Highlighted
Super Contributor.
Super Contributor.
397 views

RCP. Queued CONNECT connections

Jump to solution

sometimes i have several nodes in 'Queued CONNECT connections' block from ovbbcrcp status output.

What does it meen?

 

If rcp serves about 20-30 nodes i see that any node could get in this list and quickly get out.

But if rcp serves 40 and more nodes i have problem that some nodes never get out from this list.

Why so?

Labels (1)
0 Likes
1 Solution

Accepted Solutions
Highlighted
Outstanding Contributor.. Outstanding Contributor..
Outstanding Contributor..

Hello,

 

You can try the following paramer changes at the RCP for better scalability:

> ovconfchg -ns bbc.http -set MAX_SOCKETSET_THREADS 11

> ovconfchg -ns xpl.net -set SOCKETS_PER_SOCKETSET 20

> ovconfchg -ns xpl.net -set SocketPoll TRUE  (Only for Linux or Solaris RCPs)

 

 

View solution in original post

5 Replies
Highlighted
Outstanding Contributor.. Outstanding Contributor..
Outstanding Contributor..

Hello,

 

You can try the following paramer changes at the RCP for better scalability:

> ovconfchg -ns bbc.http -set MAX_SOCKETSET_THREADS 11

> ovconfchg -ns xpl.net -set SOCKETS_PER_SOCKETSET 20

> ovconfchg -ns xpl.net -set SocketPoll TRUE  (Only for Linux or Solaris RCPs)

 

 

View solution in original post

Highlighted
Visitor.

Hello.

 

I not found in ovconfget part of settings "xpl.net". When command will is perfomed, it's adding part of settings and will correctly working?

Also I would like to know how much increase connected per node?

 

Thank you.

0 Likes
Highlighted
Outstanding Contributor.. Outstanding Contributor..
Outstanding Contributor..

Hello,

 

The fact that the namespace [xpl.net] does not appear in the output of `ovconfget`by default does not indicate it's not valid.

 

The variables mentioned above are documented in the Operations Agent 11.14 Reference Guide, e.g:

 

 

 

0 Likes
Highlighted
Micro Focus Expert
Micro Focus Expert

Hi,

 

Recommendation 1:

Not sure which version the agent is running on OM.

I recommend upgrade the version to 11.14.

11.12/11.13 had some known issues causing reconnections (hotfix available for these versions).

 

 

Recommendation 2:

Check your "ulimit" settings, and compare with  OML documentation for recommended values.

Have seen incidents, where "open files" settings were not matching recommended settings, causing problems in creating new sockets etc, leading to above problem.

 

Recommendation 3:

If the latest OM patch installed, you might have got the variable:

[xpl.net]

SocketPoll=TRUE

If this varaible is not set, please set it on OM Linux Or OM Solaris.

 

Recommendation 3:

Thanks Cristian.

The variables SOCKETS_PER_SOCKETSET and MAX_SOCKETSET_THREADS variables can be fine tuned on need basis.

(Also refer Item 2, if you are setting this value).

 

Regards,

M.K.Prasad

0 Likes
Highlighted
Super Contributor.
Super Contributor.

Hello, Christian.

 

Your solution is right.

In my case i have over 140 node being served by proxy.

 

But i dont understand each attribute influence.

1. In my case MAX_SOCKETSET_THREADS less than 11 doesn't resolve problem. why it must be 11 and more?

2. By default (according to documentation) SOCKETS_PER_SOCKETSET is equal 60. Why do i need reduce this attribute value?

 

And as i understand this parameters is global. it must affect anb bbc too (not rcp only). Am i right?

The opinions expressed above are the personal opinions of the authors, not of Micro Focus. By using this site, you accept the Terms of Use and Rules of Participation. Certain versions of content ("Material") accessible here may contain branding from Hewlett-Packard Company (now HP Inc.) and Hewlett Packard Enterprise Company. As of September 1, 2017, the Material is now offered by Micro Focus, a separately owned and operated company. Any reference to the HP and Hewlett Packard Enterprise/HPE marks is historical in nature, and the HP and Hewlett Packard Enterprise/HPE marks are the property of their respective owners.