All of lore.kernel.org
 help / color / mirror / Atom feed
* mac error
@ 2011-01-14  1:13 hce
  2011-01-14  2:37 ` Mi Jinlong
  0 siblings, 1 reply; 9+ messages in thread
From: hce @ 2011-01-14  1:13 UTC (permalink / raw)
  To: linux-nfs

Hi,

I am running nfs server on a CentOS 5, it works fine for all
connections from linux, windows clients. Mac clients worked fine for
coping some regular files, but it could not copy mac packaging file
with the extension *.dmg, it stuck with an error "nfs server lockd not
responding", at the same time the server got a log message of
"rpc.statd[2448]: Can't callback linux_server (100021,4), giving up".

Appreciate advice what could cause that problem and how could fix it?

Thank you.

hce

^ permalink raw reply	[flat|nested] 9+ messages in thread

* Re: mac error
  2011-01-14  1:13 mac error hce
@ 2011-01-14  2:37 ` Mi Jinlong
  2011-01-14  4:01   ` hce
  0 siblings, 1 reply; 9+ messages in thread
From: Mi Jinlong @ 2011-01-14  2:37 UTC (permalink / raw)
  To: hce; +Cc: linux-nfs



hce :
> Hi,
> 
> I am running nfs server on a CentOS 5, it works fine for all
> connections from linux, windows clients. Mac clients worked fine for
> coping some regular files, but it could not copy mac packaging file
> with the extension *.dmg, it stuck with an error "nfs server lockd not
> responding", at the same time the server got a log message of
> "rpc.statd[2448]: Can't callback linux_server (100021,4), giving up".
> 
> Appreciate advice what could cause that problem and how could fix it?

Hi hce,

You use NFSv3, 
the NFSv3's file lock service depend on NLM(Network Lock Manager),
that need a user space program named rpc.statd to support service
like this:

           |     nfs client     |      nfs server 
-------------------------------------------------
           |                    |
user space |     rpc.statd    <--->      rpc.statd
           |        ^           |            ^
--------------------|------------------------|----
           |        v           |            v
kernel     |      lockd       <--->       lockd

when your nfs client is a MAC, maybe Mac's user space don't support
a program as rpc.statd, or you don't make the program run correctly.

Maybe you can use NFSv4 to avoid this problem.


-- 
----
thanks
Mi Jinlong


^ permalink raw reply	[flat|nested] 9+ messages in thread

* Re: mac error
  2011-01-14  2:37 ` Mi Jinlong
@ 2011-01-14  4:01   ` hce
  2011-01-14  6:02     ` Mi Jinlong
  2011-01-14 15:18     ` Chuck Lever
  0 siblings, 2 replies; 9+ messages in thread
From: hce @ 2011-01-14  4:01 UTC (permalink / raw)
  To: Mi Jinlong; +Cc: linux-nfs

Thanks for your response. I've just found that nlockmgr did not
respond when I run following command on the server. Is it a problem or
it is normal? Anyway, the NFS is still working fine.

$ rpcinfo -t linux_server nlockmgr
rpcinfo: RPC: Timed out
program 100021 version 0 is not available

$ rpcinfo -u linux_server nlockmgr
rpcinfo: RPC: Timed out
program 100021 version 0 is not available

$ rpcinfo -p
   program vers proto   port
    100000    2   tcp    111  portmapper
    100000    2   udp    111  portmapper
    100024    1   udp    895  status
    100024    1   tcp    898  status
    100011    1   udp    806  rquotad
    100011    2   udp    806  rquotad
    100011    1   tcp    809  rquotad
    100011    2   tcp    809  rquotad
    100003    2   udp   2049  nfs
    100003    3   udp   2049  nfs
    100003    4   udp   2049  nfs
    100021    1   udp  32794  nlockmgr
    100021    3   udp  32794  nlockmgr
    100021    4   udp  32794  nlockmgr
    100003    2   tcp   2049  nfs
    100003    3   tcp   2049  nfs
    100003    4   tcp   2049  nfs
    100021    1   tcp  57164  nlockmgr
    100021    3   tcp  57164  nlockmgr
    100021    4   tcp  57164  nlockmgr
    100005    1   udp    828  mountd
    100005    1   tcp    831  mountd
    100005    2   udp    828  mountd
    100005    2   tcp    831  mountd
    100005    3   udp    828  mountd
    100005    3   tcp    831  mountd

On Fri, Jan 14, 2011 at 1:37 PM, Mi Jinlong <mijinlong@cn.fujitsu.com> wrote:
>
>
> hce :
>> Hi,
>>
>> I am running nfs server on a CentOS 5, it works fine for all
>> connections from linux, windows clients. Mac clients worked fine for
>> coping some regular files, but it could not copy mac packaging file
>> with the extension *.dmg, it stuck with an error "nfs server lockd not
>> responding", at the same time the server got a log message of
>> "rpc.statd[2448]: Can't callback linux_server (100021,4), giving up".
>>
>> Appreciate advice what could cause that problem and how could fix it?
>
> Hi hce,
>
> You use NFSv3,
> the NFSv3's file lock service depend on NLM(Network Lock Manager),
> that need a user space program named rpc.statd to support service
> like this:
>
>           |     nfs client     |      nfs server
> -------------------------------------------------
>           |                    |
> user space |     rpc.statd    <--->      rpc.statd
>           |        ^           |            ^
> --------------------|------------------------|----
>           |        v           |            v
> kernel     |      lockd       <--->       lockd
>
> when your nfs client is a MAC, maybe Mac's user space don't support
> a program as rpc.statd, or you don't make the program run correctly.
>
> Maybe you can use NFSv4 to avoid this problem.
>
>
> --
> ----
> thanks
> Mi Jinlong
>
>

^ permalink raw reply	[flat|nested] 9+ messages in thread

* Re: mac error
  2011-01-14  4:01   ` hce
@ 2011-01-14  6:02     ` Mi Jinlong
  2011-01-14  6:24       ` hce
  2011-01-14 15:18     ` Chuck Lever
  1 sibling, 1 reply; 9+ messages in thread
From: Mi Jinlong @ 2011-01-14  6:02 UTC (permalink / raw)
  To: hce; +Cc: linux-nfs



hce :
> Thanks for your response. I've just found that nlockmgr did not
> respond when I run following command on the server. Is it a problem or
> it is normal? Anyway, the NFS is still working fine.

  No, it's not a problem.
  It just a lack of NFSv3's file lock, if you try to lock NFSv3 file,
  you must start the nfslock service both the client and server
  (maybe other platform don't named nfslock).

  Because your normal NFS operation don't need file lock,
  so NFSv3 is still working fine.

> 
> $ rpcinfo -t linux_server nlockmgr
> rpcinfo: RPC: Timed out
> program 100021 version 0 is not available
> 
> $ rpcinfo -u linux_server nlockmgr
> rpcinfo: RPC: Timed out
> program 100021 version 0 is not available
> 
> $ rpcinfo -p

  The linux_server and localhost are different machine ?

  linux_server is CentOS?? or MAC?
  localhost is?


-- 
thanks,
Mi Jinlong

>    program vers proto   port
>     100000    2   tcp    111  portmapper
>     100000    2   udp    111  portmapper
>     100024    1   udp    895  status
>     100024    1   tcp    898  status
>     100011    1   udp    806  rquotad
>     100011    2   udp    806  rquotad
>     100011    1   tcp    809  rquotad
>     100011    2   tcp    809  rquotad
>     100003    2   udp   2049  nfs
>     100003    3   udp   2049  nfs
>     100003    4   udp   2049  nfs
>     100021    1   udp  32794  nlockmgr
>     100021    3   udp  32794  nlockmgr
>     100021    4   udp  32794  nlockmgr
>     100003    2   tcp   2049  nfs
>     100003    3   tcp   2049  nfs
>     100003    4   tcp   2049  nfs
>     100021    1   tcp  57164  nlockmgr
>     100021    3   tcp  57164  nlockmgr
>     100021    4   tcp  57164  nlockmgr
>     100005    1   udp    828  mountd
>     100005    1   tcp    831  mountd
>     100005    2   udp    828  mountd
>     100005    2   tcp    831  mountd
>     100005    3   udp    828  mountd
>     100005    3   tcp    831  mountd
> 
> On Fri, Jan 14, 2011 at 1:37 PM, Mi Jinlong <mijinlong@cn.fujitsu.com> wrote:
>>
>> hce :
>>> Hi,
>>>
>>> I am running nfs server on a CentOS 5, it works fine for all
>>> connections from linux, windows clients. Mac clients worked fine for
>>> coping some regular files, but it could not copy mac packaging file
>>> with the extension *.dmg, it stuck with an error "nfs server lockd not
>>> responding", at the same time the server got a log message of
>>> "rpc.statd[2448]: Can't callback linux_server (100021,4), giving up".
>>>
>>> Appreciate advice what could cause that problem and how could fix it?
>> Hi hce,
>>
>> You use NFSv3,
>> the NFSv3's file lock service depend on NLM(Network Lock Manager),
>> that need a user space program named rpc.statd to support service
>> like this:
>>
>>           |     nfs client     |      nfs server
>> -------------------------------------------------
>>           |                    |
>> user space |     rpc.statd    <--->      rpc.statd
>>           |        ^           |            ^
>> --------------------|------------------------|----
>>           |        v           |            v
>> kernel     |      lockd       <--->       lockd
>>
>> when your nfs client is a MAC, maybe Mac's user space don't support
>> a program as rpc.statd, or you don't make the program run correctly.
>>
>> Maybe you can use NFSv4 to avoid this problem.
>>
>>
>> --
>> ----
>> thanks
>> Mi Jinlong
>>
>>
> 
> 


^ permalink raw reply	[flat|nested] 9+ messages in thread

* Re: mac error
  2011-01-14  6:02     ` Mi Jinlong
@ 2011-01-14  6:24       ` hce
  0 siblings, 0 replies; 9+ messages in thread
From: hce @ 2011-01-14  6:24 UTC (permalink / raw)
  To: Mi Jinlong; +Cc: linux-nfs

On Fri, Jan 14, 2011 at 5:02 PM, Mi Jinlong <mijinlong@cn.fujitsu.com> wrote:
>
>
> hce :
>> Thanks for your response. I've just found that nlockmgr did not
>> respond when I run following command on the server. Is it a problem or
>> it is normal? Anyway, the NFS is still working fine.
>
>  No, it's not a problem.
>  It just a lack of NFSv3's file lock, if you try to lock NFSv3 file,
>  you must start the nfslock service both the client and server
>  (maybe other platform don't named nfslock).
>
>  Because your normal NFS operation don't need file lock,
>  so NFSv3 is still working fine.

Thanks. Good to know, so I don't need to worry about it.

>>
>> $ rpcinfo -t linux_server nlockmgr
>> rpcinfo: RPC: Timed out
>> program 100021 version 0 is not available
>>
>> $ rpcinfo -u linux_server nlockmgr
>> rpcinfo: RPC: Timed out
>> program 100021 version 0 is not available
>>
>> $ rpcinfo -p
>
>  The linux_server and localhost are different machine ?
>
>  linux_server is CentOS?? or MAC?
>  localhost is?

The linux_server is a host for NFS server running CentOS. All commands
were called in the NFS server.

Anyway, the MAC clients sometime crashed or sometime worked very slow
when the NFS server got rpc.statd[2448]: Can't callback .... If I
understand it correct, it should not cause the MAC networking problem,
then we have to look at the problem in other places.

Thank you.

Kind regards.

>
> --
> thanks,
> Mi Jinlong
>
>>    program vers proto   port
>>     100000    2   tcp    111  portmapper
>>     100000    2   udp    111  portmapper
>>     100024    1   udp    895  status
>>     100024    1   tcp    898  status
>>     100011    1   udp    806  rquotad
>>     100011    2   udp    806  rquotad
>>     100011    1   tcp    809  rquotad
>>     100011    2   tcp    809  rquotad
>>     100003    2   udp   2049  nfs
>>     100003    3   udp   2049  nfs
>>     100003    4   udp   2049  nfs
>>     100021    1   udp  32794  nlockmgr
>>     100021    3   udp  32794  nlockmgr
>>     100021    4   udp  32794  nlockmgr
>>     100003    2   tcp   2049  nfs
>>     100003    3   tcp   2049  nfs
>>     100003    4   tcp   2049  nfs
>>     100021    1   tcp  57164  nlockmgr
>>     100021    3   tcp  57164  nlockmgr
>>     100021    4   tcp  57164  nlockmgr
>>     100005    1   udp    828  mountd
>>     100005    1   tcp    831  mountd
>>     100005    2   udp    828  mountd
>>     100005    2   tcp    831  mountd
>>     100005    3   udp    828  mountd
>>     100005    3   tcp    831  mountd
>>
>> On Fri, Jan 14, 2011 at 1:37 PM, Mi Jinlong <mijinlong@cn.fujitsu.com> wrote:
>>>
>>> hce :
>>>> Hi,
>>>>
>>>> I am running nfs server on a CentOS 5, it works fine for all
>>>> connections from linux, windows clients. Mac clients worked fine for
>>>> coping some regular files, but it could not copy mac packaging file
>>>> with the extension *.dmg, it stuck with an error "nfs server lockd not
>>>> responding", at the same time the server got a log message of
>>>> "rpc.statd[2448]: Can't callback linux_server (100021,4), giving up".
>>>>
>>>> Appreciate advice what could cause that problem and how could fix it?
>>> Hi hce,
>>>
>>> You use NFSv3,
>>> the NFSv3's file lock service depend on NLM(Network Lock Manager),
>>> that need a user space program named rpc.statd to support service
>>> like this:
>>>
>>>           |     nfs client     |      nfs server
>>> -------------------------------------------------
>>>           |                    |
>>> user space |     rpc.statd    <--->      rpc.statd
>>>           |        ^           |            ^
>>> --------------------|------------------------|----
>>>           |        v           |            v
>>> kernel     |      lockd       <--->       lockd
>>>
>>> when your nfs client is a MAC, maybe Mac's user space don't support
>>> a program as rpc.statd, or you don't make the program run correctly.
>>>
>>> Maybe you can use NFSv4 to avoid this problem.
>>>
>>>
>>> --
>>> ----
>>> thanks
>>> Mi Jinlong
>>>
>>>
>>
>>
>
>

^ permalink raw reply	[flat|nested] 9+ messages in thread

* Re: mac error
  2011-01-14  4:01   ` hce
  2011-01-14  6:02     ` Mi Jinlong
@ 2011-01-14 15:18     ` Chuck Lever
  2011-01-14 15:55       ` peter.staubach
  1 sibling, 1 reply; 9+ messages in thread
From: Chuck Lever @ 2011-01-14 15:18 UTC (permalink / raw)
  To: hce; +Cc: Linux NFS Mailing List


On Jan 13, 2011, at 11:01 PM, hce wrote:

> Thanks for your response. I've just found that nlockmgr did not
> respond when I run following command on the server. Is it a problem or
> it is normal? Anyway, the NFS is still working fine.

nlockmgr and NFS are two separate RPC services.  So, NFS can work while there might be a problem with nlockmgr.  IMO, nlockmgr should respond in this case.

> $ rpcinfo -t linux_server nlockmgr
> rpcinfo: RPC: Timed out
> program 100021 version 0 is not available
> 
> $ rpcinfo -u linux_server nlockmgr
> rpcinfo: RPC: Timed out
> program 100021 version 0 is not available

This is attempting to contact nlockmgr's "version 0" which is not registered (see below).  Versions 1, 3, and 4 are registered.

Try "rpcinfo -u linux_server nlockmgr 3".

> $ rpcinfo -p
>   program vers proto   port
>    100000    2   tcp    111  portmapper
>    100000    2   udp    111  portmapper
>    100024    1   udp    895  status
>    100024    1   tcp    898  status
>    100011    1   udp    806  rquotad
>    100011    2   udp    806  rquotad
>    100011    1   tcp    809  rquotad
>    100011    2   tcp    809  rquotad
>    100003    2   udp   2049  nfs
>    100003    3   udp   2049  nfs
>    100003    4   udp   2049  nfs
>    100021    1   udp  32794  nlockmgr
>    100021    3   udp  32794  nlockmgr
>    100021    4   udp  32794  nlockmgr
>    100003    2   tcp   2049  nfs
>    100003    3   tcp   2049  nfs
>    100003    4   tcp   2049  nfs
>    100021    1   tcp  57164  nlockmgr
>    100021    3   tcp  57164  nlockmgr
>    100021    4   tcp  57164  nlockmgr
>    100005    1   udp    828  mountd
>    100005    1   tcp    831  mountd
>    100005    2   udp    828  mountd
>    100005    2   tcp    831  mountd
>    100005    3   udp    828  mountd
>    100005    3   tcp    831  mountd

nlockmgr is registered, so it must have been started at some point.  It's a kernel service, so there's no daemon to die.  Do you have some kind of firewall in place that would prevent contacting the NLM service?  Does "linux_server" resolve to the IP address you expect it to?

-- 
Chuck Lever
chuck[dot]lever[at]oracle[dot]com





^ permalink raw reply	[flat|nested] 9+ messages in thread

* RE: mac error
  2011-01-14 15:18     ` Chuck Lever
@ 2011-01-14 15:55       ` peter.staubach
  2011-01-14 17:00         ` Chuck Lever
  0 siblings, 1 reply; 9+ messages in thread
From: peter.staubach @ 2011-01-14 15:55 UTC (permalink / raw)
  To: chuck.lever, webmail.hce; +Cc: linux-nfs

Hi.

Actually, that command should work.  The server should respond with an
RPC error indicating which
versions of the nlockmgr protocol are supported and then rpcinfo will
ping each of them.

		ps

-----Original Message-----
From: linux-nfs-owner@vger.kernel.org
[mailto:linux-nfs-owner@vger.kernel.org] On Behalf Of Chuck Lever
Sent: Friday, January 14, 2011 10:18 AM
To: hce
Cc: Linux NFS Mailing List
Subject: Re: mac error


On Jan 13, 2011, at 11:01 PM, hce wrote:

> Thanks for your response. I've just found that nlockmgr did not
> respond when I run following command on the server. Is it a problem or
> it is normal? Anyway, the NFS is still working fine.

nlockmgr and NFS are two separate RPC services.  So, NFS can work while
there might be a problem with nlockmgr.  IMO, nlockmgr should respond in
this case.

> $ rpcinfo -t linux_server nlockmgr
> rpcinfo: RPC: Timed out
> program 100021 version 0 is not available
> 
> $ rpcinfo -u linux_server nlockmgr
> rpcinfo: RPC: Timed out
> program 100021 version 0 is not available

This is attempting to contact nlockmgr's "version 0" which is not
registered (see below).  Versions 1, 3, and 4 are registered.

Try "rpcinfo -u linux_server nlockmgr 3".

> $ rpcinfo -p
>   program vers proto   port
>    100000    2   tcp    111  portmapper
>    100000    2   udp    111  portmapper
>    100024    1   udp    895  status
>    100024    1   tcp    898  status
>    100011    1   udp    806  rquotad
>    100011    2   udp    806  rquotad
>    100011    1   tcp    809  rquotad
>    100011    2   tcp    809  rquotad
>    100003    2   udp   2049  nfs
>    100003    3   udp   2049  nfs
>    100003    4   udp   2049  nfs
>    100021    1   udp  32794  nlockmgr
>    100021    3   udp  32794  nlockmgr
>    100021    4   udp  32794  nlockmgr
>    100003    2   tcp   2049  nfs
>    100003    3   tcp   2049  nfs
>    100003    4   tcp   2049  nfs
>    100021    1   tcp  57164  nlockmgr
>    100021    3   tcp  57164  nlockmgr
>    100021    4   tcp  57164  nlockmgr
>    100005    1   udp    828  mountd
>    100005    1   tcp    831  mountd
>    100005    2   udp    828  mountd
>    100005    2   tcp    831  mountd
>    100005    3   udp    828  mountd
>    100005    3   tcp    831  mountd

nlockmgr is registered, so it must have been started at some point.
It's a kernel service, so there's no daemon to die.  Do you have some
kind of firewall in place that would prevent contacting the NLM service?
Does "linux_server" resolve to the IP address you expect it to?

-- 
Chuck Lever
chuck[dot]lever[at]oracle[dot]com




--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html


^ permalink raw reply	[flat|nested] 9+ messages in thread

* Re: mac error
  2011-01-14 15:55       ` peter.staubach
@ 2011-01-14 17:00         ` Chuck Lever
  2011-01-16 22:09           ` hce
  0 siblings, 1 reply; 9+ messages in thread
From: Chuck Lever @ 2011-01-14 17:00 UTC (permalink / raw)
  To: peter.staubach; +Cc: webmail.hce, linux-nfs


On Jan 14, 2011, at 10:55 AM, peter.staubach@emc.com wrote:

> Hi.
> 
> Actually, that command should work.  The server should respond with an
> RPC error indicating which
> versions of the nlockmgr protocol are supported and then rpcinfo will
> ping each of them.

Yes, that and the "RPC: Timed out" message suggests that the underlying transport is not even getting through.

I expect lockd to unregister itself with the portmapper when it is shut down cleanly.  The presence of a registration for nlockmgr indicates it should still be running and contactable.

> 
> 		ps
> 
> -----Original Message-----
> From: linux-nfs-owner@vger.kernel.org
> [mailto:linux-nfs-owner@vger.kernel.org] On Behalf Of Chuck Lever
> Sent: Friday, January 14, 2011 10:18 AM
> To: hce
> Cc: Linux NFS Mailing List
> Subject: Re: mac error
> 
> 
> On Jan 13, 2011, at 11:01 PM, hce wrote:
> 
>> Thanks for your response. I've just found that nlockmgr did not
>> respond when I run following command on the server. Is it a problem or
>> it is normal? Anyway, the NFS is still working fine.
> 
> nlockmgr and NFS are two separate RPC services.  So, NFS can work while
> there might be a problem with nlockmgr.  IMO, nlockmgr should respond in
> this case.
> 
>> $ rpcinfo -t linux_server nlockmgr
>> rpcinfo: RPC: Timed out
>> program 100021 version 0 is not available
>> 
>> $ rpcinfo -u linux_server nlockmgr
>> rpcinfo: RPC: Timed out
>> program 100021 version 0 is not available
> 
> This is attempting to contact nlockmgr's "version 0" which is not
> registered (see below).  Versions 1, 3, and 4 are registered.
> 
> Try "rpcinfo -u linux_server nlockmgr 3".
> 
>> $ rpcinfo -p
>>  program vers proto   port
>>   100000    2   tcp    111  portmapper
>>   100000    2   udp    111  portmapper
>>   100024    1   udp    895  status
>>   100024    1   tcp    898  status
>>   100011    1   udp    806  rquotad
>>   100011    2   udp    806  rquotad
>>   100011    1   tcp    809  rquotad
>>   100011    2   tcp    809  rquotad
>>   100003    2   udp   2049  nfs
>>   100003    3   udp   2049  nfs
>>   100003    4   udp   2049  nfs
>>   100021    1   udp  32794  nlockmgr
>>   100021    3   udp  32794  nlockmgr
>>   100021    4   udp  32794  nlockmgr
>>   100003    2   tcp   2049  nfs
>>   100003    3   tcp   2049  nfs
>>   100003    4   tcp   2049  nfs
>>   100021    1   tcp  57164  nlockmgr
>>   100021    3   tcp  57164  nlockmgr
>>   100021    4   tcp  57164  nlockmgr
>>   100005    1   udp    828  mountd
>>   100005    1   tcp    831  mountd
>>   100005    2   udp    828  mountd
>>   100005    2   tcp    831  mountd
>>   100005    3   udp    828  mountd
>>   100005    3   tcp    831  mountd
> 
> nlockmgr is registered, so it must have been started at some point.
> It's a kernel service, so there's no daemon to die.  Do you have some
> kind of firewall in place that would prevent contacting the NLM service?
> Does "linux_server" resolve to the IP address you expect it to?
> 
> -- 
> Chuck Lever
> chuck[dot]lever[at]oracle[dot]com
> 
> 
> 
> 
> --
> To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> 

-- 
Chuck Lever
chuck[dot]lever[at]oracle[dot]com





^ permalink raw reply	[flat|nested] 9+ messages in thread

* Re: mac error
  2011-01-14 17:00         ` Chuck Lever
@ 2011-01-16 22:09           ` hce
  0 siblings, 0 replies; 9+ messages in thread
From: hce @ 2011-01-16 22:09 UTC (permalink / raw)
  To: Chuck Lever; +Cc: peter.staubach, linux-nfs

On Sat, Jan 15, 2011 at 4:00 AM, Chuck Lever <chuck.lever@oracle.com> wrote:
>
> On Jan 14, 2011, at 10:55 AM, peter.staubach@emc.com wrote:
>
>> Hi.
>>
>> Actually, that command should work.  The server should respond with an
>> RPC error indicating which
>> versions of the nlockmgr protocol are supported and then rpcinfo will
>> ping each of them.
>
> Yes, that and the "RPC: Timed out" message suggests that the underlying transport is not even getting through.

The  "rpcinfo -u linux_server nlockmgr 3" was timeout either. That
server resolving IP address was working fine. As you indicated it
might have underlying transports problem. I'll have to find it out.

Your comments confirmed my suspicious that the problem of the nlockmgr
timeout could cause the problem we currently got. I'll follow your
lead and debug it.

Thank you all.

> I expect lockd to unregister itself with the portmapper when it is shut down cleanly.  The presence of a registration for nlockmgr indicates it should still be running and contactable.
>
>>
>>               ps
>>
>> -----Original Message-----
>> From: linux-nfs-owner@vger.kernel.org
>> [mailto:linux-nfs-owner@vger.kernel.org] On Behalf Of Chuck Lever
>> Sent: Friday, January 14, 2011 10:18 AM
>> To: hce
>> Cc: Linux NFS Mailing List
>> Subject: Re: mac error
>>
>>
>> On Jan 13, 2011, at 11:01 PM, hce wrote:
>>
>>> Thanks for your response. I've just found that nlockmgr did not
>>> respond when I run following command on the server. Is it a problem or
>>> it is normal? Anyway, the NFS is still working fine.
>>
>> nlockmgr and NFS are two separate RPC services.  So, NFS can work while
>> there might be a problem with nlockmgr.  IMO, nlockmgr should respond in
>> this case.
>>
>>> $ rpcinfo -t linux_server nlockmgr
>>> rpcinfo: RPC: Timed out
>>> program 100021 version 0 is not available
>>>
>>> $ rpcinfo -u linux_server nlockmgr
>>> rpcinfo: RPC: Timed out
>>> program 100021 version 0 is not available
>>
>> This is attempting to contact nlockmgr's "version 0" which is not
>> registered (see below).  Versions 1, 3, and 4 are registered.
>>
>> Try "rpcinfo -u linux_server nlockmgr 3".
>>
>>> $ rpcinfo -p
>>>  program vers proto   port
>>>   100000    2   tcp    111  portmapper
>>>   100000    2   udp    111  portmapper
>>>   100024    1   udp    895  status
>>>   100024    1   tcp    898  status
>>>   100011    1   udp    806  rquotad
>>>   100011    2   udp    806  rquotad
>>>   100011    1   tcp    809  rquotad
>>>   100011    2   tcp    809  rquotad
>>>   100003    2   udp   2049  nfs
>>>   100003    3   udp   2049  nfs
>>>   100003    4   udp   2049  nfs
>>>   100021    1   udp  32794  nlockmgr
>>>   100021    3   udp  32794  nlockmgr
>>>   100021    4   udp  32794  nlockmgr
>>>   100003    2   tcp   2049  nfs
>>>   100003    3   tcp   2049  nfs
>>>   100003    4   tcp   2049  nfs
>>>   100021    1   tcp  57164  nlockmgr
>>>   100021    3   tcp  57164  nlockmgr
>>>   100021    4   tcp  57164  nlockmgr
>>>   100005    1   udp    828  mountd
>>>   100005    1   tcp    831  mountd
>>>   100005    2   udp    828  mountd
>>>   100005    2   tcp    831  mountd
>>>   100005    3   udp    828  mountd
>>>   100005    3   tcp    831  mountd
>>
>> nlockmgr is registered, so it must have been started at some point.
>> It's a kernel service, so there's no daemon to die.  Do you have some
>> kind of firewall in place that would prevent contacting the NLM service?
>> Does "linux_server" resolve to the IP address you expect it to?
>>
>> --
>> Chuck Lever
>> chuck[dot]lever[at]oracle[dot]com
>>
>>
>>
>>
>> --
>> To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
>> the body of a message to majordomo@vger.kernel.org
>> More majordomo info at  http://vger.kernel.org/majordomo-info.html
>>
>
> --
> Chuck Lever
> chuck[dot]lever[at]oracle[dot]com
>
>
>
>
>

^ permalink raw reply	[flat|nested] 9+ messages in thread

end of thread, other threads:[~2011-01-16 22:09 UTC | newest]

Thread overview: 9+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2011-01-14  1:13 mac error hce
2011-01-14  2:37 ` Mi Jinlong
2011-01-14  4:01   ` hce
2011-01-14  6:02     ` Mi Jinlong
2011-01-14  6:24       ` hce
2011-01-14 15:18     ` Chuck Lever
2011-01-14 15:55       ` peter.staubach
2011-01-14 17:00         ` Chuck Lever
2011-01-16 22:09           ` hce

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.