我已经在一台机器上安装了 Ubuntu 22.04,我正在尝试挂载一个 NFS 共享,它在其他 Linux Ubuntu 机器(21.04、20.04、18.04 等)的阵列上挂载没有问题。
我们使用的命令是:
sudo apt install nfs-common # Only required once for installation
sudo mount -t nfs 10.234.123.11:/bfx_share1 /bfx_share1
如果我这样做:
sudo showmount -e 10.234.123.11
它显示了可能的坐骑列表,因此在这方面一切都是犹太教。
它只是卡在那里,什么也没做。有任何想法吗?我可以做些什么或添加什么来调试情况?没有其他客户nfs-common
吗?
我已经安装nfstrace
并在尝试挂载 nfs 文件夹时,我在另一个终端中启动了它,以获得以下提示:
## On one terminal
(base) user@LS6-MS-7D04:~$ ping 10.234.123.11
PING 10.234.123.11 (10.234.123.11) 56(84) bytes of data.
64 bytes from 10.234.123.11: icmp_seq=1 ttl=64 time=0.202 ms
64 bytes from 10.234.123.11: icmp_seq=2 ttl=64 time=0.122 ms
64 bytes from 10.234.123.11: icmp_seq=3 ttl=64 time=0.172 ms
^C
--- 10.234.123.11 ping statistics ---
3 packets transmitted, 3 received, 0% packet loss, time 2027ms
rtt min/avg/max/mdev = 0.122/0.165/0.202/0.033 ms
(base) user@LS6-MS-7D04:~$ sudo mount 10.234.123.11:/bfx_share1 /bfx_share1
## In the other terminal with nfstrace running
root@LS6-MS-7D04:/home/user# nfstrace -Z user
Log file: nfstrace.log
Read from interface: enp111s0
BPF filter : port 2049 or port 445
snapshot len: 65535 bytes
read timeout: 100 ms
buffer size : 20971520 bytes
promiscuous mode: on
capture traffic : inout
Processing packets. Press CTRL-C to quit and view results.
Detect session 10.234.123.8:902 --> 10.234.123.11:2049 [TCP]
Detect session 10.234.123.8:772 --> 10.234.123.11:2049 [TCP]
10.234.123.8:772 --> 10.234.123.11:2049 [TCP]NULL
CALL []
REPLY []
Detect session 10.234.123.8:947 --> 10.234.123.11:2049 [TCP]
10.234.123.8:947 --> 10.234.123.11:2049 [TCP]NULL
CALL []
REPLY []
查看 Ubuntu 22.04 的发行说明(Jammy Jellyfish 发行说明): https ://discourse.ubuntu.com/t/jammy-jellyfish-release-notes/24668?_ga=2.137381111.630065420.1651222708-2033241278.1650531954
有一段:
UDP disabled for NFS mounts
Since Ubuntu 20.10 (“Groovy Gorilla”), the kernel option CONFIG_NFS_DISABLE_UDP_SUPPORT=y is set and this disables using UDP as the transport for NFS mounts, regardless of NFS version.
In practice, if you try to use udp, you will get this error:
$ sudo mount f1:/storage /mnt -o udp
mount.nfs: an incorrect mount option was specified
###
NFS server
The NFS server and client packages have finally been updated to the latest upstream version.
All NFS services now read their configuration from /etc/nfs.conf and /etc/nfs.conf.d/*.conf, which is an INI-style configuration file, where each section is about one daemon or aspect of the NFS service. The old /etc/defaults/nfs-* configuration files are still left around, but are unused.
During upgrade, a conversion script is run if the package detects that the /etc/default/nfs-* files have been changed. This script is /usr/share/nfs-common/nfsconvert.py and it will read the options from /etc/defaults/nfs-* and generate /etc/nfs.conf.d/local.conf, which overrides the defaults in /etc/nfs.conf.
If the conversion script fails for some reason, the package installation or upgrade will fail, and the issue will have to be resolved. Please file a bug against nfs-utils in Launchpad 2 if you encounter such a scenario.
A new tool called nfsconf(8) can be used to query the configuration settings of /etc/nfs.conf and /etc/nfs.conf.d/*.conf.
##
the CTDB package was adjusted to work with the new NFS server version shipped in this Ubuntu 22.04
我已经提交了一份错误报告: https ://bugs.launchpad.net/ubuntu/+source/nfs-utils/+bug/1970898