This post is more than 5 years old
3 Posts
0
3070
April 11th, 2011 07:00
Celerra NTP Error
Hello All,
Recently, we had a Celerra Dual Blade NX4 installed. Initially, we received NTP errors for Time Synchronization. Our network guru configured a workaround to allow the isolated iSCSI VLAN to communicate to our NTP server. Now, server_2 appears to be synching time correctly but, server_3 is not able to sync time, nor ping the gateway. It appears this is because server_3 is a standby datamover as opposed to the primary datamover server_2. Has anyone else encountered this issue? If so, what was the workaround to ensure NTP is synchronized for both primary/standby datamovers? Would configuring both datamovers as primary/primary solve this issue?
Thank you for your time,
Chris
Rainer_EMC
4 Operator
•
8.6K Posts
0
April 11th, 2011 11:00
Hi,
just ignore any NTP warning for standby data movers.
On failover a standby data mover will take over the addresses and config of the failing one – so if your primary has a good working NTP and the network cabling on your standby is good it will be fine with NTP after a failover
Rainer
cholcomb
3 Posts
0
April 11th, 2011 11:00
Hello Rainer,
Thanks for the repsonse. I had suspicions that the NTP server sync issue was not easily correctable but, I still recevie daily alarms for "critical" NTP failures. Short of removing the error messages permantly from email notifications, I would assume there is a workaround to stop the server from attempting to sync???
Thanks again for the response.
Chris
nandas
4 Operator
•
1.5K Posts
0
April 11th, 2011 12:00
What type of error messages you receive? You may provide a sample copy here.
You may check whether the NTP server is accessible from the data mover
$ server_ping server_2
Also, You may please also run the following -
$ server_date server_2 timesvc stats ntp
This will show the NTP status on the data mover and whether the NTP requests are getting hit or miss.
And as Rainer mentioned earlier, the standby data mover is not doing any network activity - when it becomes active, it assumes the role of server_2 and get all the network configuration for server_2. So no need to worry about standby data mover configuration - but must have identical Network connectivity and switch configuration for bothe server_2 and server_3.
My 2 cents
Sandip
Rainer_EMC
4 Operator
•
8.6K Posts
1
April 12th, 2011 08:00
I don't think so
Open a case with support to check if this is maybe fixes in a newer release
cholcomb
3 Posts
0
April 13th, 2011 13:00
Thanks for the help guys. I am working thru EMC channels right now for a workaround or something close to a resolution to get the errors to stop. I dont see much sense in having logs fill up with error messages that shouldn't exist.
BTW, server_2 shows Hits for timesync with no misses. I validated this before originally posting.
Thanks again
glparker_e38cb2
58 Posts
0
July 31st, 2011 23:00
Hi cholcomb,
Did you get a resolution to this issue. I'm experiencing exactly the same issue as you on my VNX5300 with dual blades. Server_2 can see and resolve the NTP server, but server_3 can't and it keeps logging timesync alerts.
Regards,
George.
Rainer_EMC
4 Operator
•
8.6K Posts
0
August 1st, 2011 06:00
Is your server_3 a standby ?
What OS release are you using ?
glparker_e38cb2
58 Posts
0
August 1st, 2011 16:00
Yep, my server_3 is a standby.
I'm using VNX OE for File v7.0.14.0 and VNX OE for Block v05.31.000.5.011.
George.