With PSA Replica and parameter enableMajorityReadConcern false ,Secondary becomes blacklisted

[saurav gupta]

Hi All,

With P-S-A Replica  architecture with enableMajorityReadConcern false, bulk inserts and reads are going on.

But after one hour or so, replication lags increases and keeps on increasing and ultimately Secondary node goes down and becomes blacklisted

Please recommend correct configuration for P-S-A Replica  architecture and  how to recover blacklisted node to secondary server

Regards,

Saurav

[MH]

Can you share your infrastructure, output from RS.config redacted as needed, version numbers, and what you're seeing in the logs? Bandwidth between nodes? Versions? I have a number of PSA nodes out in production and am not seeing this.

[saurav gupta]

Hi All,

Below are the logs and parameters :

2020-01-29T17:21:47.124+0530 I REPL [replication-0] We are too stale to use 10.95.147.92:27017 as a sync source. Blacklisting this sync source because our last fetched timestamp: Timestamp(1580295549, 1) is before their earliest timestamp: Timestamp(1580297909, 30285) for 1min until: 2020-01-29T17:22:47.124+0530 

replication:
   replSetName: rs1
   enableMajorityReadConcern: false

Regards,

Saurav

On Thu, Feb 20, 2020 at 5:11 AM MH <disgus...@gmail.com> wrote:

Can you share your infrastructure, output from RS.config redacted as needed, version numbers, and what you're seeing in the logs? Bandwidth between nodes? Versions? I have a number of PSA nodes out in production and am not seeing this.

--
You received this message because you are subscribed to the Google Groups "mongodb-user"
group.

For other MongoDB technical support options, see: https://docs.mongodb.com/manual/support/
---
You received this message because you are subscribed to the Google Groups "mongodb-user" group.
To unsubscribe from this group and stop receiving emails from it, send an email to mongodb-user...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/mongodb-user/2125d1b1-4957-4268-9a20-ea8846c02023%40googlegroups.com.

[saurav gupta]

[MH]

based on that log message I believe your node is too stale and will need to be synced fresh

https://docs.mongodb.com/manual/tutorial/resync-replica-set-member/

perhaps this members cant get a copy of all of the data due to the oplog not being large enough to store enough data that the secondary can sync to before the oldest entry in the op log falls off. maybe your primary needs a larger op log?

[Stennie]

This question was asked & answered on the MongoDB Community site: PSA Architecture with enableMajorityReadConcern false- Secondary down.

It would be best to continue any discussion on the MongoDB Community site, which is the replacement for this Google Group.

Thanks,

Stennie