Abrupt/Sudden Increases in Server Load Issues by XINETD Processes !!

[Jai Singh VistA]

Hi Everyone,

We are using mgwsi + ewd + apache architecture for our application. Also, the OS used is RHEL 7.5.

 Frequently, We are facing an issue related to abrupt server load increases to more than 1000.

In general, there is 300 to 500 xinetd process that exists in our server.

Whenever multiple processes of mgwsi get started, it starts creating more than 1,000 - 15,000 mgwsi-xinetd processes from 127.0.0.1. Which causes the sudden server load increase to more than 100 to 1000.

To normalize it always we needs to stop the entire process and restart the Apache services.

Below is the code mgwsi service created under /etc/xinetd/ directory.

1. mgwsi service - defined in /etc/xinetd/ directory

service mgwsi-rpcproc-7041
{
        port        = 7041
        socket_type = stream
        protocol    = tcp
        type        = UNLISTED
        user        = prd-ser
        server      = /bin/bash
        server_args = /opt/worldvista/prd/ser/mgwsi7041
        wait        = no
        disable     = no
        flags       = REUSE
        per_source  = UNLIMITED
        instances   = UNLIMITED
        cps         = 300 10

}

2. Logs from /var/logs/messages 

Oct 11 01:21:49 localhost xinetd[77487]: START: mgwsi-rpcproc-7041 pid=61325 from=::ffff:127.0.0.1
Oct 11 01:21:49 localhost xinetd[77487]: START: mgwsi-rpcproc-7041 pid=61326 from=::ffff:127.0.0.1
Oct 11 01:21:49 localhost xinetd[77487]: START: mgwsi-rpcproc-7041 pid=61348 from=::ffff:127.0.0.1
Oct 11 01:21:49 localhost xinetd[77487]: START: mgwsi-rpcproc-7041 pid=61358 from=::ffff:127.0.0.1
Oct 11 01:21:49 localhost xinetd[77487]: START: mgwsi-rpcproc-7041 pid=61359 from=::ffff:127.0.0.1
Oct 11 01:21:49 localhost xinetd[77487]: START: mgwsi-rpcproc-7041 pid=61375 from=::ffff:127.0.0.1
Oct 11 01:21:49 localhost xinetd[77487]: START: mgwsi-rpcproc-7041 pid=61384 from=::ffff:127.0.0.1
Oct 11 01:21:49 localhost xinetd[77487]: START: mgwsi-rpcproc-7041 pid=61385 from=::ffff:127.0.0.1
Oct 11 01:21:49 localhost xinetd[77487]: START: mgwsi-rpcproc-7041 pid=61414 from=::ffff:127.0.0.1
Oct 11 01:21:49 localhost xinetd[77487]: START: mgwsi-rpcproc-7041 pid=61422 from=::ffff:127.0.0.1
Oct 11 01:21:49 localhost xinetd[77487]: START: mgwsi-rpcproc-7041 pid=61423 from=::ffff:127.0.0.1
Oct 11 01:21:49 localhost xinetd[77487]: START: mgwsi-rpcproc-7041 pid=61425 from=::ffff:127.0.0.1
Oct 11 01:21:49 localhost xinetd[77487]: START: mgwsi-rpcproc-7041 pid=61426 from=::ffff:127.0.0.1
Oct 11 01:21:49 localhost xinetd[77487]: START: mgwsi-rpcproc-7041 pid=61427 from=::ffff:127.0.0.1
Oct 11 01:21:49 localhost xinetd[77487]: START: mgwsi-rpcproc-7041 pid=61428 from=::ffff:127.0.0.1
Oct 11 01:21:49 localhost xinetd[77487]: START: mgwsi-rpcproc-7041 pid=61443 from=::ffff:127.0.0.1
Oct 11 01:21:49 localhost xinetd[77487]: START: mgwsi-rpcproc-7041 pid=61444 from=::ffff:127.0.0.1
Oct 11 01:21:49 localhost xinetd[77487]: START: mgwsi-rpcproc-7041 pid=61445 from=::ffff:127.0.0.1
Oct 11 01:21:49 localhost xinetd[77487]: START: mgwsi-rpcproc-7041 pid=61446 from=::ffff:127.0.0.1
Oct 11 01:21:49 localhost xinetd[77487]: START: mgwsi-rpcproc-7041 pid=61447 from=::ffff:127.0.0.1
Oct 11 01:21:49 localhost xinetd[77487]: START: mgwsi-rpcproc-7041 pid=61448 from=::ffff:127.0.0.1
Oct 11 01:21:49 localhost xinetd[77487]: START: mgwsi-rpcproc-7041 pid=61449 from=::ffff:127.0.0.1
Oct 11 01:21:49 localhost xinetd[77487]: START: mgwsi-rpcproc-7041 pid=61456 from=::ffff:127.0.0.1
Oct 11 01:21:49 localhost xinetd[77487]: START: mgwsi-rpcproc-7041 pid=61457 from=::ffff:127.0.0.1
Oct 11 01:21:49 localhost xinetd[77487]: START: mgwsi-rpcproc-7041 pid=61462 from=::ffff:127.0.0.1

[rtweed]

I suggest you contact me and pay for support, since you're clearly using EWD and mgwsi for a large production, business-critical platform.

Email me for details - rob dot tweed at gmail dot com

Rob

[rtweed]

Still awaiting an email from you to provide support for your system