Monday, March 26, 2012

Replication Distributor Job History is shocking!

Sorry for the odd format but that is what happens when you paste here! This is data from our Job History for a "Replication Distributor" Job Step Type. Here is how I interprete it: The job step started on the 19th. It restarted itself twice on the 20th. Then it restarted itself twice on the 21st. Then it restarted itself 3 times on the 22nd. And finally it restarted itself on the 23rd. Notice the progression of Retries_Attempted. When Retries_Attempted exceeds 10, the step will fail and the service will shut-down, which is what I have seen in the past. I conclude the Advanced Settings of Retry Attempts is not per "Event" but "total tries no matter how many successful retries are encountered." I thought a successful retry would reset the Retries_Attempted. But it does not. Is this correct? This implies that to keep the distributor up, we should set the Retry Attempts" setting to its max!

Status Run_Date Run_Time Run_Duration Retries_attempted

1

1/19/2007

44025

0

0

2

1/20/2007

31200

1429

1

2

1/20/2007

32729

234250

2

2

1/21/2007

31119

1558

3

2

1/21/2007

32817

234225

4

2

1/22/2007

31142

1513

5

2

1/22/2007

32755

508

6

2

1/22/2007

33403

233615

7

2

1/23/2007

31118

1503

8

2

1/23/2007

32721

501

9

Topping this one so I can remember it, will look into it.

No comments:

Post a Comment