perfSONAR User Q&A and Other Discussion

[Andrew Lake <>]

Hi Carlo,

Thanks for the info. An upgrade from 3.2.2 might help. It's a migration from 
CentOS 5 to CentOS 6 so a lot of the involved components will change. 3.2.2 
saw its end-of-life a couple weeks ago so I would definitely recommend an 
upgrade. If after the upgrade the problem persists, do not hesitate to let us 
know.

Thanks,
Andy

On Apr 10, 2014, at 4:09 PM, Carlo Musante 
<>
 wrote:

> Andy,
> 
> The server is running pS-Performance Toolkit v3.2.2. I guess it's time to 
> upgrade.  For the record, netstat reports there is a CLOSE-WAIT state still 
> active on the port used by npad. 
> 
> tcp        1      0 192.168.1.2:8001            192.168.2.69:40109          
> CLOSE_WAIT  
> 
> Killing the process dropped the connection.  The server which created the 
> connection runs nessus security scans. I suppose it depends to what level 
> the nessus scan is trying to intrude into the system as to why the process 
> is still running.  Since this is an older version of perfSonar I will 
> upgrade to the latest version.
> 
> Thanks,
> 
> 
> Carlo
> 
> 
> 
> 
> ----- Original Message -----
> From: "Andrew Lake" 
> <>
> To: "Carlo Musante" 
> <>
> Cc: 
> 
> Sent: Thursday, April 10, 2014 10:13:32 AM
> Subject: Re: Npad Issue
> 
> Hi,
> 
> Copying the perfsonar-user list in case others have insights. I do recall a 
> thread about the usage shooting up last year. We hypothesized at the time a 
> web crawler or similar was hitting the server and confusing it, but we 
> could never confirm. It's made more difficult by the fact that it's unclear 
> who the maintainer of the NPAD source is these days. Are you running 
> version 3.3.2 of the toolkit? I haven't heard reports in awhile but that 
> doesn't mean people have stopped seeing it (just stopped emailing about 
> it). It's also possible (though probably unlikely) that if you are running 
> an old version of the toolkit that some update fixed it.
> 
> Thanks,
> Andy
> 
> 
> 
> 
> 
> On Apr 10, 2014, at 9:58 AM, Carlo Musante 
> <>
>  wrote:
> 
>> Hi Andy,
>> 
>> I am seeing a problem with npad and the DiagServer.py process running 
>> constantly and shooting up the CPU to 100%. In checking for resolutions I 
>> found that you had asked for info from several users having the problem 
>> back in August of last year. I couldn't find a resolution posted.  Has 
>> something been fixed?
>> 
>> top - 09:57:14 up 61 days, 21:00,  1 user,  load average: 1.02, 1.04, 1.00
>> Tasks: 133 total,   3 running, 121 sleeping,   0 stopped,   9 zombie
>> Cpu(s):100.0%us,  0.0%sy,  0.0%ni,  0.0%id,  0.0%wa,  0.0%hi,  0.0%si,  
>> 0.0%st
>> Mem:   1034660k total,   994180k used,    40480k free,    43988k buffers
>> Swap:        0k total,        0k used,        0k free,   611504k cached
>> 
>> PID USER      PR  NI  VIRT  RES  SHR S %CPU %MEM    TIME+  COMMAND         
>>                          
>> 3413 npad      15   0 52092 3816 1192 S 98.8  0.4   2330:41 DiagServer.py  
>>         
>> 
>> If you need additional information or if there is a better forum to post 
>> this to, please let me know.
>> 
>> Thanks,
>> 
>> 
>> Carlo
>> 
>> -- 
>> http://security.wayne.edu 
>