A "ps -ef | grep vx" shows there are several vxnotify processes running

book

Article ID: 100016540

calendar_today

Updated On:

Resolution

Consider a sample output as below

root   502     1  0  Sep 11 ?        0:00 /sbin/sh -/usr/lib/vxvm/bin/vxconfigbackupd
root   500    1  0   Sep 11 ?        0:00 /sbin/sh -/usr/lib/vxvm/bin/vxrelocd root
root   501    1  0   Sep 11 ?        0:00 /sbin/sh -/usr/lib/vxvm/bin/vxcached root
root   547   500  0   Sep11 ?        0:00 /sbin/sh - /usr/lib/vxvm/bin/vxrelocdroot
root   531   530  0   Sep 11?        0:00 vxnotify -C -w 15
root  530   501  0   Sep 11 ?        0:00 /sbin/sh- /usr/lib/vxvm/bin/vxcachedroot
root  1006  1005  0   Sep 11?        0:00 vxnotify
root  1005  502  0   Sep 11 ?        0:00 /sbin/sh -/usr/lib/vxvm/bin/vxconfigbackupd
root   548  547  0   Sep 11 ?        0:00 vxnotify -f -w15
root 10974  6153  0   Sep 11?        0:00  /usr/sbin/vxnotify -gsharedg

From above output

1) The  vxnotify -f -w15 is started by vxrelocd to display plex, volume, and disk detach events.

2) The vxnotify -C -w 15 is started by vxcached so it can grow cache volumes used by space-optimized snapshots if required.

3) The vxnotify (no arguments) is started by vxconfigbackupd so it can be notified of configuration changes and therefore back up the new configuration.

4) The  vxnotify -g (where <> is not part of the name) is started when Veritas Cluster Server brings a CVMVolDg resource for that diskgroup online.

 
 

 

Issue/Introduction

A "ps -ef | grep vx" shows there are several vxnotify processes running, however each instance of the process signifies a particular vxvm action.