- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page
Question about persistency in DTC
- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
07-26-2021 01:53 PM
This in bit urgent. What's the maximum time can be configured for persistency in DTC. I am looking for persistency or stickiness for 24 hours, is it possible somehow?
Also, I see it is currently seems to be maximum 2 hours but is it renewed each time user reconnects within the persistant time configured.
Re: Question about persistency in DTC
- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
07-27-2021 01:17 PM
Hi,
This is from the official documentation:
You can specify a period between one second to 2 hours. Even if the DNS restart takes longer than the value specified in the Persistence field, the DNS server directs the request to the same server. If you specify zero, the appliance does not cache the requests. When you enable persistence for an LBDN, the appliance stores the results for specific LBDN responses in the DNS Traffic Control cache. When a request originates from the respective FQDN or an IP address within the specified period, the DNS server directs the request to the same server.
Regards,
Krishna Vasudevan
Re: Question about persistency in DTC
[ Edited ]- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
07-29-2021 10:59 AM - edited 06-27-2023 01:38 PM
Hello blox_user,
To the best i know, the persistence value isn't renewed upon subsequent queries which matches the same LBDN pattern(I could be wrong, but I'm taking regular bind cache into account for comparison). I see where your concern is & I think one of the new load balancing methods introduced by Infoblox recently could be handy :
From this link, can you please read if "Source IP Hash" load balancing method would be helpful to you ? I haven't tried this specific LB method, but sounds like that'd help you maintain the *same* response back to the *same* client as long as that specific DTC-server's *health* is 'GREEN' within the health_status.txt ~ record of health!
Best regards,