RE: async vs. sync

RE: async vs. sync

[Lever, Charles]

is the effect of the "sync" export option limited to NFSv3 COMMIT, or is
it limited to both NFSv3 COMMIT and NFSv3 FILE_SYNC/DATA_SYNC WRITE?

what are the effects on NFSv4 writes and commits?

> -----Original Message-----
> From: Olaf Kirch [mailto:okir@suse.de]
> Sent: Wednesday, July 28, 2004 4:57 AM
> To: Bernd Schubert
> Cc: nfs@lists.sourceforge.net
> Subject: Re: [NFS] async vs. sync
>=20
>=20
> Hi,
>=20
> the way the sync export option affects NFSv3 writes is
> limited to COMMITs, so if you see a slow-down here it must be=20
> bottle-necking in that part of the code.


-------------------------------------------------------
This SF.Net email is sponsored by: InterSystems CACHE
FREE OODBMS DOWNLOAD - A multidimensional database that combines
robust object and relational technologies, making it a perfect match
for Java, C++,COM, XML, ODBC and JDBC. www.intersystems.com/match8
_______________________________________________
NFS maillist  -  NFS@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nfs

RE: async vs. sync

[Trond Myklebust]

ty den 16.11.2004 Klokka 08:15 (-0800) skreiv Lever, Charles:
> is the effect of the "sync" export option limited to NFSv3 COMMIT, or is
> it limited to both NFSv3 COMMIT and NFSv3 FILE_SYNC/DATA_SYNC WRITE?
> 
> what are the effects on NFSv4 writes and commits?

No! It is clearly not just limited to writes and commits.

Look at the code in fs/nfs/vfs.c: there are EX_ISSYNC() exceptions that
wrap calls to nfsd_sync_dir() in nfsd*_create(), nfsd_symlink(),
nfsd_link(), nfsd_rename(), and nfsd_unlink().

Cheers,
  Trond

-- 
Trond Myklebust <trond.myklebust@fys.uio.no>



-------------------------------------------------------
This SF.Net email is sponsored by: InterSystems CACHE
FREE OODBMS DOWNLOAD - A multidimensional database that combines
robust object and relational technologies, making it a perfect match
for Java, C++,COM, XML, ODBC and JDBC. www.intersystems.com/match8
_______________________________________________
NFS maillist  -  NFS@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nfs

Re: async vs. sync

[jehan.procaccia]

Trond Myklebust wrote:

>ty den 16.11.2004 Klokka 08:15 (-0800) skreiv Lever, Charles:
>  
>
>>is the effect of the "sync" export option limited to NFSv3 COMMIT, or is
>>it limited to both NFSv3 COMMIT and NFSv3 FILE_SYNC/DATA_SYNC WRITE?
>>
>>what are the effects on NFSv4 writes and commits?
>>    
>>
>
>No! It is clearly not just limited to writes and commits.
>
>Look at the code in fs/nfs/vfs.c: there are EX_ISSYNC() exceptions that
>wrap calls to nfsd_sync_dir() in nfsd*_create(), nfsd_symlink(),
>nfsd_link(), nfsd_rename(), and nfsd_unlink().
>
>Cheers,
>  Trond
>
>  
>
By the way, I noticed on a performance factor of 30 (!) from an sync to 
async nfs v3 export FS :

sync export:

$time tar xvfz linux-2.6.8.tar.gz
real    64m18.618s
user    0m5.742s
sys     0m15.658s

async export:

$time tar xvfz linux-2.6.8.tar.gz
real    2m0.552s
user    0m5.838s
sys     0m15.678s


Is it really dangerous to use async ? why recent OS uses sync by default 
(My client is a Fedora Core 2 (kernel 2.6)  and server an RedHat 
Entreprise server 3 (kernel 2.4))

thanks.


-------------------------------------------------------
This SF.Net email is sponsored by: InterSystems CACHE
FREE OODBMS DOWNLOAD - A multidimensional database that combines
robust object and relational technologies, making it a perfect match
for Java, C++,COM, XML, ODBC and JDBC. www.intersystems.com/match8
_______________________________________________
NFS maillist  -  NFS@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nfs

Re: async vs. sync

[Trond Myklebust]

ty den 16.11.2004 Klokka 18:18 (+0100) skreiv jehan.procaccia:
> >
> By the way, I noticed on a performance factor of 30 (!) from an sync to 
> async nfs v3 export FS :
> 
> sync export:
> 
> $time tar xvfz linux-2.6.8.tar.gz
> real    64m18.618s
> user    0m5.742s
> sys     0m15.658s
> 
> async export:
> 
> $time tar xvfz linux-2.6.8.tar.gz
> real    2m0.552s
> user    0m5.838s
> sys     0m15.678s

What mount options are you using? I don't see a factor 30 difference
using my setup.

> Is it really dangerous to use async ? why recent OS uses sync by default 
> (My client is a Fedora Core 2 (kernel 2.6)  and server an RedHat 
> Entreprise server 3 (kernel 2.4))

"async" is bad because it lies to you about whether or not the data is
on disk or not. Type "sync", and it will happily return, and tell your
application that all is well but you will still lose your data if the
server crashes on you...

Cheers,
  Trond

-- 
Trond Myklebust <trond.myklebust@fys.uio.no>



-------------------------------------------------------
This SF.Net email is sponsored by: InterSystems CACHE
FREE OODBMS DOWNLOAD - A multidimensional database that combines
robust object and relational technologies, making it a perfect match
for Java, C++,COM, XML, ODBC and JDBC. www.intersystems.com/match8
_______________________________________________
NFS maillist  -  NFS@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nfs

Problem with NFS-Exports

[Oliver Beowulf Friedrich]

Greetings @ all,

I'm new here on the list, but already have a simple question to you.

I'm using Gentoo linux, currently, having a working server@home with 
Gentoo which exports Gentoo's temporary folders out to my workstations.

server root # cat /etc/exports
/home           192.168.0.0/24(rw,async)
/usr/portage    192.168.0.0/24(rw,no_root_squash,async,mp)
/tmp 
192.168.0.0/24(rw,no_root_squash,insecure_locks,async,mp)
/mnt/data       192.168.0.0/24(rw,no_root_squash,mp,sync)
/mnt/backup     192.168.0.0/24(rw,no_root_squash,mp,sync)


One Workstation is now on compiling a new fresh Gentoo Linux, so I 
startet with booting from Gentoo-2004.3 liveCD, mounting local 
filesystems, mounting nfs filesystems, chrooting and starting to 
compile.

So fine, but while compiling, I get the following Error messages 
printed out quit often:

lockd: failed to monitor 192.168.0.2
lockd: cannont monitor 192.168.0.2

So whats going on there? Sometimes it's going so far, that my new 
WKS cannot compile, cause of:

lockd: server ist not responding
lockd: still trying OK

How do i get this solved? With my previus installation it worked 
fine, I had only NFS-Server Support compiled into Kernel, now I have 
NFSV3-Server Support compiled in.

My Server now is not used to heavy, only a little Samba Server for 
my (only) Windows-Box where MP3's are shared to...

On my last installation on my Server I used distcc while compiling, 
and even that did not lead to such Error Messages...

Thanks for reading

BeowulfOF




-------------------------------------------------------
This SF.Net email is sponsored by: InterSystems CACHE
FREE OODBMS DOWNLOAD - A multidimensional database that combines
robust object and relational technologies, making it a perfect match
for Java, C++,COM, XML, ODBC and JDBC. www.intersystems.com/match8
_______________________________________________
NFS maillist  -  NFS@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/nfs