#StackBounty: #sql-server #high-availability #sql-server-2019 #alwayson #availability-group SQL Server :: Replication Distribution Agen…

Bounty: 50

I’m running SQL Server 2019 Always ON Availability Group with an asynchronous replication.

I use a free tool called IDERA SQL Check and I have spotted the SPID 69 which program name is Replication Distribution Agent. It’s always there, staring at me like a bored cat.

enter image description here

This SPID 69 is pointing to a specific database which is mirrored I investigated it with this the query:

select 
s.session_id
,login_name
,login_time
,host_name
,program_name
,status
,cpu_time
,memory_usage
,total_scheduled_time
,total_elapsed_time
,last_request_start_time
,reads
,writes
,logical_reads
from sys.dm_exec_sessions s
inner join sys.dm_exec_connections c
on s.session_id = c.session_id
outer apply sys.dm_exec_sql_text(c.most_recent_sql_handle) st
where s.is_user_process = 1
and s.open_transaction_count > 0;

Which gave me this response:

  • session_id = 69
  • text = begin tran
  • login_time = 2020-09-08 18:40:57.153
  • program_name = Replication Distribution Agent
  • status = sleeping
  • cpu_time = 1362772
  • memory_usage = 4
  • total_scheduled_time = 1689634
  • total_elapsed_time = 22354857
  • last_request_start_time = 2020-09-28 16:28:39.433
  • reads = 18607577
  • writes = 5166597
  • logical_reads = 112256365

Now, on internet I find that when you see Replication Distribution Agent is all good, that agent should be going and there should be no problem. But why:

  1. The text says begin tran and nothing more?
  2. IDERA SQL Check is labelling it as connection idling transaction?
  3. The status is sleeping?
  4. I’m concerned that CPU time, reads and writes are basically telling me that this process is frying the drive with never ending I/O, am I right?


Get this bounty!!!

#StackBounty: #vmware-vsphere #high-availability #fault-tolerance VMWare Fault Tolerance vLockStep

Bounty: 100

I would like to understand how vLockStep is working and could only find very high level description.

The current documentation say that disk read but not disk write performed by the primary are replicated to the secondary. What is the reason for that ?

Also non-deterministict cpu instruction are also logged and replicated.
I can only assume it’s to cause the execution on the secondary to not execute those “non-deterministic instruction” and instead sync the effect of calling it from primary to secondary.

If this is the case, this would mean the Primary need to wait for the secondary to finish (executing/syncing) the non-deterministic instruction before sending the response to a network request. This seem like this would add significant latency.

Am I completely wrong in my understanding, or is this more or less correct request latency on the primary would be significantly slower than when not using vLockstep?


Get this bounty!!!