Re: [CLOSED] Re: Problem with Linux >= 3.3 as NFSv4 server

2012-08-22 Thread Rick Macklem
Norbert Aschendorff wrote: > As this is obviously a Linux (kernel(?)) problem, I think I can close > this thread. And thank you for your answers, Rick :) > As noted in my other post, it comes down to which NFSv4 spec is the current one, since although I haven't looked, I'm fairly sure that numeric

Re: Problem with Linux >= 3.3 as NFSv4 server

2012-08-22 Thread Rick Macklem
Norbert Aschendorff wrote: > Sorry, I was wrong :|. This time, the packets contain the numeric user > and group IDs (in my case, 1000:1000). A pcap file with some NFS > requests and responses can be found here: > http://lbo.spheniscida.de/Files/nfs.pcap -- sorry for that > misinformation in the pre

[CLOSED] Re: Problem with Linux >= 3.3 as NFSv4 server

2012-08-22 Thread Norbert Aschendorff
As this is obviously a Linux (kernel(?)) problem, I think I can close this thread. And thank you for your answers, Rick :) Norbert ___ freebsd-stable@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-stable To unsubscribe, send a

Re: Problem with Linux >= 3.3 as NFSv4 server

2012-08-22 Thread Norbert Aschendorff
And just another thing: Using the Fedora machine (also kernel 3.5) as server, I see the exactly same behavior. Sending '1000' (numeric UID) instead of actual username. Using Kernel 3.2 on the server, the translation even works when using the same username and different IDs (as it should...). _

Re: Problem with Linux >= 3.3 as NFSv4 server

2012-08-22 Thread Norbert Aschendorff
It looks as if the problem is related to this bug which shows the exactly same symptoms (if you look at the package dumps). It's also Kernel 3.3 with which it also begun here. And with kernel 3.1, it works (same here). --> https://bugzilla.novell.com/show_bug.cgi?id=756897 I don't know what's the

Re: Problem with Linux >= 3.3 as NFSv4 server

2012-08-22 Thread Norbert Aschendorff
Sorry, I was wrong :|. This time, the packets contain the numeric user and group IDs (in my case, 1000:1000). A pcap file with some NFS requests and responses can be found here: http://lbo.spheniscida.de/Files/nfs.pcap -- sorry for that misinformation in the previous mail :( Norbert __

Re: Problem with Linux >= 3.3 as NFSv4 server

2012-08-22 Thread Rick Macklem
Norbert Aschendorff wrote: > I already had a problem with this, but this is fixed: 1. The hostnames > are correct 2. in the debian /etc/idmapd.conf, the right domain name > is > specified 3. The names are transmitted correctly -- tested with > Wireshark. If the problem was wrong domain names, I had

Re: Problem with Linux >= 3.3 as NFSv4 server

2012-08-22 Thread Norbert Aschendorff
I already had a problem with this, but this is fixed: 1. The hostnames are correct 2. in the debian /etc/idmapd.conf, the right domain name is specified 3. The names are transmitted correctly -- tested with Wireshark. If the problem was wrong domain names, I had found it because I always take a loo

Re: Problem with Linux >= 3.3 as NFSv4 server

2012-08-20 Thread Rick Macklem
Norbert Aschendorff wrote: > Hi all, > I recently noticed a problem in my network. I use some desktop > machines > there, two with Linux kernel (Debian and Fedora, both using Kernel > 3.5) > and a FreeBSD 9.0 (p4) machine. > Some days ago, I updated the Debian machine from Kernel 3.2 to Kernel > 3.

Problem with Linux >= 3.3 as NFSv4 server

2012-08-20 Thread Norbert Aschendorff
Hi all, I recently noticed a problem in my network. I use some desktop machines there, two with Linux kernel (Debian and Fedora, both using Kernel 3.5) and a FreeBSD 9.0 (p4) machine. Some days ago, I updated the Debian machine from Kernel 3.2 to Kernel 3.5 (from the experimental branch, 3.5-trunk-