Re: AWS RDS Detecting failover
Date: Thu, 19 Oct 2017 14:19:56 -0400 (EDT)
Message-ID: <372665825.13983286.1508437196685.JavaMail.zimbra_at_earlham.edu>
Hi Steve
Sorry for the delay in followup.
Here's what I have from the team:
The general failover process is described in http://docs.aws.amazon.com/AmazonRDS/latest/UserGuide/Concepts.MultiAZ.html , under "Failover Process for Amazon RDS".
The general rule is that the application need to set a short TTL on the DNS cache on the application side, so that your database quickly notices changes to the underlying IP address for your database's DNS name.
Doing this in Java is described here:
http://docs.aws.amazon.com/sdk-for-java/v1/developer-guide/java-dg-jvm-ttl.html
Event Service will solve the delay in sending SNS notifications we can have now, which will come soon in Q4. At present, we make a best effort to send events. The ordering and timing of notifications cannot be guaranteed, and thus are not good replacement for having the proper DNS strategy. However, they should usually arrive faster than 8 minutes.
I'm hoping this helps. Also the team went and answered the forum post as well. Please let me know if I can do anything else to assist, or if you have followup questions for the team.
Regards,
Jeremiah
From: "stbaldwin" <stbaldwin_at_msts.com> To: "Jeremiah Cetlin Wilton" <jcwilton93_at_earlham.edu> Cc: "Oracle Mailing List" <Oracle-L_at_freelists.org> Sent: Wednesday, October 18, 2017 11:41:37 PM Subject: Re: AWS RDS Detecting failover
Hi Jeremiah,
Did you hear anything back from the Multi-AZ folks? I have confirmed this 'issue' is not Oracle specific. It also occurs with Postgres failovers. It got me wondering if the issue would still present in a 'real' failover (i.e. catastrophic failure of an AZ) or if maybe this 'issue' is related to the simulated 'reboot-with-failover' implementation.
Thanks,
Steve
From: Jeremiah Cetlin Wilton <jcwilton93_at_earlham.edu>
Sent: Thursday, 5 October 2017 8:55:05 AM
To: Steve T. Baldwin
Cc: Oracle Mailing List
Subject: Re: AWS RDS Detecting failover
Hi all, let me check with the Multi-AZ folks and get back to you.
Thanks,
Jeremiah
From: "Steve T. Baldwin" <stbaldwin_at_msts.com> To: "Oracle Mailing List" <Oracle-L_at_freelists.org> Sent: Tuesday, October 3, 2017 1:06:16 PM Subject: Re: AWS RDS Detecting failover
For information, here's an AWS forum post ( [ https://forums.aws.amazon.com/thread.jspa?messageID=689877 | https://forums.aws.amazon.com/thread.jspa?messageID=689877 ] ) from around 2 years ago where someone asks about the lag in publishing DB events.
The forum post was never answered and the support analyst responding to my query couldn't give an answer to me asking if/when this would be improved.
Steve
This email is intended solely for the use of the addressee and may contain information that is confidential, proprietary, or both. If you receive this email in error please immediately notify the sender and delete the email.
This email is intended solely for the use of the addressee and may contain information that is confidential, proprietary, or both. If you receive this email in error please immediately notify the sender and delete the email.
-- http://www.freelists.org/webpage/oracle-lReceived on Thu Oct 19 2017 - 20:19:56 CEST