done at 2,3,4 am in the morning.... >> >> We may have 600 deletes/sec at busy times (acording to what I see in gstat and calculating when having two masters).... > > As I've said, you need to monitor ALL values, there no unimportant ones on that screen. > Look also at ms/r, ms/w and ms/d - that is, milliseconds per read operation, per write and per delete. > Do these times grow at peak time? --=_03a51938caf864ccc70ca989be3d2b44 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>Hi Eugene,</p> <p><br /></p> <p>Thank you so much really!</p> <p><br /></p> <p>As those values seem to be related to basically the average service time= in the disks and the average service times increases in crisis moments..= =2E I assume yes it does...</p> <p><br /></p> <p>I will obviously keep an eye on them... sure...</p> <p><br /></p> <p>Thanks again!</p> <p><br /></p> <p>Regards,</p> <p><br /></p> <div> </div> <p><br /></p> <p>El 2022-04-06 14:23, Eugene Grosbein escribió:</p> <blockquote type=3D"cite" style=3D"padding: 0 0.4em; border-left: #1010ff 2= px solid; margin: 0"> <div class=3D"pre" style=3D"margin: 0; padding: 0; font-family: monospace">= <br /> 06.04.2022 19:02, <a href=3D"mailto:egoitz@ramattack.net">egoitz@ram= attack.net</a> wrote: <blockquote type=3D"cite" style=3D"padding: 0 0.4em; border-left: #1010ff 2= px solid; margin: 0">Hi Eugene,<br /> <br /> <br /> No... I normally don't = have many delete operations..... in fact the bast majority of them are left= for the night... they are done at 2,3,4 am in the morning....<br /> <br />= We may have 600 deletes/sec at busy times (acording to what I see in gstat= and calculating when having two masters)....</blockquote> <br /> As I've said, you need to monitor ALL values, there no unimportant o= nes on that screen.<br /> Look also at ms/r, ms/w and ms/d - that is, milli= seconds per read operation, per write and per delete.<br /> Do these times = grow at peak time?<br /> <br /> </div> </blockquote> </body></html> --=_03a51938caf864ccc70ca989be3d2b44--
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?15ce19eff500edfead2b56ea38cace7b>