1 |
On Tue, Jul 19, 2016 at 12:49 AM, Adam Carter <adamcarter3@×××××.com> wrote: |
2 |
> |
3 |
> I'm trying to troubleshoot a newly setup nfs server, which, sometimes has a |
4 |
> 30 second pause (tcpdump shows its server waiting). |
5 |
> |
6 |
> # time touch /usr/portage/distfiles/testfile |
7 |
> |
8 |
> real 0m30.088s |
9 |
> user 0m0.000s |
10 |
> sys 0m0.001s |
11 |
> |
12 |
> I cant see anything in the nfs server debugging so i want to strace nfsd. |
13 |
> First i tell it to use single thread so i know i;m stracing the correct |
14 |
> thread, but; |
15 |
> |
16 |
> # grep OPTS_RPC_NFSD nfs |
17 |
> #OPTS_RPC_NFSD="8 -N2 -V 3 -V 4 -V 4.1" |
18 |
> OPTS_RPC_NFSD="1 -N2 -V 3 -V 4 -V 4.1" |
19 |
> # systemctl restart nfs-server |
20 |
> # pgrep -lf nfsd |
21 |
> 23546 nfsd4_callbacks |
22 |
> 23548 nfsd |
23 |
> 23549 nfsd |
24 |
> 23550 nfsd |
25 |
> 23551 nfsd |
26 |
> 23552 nfsd |
27 |
> 23553 nfsd |
28 |
> 23554 nfsd |
29 |
> 23555 nfsd |
30 |
> |
31 |
> So its not respecting the nproc setting. Any ideas? I also tried changing |
32 |
> EXPORTFS_TIMEOUT= since its currently set at 30. It didnt help, but perhaps |
33 |
> that's because its being ignored too. |
34 |
|
35 |
Are the nfsd versions that you're setting being respected? You can |
36 |
check with "rpcinfo -s" or "cat /proc/fs/nfsd/versions". |
37 |
|
38 |
You can change the number of threads on the fly with "echo 1 > |
39 |
/proc/fs/nfsd/threads". |
40 |
|
41 |
I don't use systemd on Gentoo but for the nfs-utils upstream-shipped |
42 |
systemd units that I think that Gentoo's using, you have to re-run |
43 |
nfs-config.service - or run the script that it calls - in order to |
44 |
update the "/run/sysconfig/nfs-utils" environment file that's sourced |
45 |
by the nfs-server.service unit. |
46 |
|
47 |
Does "/var/lib/nfs/v4recovery/" exist? |
48 |
|
49 |
Does adding the client to "/etc/hosts" - or to your reverse dns zone - |
50 |
eliminate the delay? |