Quantcast
Channel: Data Protector Practitioners Forum topics
Viewing all articles
Browse latest Browse all 3189

DataProtector 9 Storeonce Network Sensitivity

$
0
0

Hi,

We are running DP 9.03 on our Windows Server 2012 R2 Cell Manager. All Media Agents are also upgraded to 9.03 and also Windows Server 2012 R2.

Recently we have had some very brief network outages between our main datacentres, the outages last about 1 - 2 seconds. During these any Data Protector jobs going over the WAN will fail. We just had a network drop where we only missed one ping between our datacentres, however an active job in Data Protector failed. Below is what occurred in the Session Log:

 

--------------------------------------

 [Major] From: BMA@Site2-Server01.Domain.com "Site1-StoreOnce01-Veeam_gw4 [GW 7380:0:15697104232020443816]"  Time: 15/06/2015 1:06:52 PM


[90:51]                  \\Site1-STOREONCE01\Site1Veeam\042c3c9f_557cc22d_0828_0375
                Cannot write to device (StoreOnce error: StoreOnce device offline, network error occurred or secure communication failed while contacting the StoreOnce device)

 

[Critical] From: VBDA@Site2-Server01.Domain.com "Site2-Backup-Veeam-Prod-Exchange2013-All-Weekly"  Time: 15/06/2015 1:06:52 PM


                Unexpected close reading NET message => aborting.

 

[Critical] From: VBDA@Site2-Server01.Domain.com "Site2-Backup-Veeam-Prod-Exchange2013-All-Weekly"  Time: 15/06/2015 1:06:52 PM


                Connection to Media Agent broken => aborting.

--------------------------------------

 

This job was backing up files sitting on a Storeonce NAS share in Site2 to a Storeonce catalyst in Site1.

 

I can see the following values in the OMNIRC on the Cell Manager:

 

#  OB2RECONNECT_RETRY=s
#    Default: 600
#    Time-out (in seconds) before either a Disk Agent (DA) or Media Agent (MA)
#    should stop trying to reconnect to the Session Manager or Disk Agent
#    should stop trying to reconnect to the Media Agent during backup.
#
#  OB2RECONNECT_ACK=s
#    Default: 1200
#    Timeout (in seconds) before an agent gives up waiting to acknowledge a
#    message from the session manager.

 

These values have not been defined on the Cell Manager, so the defaults should be taking effect. Are these the only settings that control session timeouts due to a loss of network?

 

It seems the connectivity between the Cell Manager and the Storeonce devices is very sensitive, is there any way to make it less so?

 

Cheers
Wal


Viewing all articles
Browse latest Browse all 3189

Trending Articles



<script src="https://jsc.adskeeper.com/r/s/rssing.com.1596347.js" async> </script>