Fixing EMC VNX 5×00 naviagent not starting failover
Fixing EMC VNX 5300 naviagent not starting on failover control station when the failover actually occured.
** from emc eservice doc
The following is a Primus(R) eServer solution: ID: emc262797 Domain: EMC1 Solution Class: 3.X Compatibility Goal How to correct Naviagent errors when Control Station is failed over to CS1 Fact Product: Celerra Fact Product: VNX Series Fact Product: VNX Unified/File Fact Product: VNX VG8 Fact Product: VNX VG2 Fact EMC SW: NAS Code 6.x Fact EMC SW: NAS Code 7.x Fact EMC SW: NAS Code 5.6.50 and later Symptom Naviagent messages unexpectedly exited logged in /var/log/messages every five minutes for Navisphere agent "cannot create messages directories" Symptom Naviagent errors occurs when Control Station is failed over to CS1. Symptom Call home message: Slot 1 becomes primary because of timeout. Symptom /var/log/messages CS_PLATFORM:NaviEventMonitor:CRITICAL:4:::::CLARiiON event number 0x0 to /etc/Unisphere/backup/navimon.log. ExtCode1=107373568<http://remedytools.rtp.lab.emc.com/nsgview.pl?entry=107373568>,; ExtCode2=log. Symptom /var/log/messages show the following errors: Feb 11 11:22:19 emcnas_i0 EMCServer: nas_mcd: daemon "Apache daemon" unexpectedly exited (status = 0) ifexit=1 exitstatus=0 ifsignal=0 termsig=0 ifstop=0 stopsig=0 ifdump=0 Feb 11 11:22:19 emcnas_i0 EMCServer: nas_mcd: daemon "Naviagent" unexpectedly exited (status = 11) ifexit=0 exitstatus=0 ifsignal=1 termsig=11 ifstop=0 stopsig=0 ifdump=0 Feb 11 11:22:19 emcnas_i0 login(pam_unix)[27969]: session closed for user nasadmin Feb 11 11:22:22 emcnas_i0 Navisphere Agent[31217]: Error: Cannot create messages directory /etc/Unisphere/messages/ Feb 11 11:22:22 emcnas_i0 Navisphere Agent[31217]: Error: Cannot create event log backup directory /etc/Unisphere/backup/ Feb 11 11:22:22 emcnas_i0 Navisphere Agent[31217]: Open application log failed Error re-occurs after five minutes: Feb 11 17:27:34 emcnas_i0 Navisphere Agent[1287]: Error: Cannot create messages directory /etc/Unisphere/messages/ Feb 11 17:27:34 emcnas_i0 Navisphere Agent[1287]: Error: Cannot create event log backup directory /etc/Unisphere/backup/ Feb 11 17:27:34 emcnas_i0 Navisphere Agent[1287]: Open application log failed Cause Installations at these code levels don't create the symbolic link for the Naviagent service to start properly on CS1. The symbolic links created from installation point at /etc/Navisphere instead of /etc/Unisphere on CS1, which is what the naviagent service is looking for on start-up. Fix Create the symbolic links and directories while the Control Station is failed over to CS1: 1. ln -s /nas/etc/Navisphere/ /etc/Unisphere/ 2. mkdir /etc/Unisphere/messages 3. mkdir /etc/Unisphere/backup Monitor for naviagent service to start on next 5-minute interval: $ ps -ef | grep -i navia root 9198 2831 0 Feb11 ? 00:00:02 /nas/opt/Navisphere/bin/naviagent -d -f /nas/etc/Navisphere/agent.config -r /nas/log Note CS1 has been re-installed and problem persists.
Howto remove errors alerts in EMC VNX unisphere GUI
As you probably already know, you can’t delete an “error” alert in EMC VNX unisphere GUI.
This thing needs to be done with the CLI on the control station.
So here is howto do that.
ssh nasadmin@[vnx cs]
>/nas/log/webui/alert_log
The error should be removed, if not logout of Unisphere and log back in.
EMC – Howto set IP alias on primary control station
This is used to set the IP address that will move with the primary control station. That way you will always be on the primary CS if you connect to that particular IP address, it is an easy way of knowing that you will be on the primary.
- Log on to the CS as root
- Create the IP alias for the CS
/nas/sbin/nas_conrig -IPalias -create 0 - Answer no to the question “Do you want slot_0 IP address x as your alias.
This will allow you to set another IP that points to the active unit, and still be able to talk to the standby unit. - Then specify the IP
- Then view what you did with :
/nas/sbin/nas_config -IPalias -list
