Skip site navigation (1)Skip section navigation (2)
Date:      Sat, 9 Oct 2021 16:09:52 +0000
From:      Rick Macklem <rmacklem@uoguelph.ca>
To:        Felix Palmen <felix@palmen-it.de>, "freebsd-current@freebsd.org" <freebsd-current@freebsd.org>
Subject:   Re: Writing large build logs to NFS extremely slow?
Message-ID:  <YQXPR0101MB0968CA670A7AD47437D64E5FDDB39@YQXPR0101MB0968.CANPRD01.PROD.OUTLOOK.COM>
In-Reply-To: <20211009155033.fckg4i7zqo4yt2te@nexus.home.palmen-it.de>
References:  <20211007021643.bwglyvrswk2nm3fl@nexus.home.palmen-it.de> <CAOgwaMsLTWaMddx1HPY0eiY4t-12x-bEiZmhUpFsuEqkNhM27w@mail.gmail.com> <20211009141852.3cmjh7tysnehum7b@nexus.home.palmen-it.de> <YQXPR0101MB096832C878E88DF877B39DB8DDB39@YQXPR0101MB0968.CANPRD01.PROD.OUTLOOK.COM> <20211009155033.fckg4i7zqo4yt2te@nexus.home.palmen-it.de>

next in thread | previous in thread | raw e-mail | index | archive | help
Felix Palmen wrote:=0A=
>* Rick Macklem <rmacklem@uoguelph.ca> [20211009 15:35]:=0A=
>> Felix Palmen wrote:=0A=
>> Assuming your NFS performance is acceptable for other things and it=0A=
>> is only this log file that is a problem, then I doubt there is much you=
=0A=
>> can do to improve it.=0A=
>=0A=
>Yes, that's the only problem I found so far=85=0A=
Another thing you could try is turning off synchronous writing at the NFS s=
erver.=0A=
# sysctl vfs.nfsd.async=3D1=0A=
and "sync=3Ddisabled" for ZFS if it is a ZFS file system that is exported.=
=0A=
=0A=
However, BE FORWARNED:=0A=
Doing so violates the NFS RFCs because it can result is data loss/corrupted=
 files=0A=
that are being written when the NFS server crashes/reboots.=0A=
Only do it if the data being written to the NFS server is not particularly =
important stuff.=0A=
=0A=
rick=0A=
=0A=
> --> Append (as in O_APPEND opens) are a poor case for NFS, since there=0A=
>       is no append write in NFS. To approximate append write, it must flu=
sh=0A=
>       all dirty data to the server, do a Getattr to find out the file's c=
urrent=0A=
>       size and then do the write (over and over and over again).=0A=
=0A=
Ok, thanks for the insight here! So maybe, I could have a look at=0A=
poudriere's code? If I understand you correctly, keeping the file open=0A=
and just issuing more write() calls wouldn't trigger that problem?=0A=
=0A=
> You could try the "nocto" mount option, which might help, if the log file=
=0A=
> is being re-opened many times, but I doubt it will help.=0A=
=0A=
Well, I take any suggestion, so will try that as well and report back,=0A=
thanks again!=0A=
=0A=
Side note, please don't CC/To me, I'm subscribed the list ;)=0A=
=0A=
--=0A=
 Dipl.-Inform. Felix Palmen  <felix@palmen-it.de>   ,.//..........=0A=
 {web}  http://palmen-it.de  {jabber} [see email]   ,//palmen-it.de=0A=
 {pgp public key}     http://palmen-it.de/pub.txt   //   """""""""""=0A=
 {pgp fingerprint} A891 3D55 5F2E 3A74 3965 B997 3EF2 8B0A BC02 DA2A=0A=



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?YQXPR0101MB0968CA670A7AD47437D64E5FDDB39>