connpool | Connection pool for Go 's net.Conn interface

Issue got fixed. Primary database was able to connect to secondary hosts but in return secondary host can't. So I adjusted the security groups and connected. Takeaway is all the primary and secondary hosts should be able to connect with each other in mongodb port (27017)

I think you are looking for a thread-safe implementation of a Singleton class here. Of course, the way to achieve this is using enums, but for your case, you can implement double-check locking to ensure no two threads can call new MyPool() simultaneously.

Also, I think in your code, you are actually implementing a factory class, not really a singleton. Your MyPool is a different class than Singleton, which could have a public constructor.

I have made the appropriate changes with comments.

Double check-locking basically just checks the thread-safety before and after null-check, because the whole method is not synchronized, so two threads can indeed get the null value in the condition even after synchronization block, hence the second synchronization.

Also, I think your getPool() should be static. You won't be able to call getPool() without an explicit object of Singleton, which I think you don't need.

Corrected version:

The secondary node was unable to become primary because the arbiter voted no, which might have been related to it being unable to resolve the DNS name:

2020-07-14T14:06:39.449+0300 I ELECTION [replexec-1151] VoteRequester(term 2) received a no vote from srd01-arbiter:27017 with reason "can see a healthy primary (srd01-primary:27017) of equal or greater priority"; response message: { term: 2, voteGranted: false, reason: "can see a healthy primary (srd01-primary:27017) of equal or greater priority", ok: 1.0 }

The arbiter voted no because it wasn't yet aware that the primary was shutting down:

2020-07-14T14:06:39.448+0300 I ELECTION [conn10] Sending vote response: { term: 2, voteGranted: false, reason: "can see a healthy primary (srd01-primary:27017) of equal or greater priority" }

2020-07-14T14:07:01.022+0300 I NETWORK [ftdc] getaddrinfo("srd01-primary") failed: Temporary failure in name resolution

2020-07-14T14:07:01.024+0300 I REPL [replexec-1395] Member srd01-primary:27017 is now in state RS_DOWN - Couldn't get a connection within the time limit

Suggestions:

• Run the replSetStepDown command and wait for a new primary to be elected before shutting down the primary's container.
• Check the primary logs to see if it tried to step down prior to shutting down.
• Don't use arbiters unless absolutely required

Issue has been fixed.

I just needed to remove conflicting custom imports and re-install Google Vision library.

Well that was a dumb mistake, sorry to bother you.

I think you are consuming the OResultSet outside the Try block, this means that the session is closed and the rs can't fetch more data from the database anymore.

According to the analysis displayed in the bottom, the port is not open for the rest of the Ips.

Make sure the Firewall rules are enabling this. To do

• configure the firewall rule in Google cloud console
• provide a tag in your firewall rule
• tag your instance with the same tag as the firewall rule

Your sysman container should connect to the database container using sysman-db:3306 (not 4000). Remember that it is communication between 2 services so the published ports don't count.