Re: NFS on 10G interfaces still painfully slow

2016-08-02 Thread Gerrit Kühn
On Wed, 3 Aug 2016 07:59:54 +0200 Gerrit Kühn wrote about Re: NFS on 10G interfaces still painfully slow: GK> 1048576000 bytes transferred in 1.403620 secs (747051194 bytes/sec) GK> 1048576000 bytes transferred in 1.380546 secs (759537249 bytes/sec) Argh! ;-) Obviously, I cannot read and misco

Re: Unstable local network throughput

2016-08-02 Thread Ben RUBSON
> On 03 Aug 2016, at 04:32, Eugene Grosbein wrote: > > If you have gateway_enable="YES" (sysctl net.inet.ip.forwarding=1) > then try to disable this forwarding setting and rerun your tests to compare > results. Thank you Eugene for this, but net.inet.ip.forwarding is disabled by default and I

Re: NFS on 10G interfaces still painfully slow

2016-08-02 Thread Gerrit Kühn
On Tue, 2 Aug 2016 08:45:07 -0600 Alan Somers wrote about Re: NFS on 10G interfaces still painfully slow: > > Is there anyone around here who can confirm that nfs can go faster > > over 10G links? > > Any hints for further tuning/debugging are greatly appreciated. AS> I can get 1GB/s over NFS on

Re: Unstable local network throughput

2016-08-02 Thread Eugene Grosbein
03.08.2016 1:43, Ben RUBSON пишет: Hello, I'm trying to reach the 40Gb/s max throughtput between 2 hosts running a ConnectX-3 Mellanox network adapter. If you have gateway_enable="YES" (sysctl net.inet.ip.forwarding=1) then try to disable this forwarding setting and rerun your tests to compar

Re: Is NFSv4 mount interruptible?

2016-08-02 Thread Slawa Olhovchenkov
On Wed, Aug 03, 2016 at 01:55:02AM +0300, Slawa Olhovchenkov wrote: > On Tue, Aug 02, 2016 at 10:48:52PM +, Rick Macklem wrote: > > > Slawa wrote: > > >I have autofs NFSv4 mount with /etc/nfs.map: > > > > > >/NFS-nfsv4,intr,soft,sec=krb5i,gssname=host storage01:/ > > > > > >Currently I am

[Bug 211386] Kernel build fails when TCP_RFC7413 option set

2016-08-02 Thread bugzilla-noreply
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=211386 Ngie Cooper changed: What|Removed |Added Severity|Affects Only Me |Affects Some People Summa

Re: NFS on 10G interfaces still painfully slow

2016-08-02 Thread Rick Macklem
Alan Somers wrote: >On Tue, Aug 2, 2016 at 2:49 AM, Gerrit Kühn wrote: >> Hi all, >> >> I already reported this issue here a year ago and unfortunately was not >> able to fix it back then. Now I had another run at it, using two recent >> 10.3-machines with a direct 10G link. I still see nfs is pa

Re: Is NFSv4 mount interruptible?

2016-08-02 Thread Rick Macklem
Slawa wrote: >I have autofs NFSv4 mount with /etc/nfs.map: > >/NFS-nfsv4,intr,soft,sec=krb5i,gssname=host storage01:/ > >Currently I am see NFS hung and I am can't kill shell and can't >unmount /NFS. > >Is NFSv4 mount interruptible? >Or intr support only by NFSv3? Well, in the BUGS section of "

Re: Is NFSv4 mount interruptible?

2016-08-02 Thread Slawa Olhovchenkov
On Tue, Aug 02, 2016 at 10:48:52PM +, Rick Macklem wrote: > Slawa wrote: > >I have autofs NFSv4 mount with /etc/nfs.map: > > > >/NFS-nfsv4,intr,soft,sec=krb5i,gssname=host storage01:/ > > > >Currently I am see NFS hung and I am can't kill shell and can't > >unmount /NFS. > > > >Is NFSv4 mo

Is NFSv4 mount interruptible?

2016-08-02 Thread Slawa Olhovchenkov
I have autofs NFSv4 mount with /etc/nfs.map: /NFS-nfsv4,intr,soft,sec=krb5i,gssname=host storage01:/ Currently I am see NFS hung and I am can't kill shell and can't unmount /NFS. Is NFSv4 mount interruptible? Or intr support only by NFSv3? ___ free

[Bug 211386] Kernel build fails when TCP_RFC7413 option set [regression]

2016-08-02 Thread bugzilla-noreply
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=211386 Sergey Kandaurov changed: What|Removed |Added CC||pluk...@freebsd.org --- Comment

[Bug 211386] Kernel build fails when TCP_RFC7413 option set [regression]

2016-08-02 Thread bugzilla-noreply
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=211386 Ngie Cooper changed: What|Removed |Added Summary|Kernel build failes when|Kernel build fails when

[Bug 211441] incorrect handling of TCPS_SYN_SENT and TCPS_SYN_RECEIVED in API tcp_usrclosed() in file tcp_usrreq.c

2016-08-02 Thread bugzilla-noreply
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=211441 Mark Linimon changed: What|Removed |Added Assignee|freebsd-standards@FreeBSD.o |freebsd-net@FreeBSD.org

[Bug 211386] Kernel build failes when TCP_RFC7413 option set [regression]

2016-08-02 Thread bugzilla-noreply
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=211386 Mark Linimon changed: What|Removed |Added Assignee|freebsd-b...@freebsd.org|freebsd-net@FreeBSD.org K

[Bug 211219] NIC status does not pass into a state of "no carrier" after disconnecting the cable.

2016-08-02 Thread bugzilla-noreply
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=211219 Mark Linimon changed: What|Removed |Added Assignee|freebsd-b...@freebsd.org|freebsd-net@FreeBSD.org -- You are

[Bug 211062] [ixv] sr-iov virtual function driver fails to attach

2016-08-02 Thread bugzilla-noreply
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=211062 Mark Linimon changed: What|Removed |Added Keywords||IntelNetworking Assignee|

Re: Unstable local network throughput

2016-08-02 Thread Ben RUBSON
> On 02 Aug 2016, at 21:35, Hans Petter Selasky wrote: > > Hi, Thank you for your answer Hans Petter ! > The CX-3 driver doesn't bind the worker threads to specific CPU cores by > default, so if your CPU has more than one so-called numa, you'll end up that > the bottle-neck is the high-speed

[Bug 208136] kldunload if_cxgbe causes kernel panic

2016-08-02 Thread bugzilla-noreply
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=208136 John Baldwin changed: What|Removed |Added Resolution|--- |FIXED CC|

Re: Unstable local network throughput

2016-08-02 Thread Hans Petter Selasky
On 08/02/16 20:43, Ben RUBSON wrote: Hello, I'm trying to reach the 40Gb/s max throughtput between 2 hosts running a ConnectX-3 Mellanox network adapter. FreeBSD 10.3 just installed, last updates performed. Network adapters running last firmwares / last drivers. No workload at all, just iPerf

Unstable local network throughput

2016-08-02 Thread Ben RUBSON
Hello, I'm trying to reach the 40Gb/s max throughtput between 2 hosts running a ConnectX-3 Mellanox network adapter. FreeBSD 10.3 just installed, last updates performed. Network adapters running last firmwares / last drivers. No workload at all, just iPerf as the benchmark tool. ### Step 1 : I

Re: NFS on 10G interfaces still painfully slow

2016-08-02 Thread Alan Somers
On Tue, Aug 2, 2016 at 2:49 AM, Gerrit Kühn wrote: > Hi all, > > I already reported this issue here a year ago and unfortunately was not > able to fix it back then. Now I had another run at it, using two recent > 10.3-machines with a direct 10G link. I still see nfs is painfully > slow (around 20-

Re: NFS on 10G interfaces still painfully slow

2016-08-02 Thread Borja Marcos
> On 02 Aug 2016, at 10:49, Gerrit Kühn wrote: > > Is there anyone around here who can confirm that nfs can go faster over > 10G links? > Any hints for further tuning/debugging are greatly appreciated. Can you show us ifconfig output, please? Borja.

NFS on 10G interfaces still painfully slow

2016-08-02 Thread Gerrit Kühn
Hi all, I already reported this issue here a year ago and unfortunately was not able to fix it back then. Now I had another run at it, using two recent 10.3-machines with a direct 10G link. I still see nfs is painfully slow (around 20-80MB/s). I tried both nfsv3 and nfsv4, with almost the same res