Groups | Search | Server Info | Login | Register

Groups > comp.protocols.nfs > #21

tcp socket to mountd port in time_wait state in linux-- how to recover??

X-Received by 10.58.242.70 with SMTP id wo6mr2440058vec.16.1388157149956; Fri, 27 Dec 2013 07:12:29 -0800 (PST)
X-Received by 10.49.94.144 with SMTP id dc16mr21626qeb.21.1388157149938; Fri, 27 Dec 2013 07:12:29 -0800 (PST)
Path csiph.com!v102.xanadu-bbs.net!xanadu-bbs.net!usenet.blueworldhosting.com!feeder01.blueworldhosting.com!peer02.iad.highwinds-media.com!news.highwinds-media.com!feed-me.highwinds-media.com!p15no45212647qaj.0!news-out.google.com!l9ni207qay.0!nntp.google.com!p15no45212645qaj.0!postnews.google.com!glegroupsg2000goo.googlegroups.com!not-for-mail
Newsgroups comp.protocols.nfs
Date Fri, 27 Dec 2013 07:12:29 -0800 (PST)
Complaints-To groups-abuse@google.com
Injection-Info glegroupsg2000goo.googlegroups.com; posting-host=122.166.93.154; posting-account=54nESwoAAAA5nZHbdfLeq2V8VrpdB9Ko
NNTP-Posting-Host 122.166.93.154
User-Agent G2/1.0
MIME-Version 1.0
Message-ID <acd2f06e-3182-4376-b847-b10a06c0e1b1@googlegroups.com> (permalink)
Subject tcp socket to mountd port in time_wait state in linux-- how to recover??
From smilesonisamal@gmail.com
Injection-Date Fri, 27 Dec 2013 15:12:29 +0000
Content-Type text/plain; charset=ISO-8859-1
X-Received-Bytes 2460
X-Received-Body-CRC 141346147
Xref csiph.com comp.protocols.nfs:21

Show key headers only | View raw

Hi all,
  I am using Linux 6.4. When I am trying to mount from the client from a program via the tcp socket the server closes the tcp socket after reaching the state TIME_WAIT.
I have disabled the iptables rules. But still it doesnot help.

I think I am having similar problem as mentioned in below thread.

http://www.spinics.net/lists/linux-nfs/msg13036.html

I have a couple of questions:

1. How can i recover from this state?

[root@hostarena~]# netstat -tp
Active Internet connections (w/o servers)
Proto Recv-Q Send-Q Local Address               Foreign Address             State       PID/Program name   
tcp        0      0 hostarena.yyy.org:49423 hostblob.yyy.org:33973 TIME_WAIT   -                           
[root@hostarena ~]

the /var/log/messages after enabling the rpc_debug didnot show anything meaningful. 

NFS Server is listening for mount requests at port 49423. I am suspecting time_wait state is due to the fact that source-port is > 1024(i.e 33973) for root.

2. Is there a way i can configure the linux nfs server to accept from non- reserved port > 1024?

3. Is there a way i can tell the client socket to choose the source ports < 1024 for mount without changing the /proc/sys/net/ipv4/ip_local_port_range at the client?

I appreciate somebody can throw some light to fix the issue?

Note: All the RPC services are running at the NFS server(confirmed by rpcinfo -p)

Regards
Pradeep

Back to comp.protocols.nfs | Previous | Next — Next in thread | Find similar

Thread

tcp socket to mountd port in time_wait state in linux-- how to recover?? smilesonisamal@gmail.com - 2013-12-27 07:12 -0800
  Re: tcp socket to mountd port in time_wait state in linux-- how to recover?? "J.O. Aho" <user@example.net> - 2013-12-27 22:06 +0100
    Re: tcp socket to mountd port in time_wait state in linux-- how to recover?? smilesonisamal@gmail.com - 2013-12-29 11:49 -0800

csiph-web