Skip to Main Content

ttRepAdmin show high latency

3206761Mar 26 2016 — edited Mar 31 2016

Hi,

1- when looking to the ttRepAdmin output i see high number in latency , which i guess is the reason why the logs accumulation start increasing .

# ttRepAdmin -dsn db1 -receiver -list

Peer name         Host name                 Port    State  Proto Track

----------------  ------------------------                ------     ------- ----- -----

DB_standby            GeoDB_040          14550  Start      38     0


LastMsgSent       LastMsgRecv   Latency      TPS     RecordsPS      Logs

-------------                ------------          -------         -------     ---------            ------------

00:00:00            00:00:00             507.30       1544     1956             31

2- Checking the Network i see the Send-Q in netstat command is releasing quickly but always stay on high numbers too

(Send-Q will be that data which the sending application has given to the transport, but has yet to be Acknowledged by the receiving TCP)

Proto Recv-Q Send-Q Local Address               Foreign Address                            State       PID/Program name

tcp      120 147696      localDB_address:35332       GeoDB_040:14550              ESTABLISHED 1678/timestenrepd  

tcp        0    120          localDB_address:14550          GeoDB_040:47707              ESTABLISHED 1678/timestenrepd  

3- during the issue also check inserting manually data in replication tables and some time , it return with error for time exceed  to commit the change ( asynchronous replication )

4- Checking the Network it provide is is healthy and no latency in the network side.

5- I also checked the NTP but i do not see any problem there

6- I have another DB with same location and IP network setting as the two above working just fine with out latency

7- nothing up normal with CPU or memory

8- TimesTen: Release: 11.2.2.8.2

9- The issue appear only when server  (localDB_address) getting high traffic

10- after the high traffic finished replication  logs start to decreases slowly and then get back to normal and no more latency appear.

11- There is possibility the number of the non replicated logs reach to threshold which 256 files and then the replication break,, and it did happens couples of time already.

12- FYI, when i do restart the replication during the high traffic is actually help to reduce the latency.

My question here will be :-

1- is there is way i can improve the TT replication latency?

2- As restart the replication demon during the problem i guess there is something went wrong with it, but still what else i should check during the traffic to see if it application issue or TT Replication daemon issue.?

Any help will be appreciated or refer me to documentation

Many Thanks

This post has been answered by ChrisJenkins-Oracle on Mar 31 2016
Jump to Answer
Comments
Locked Post
New comments cannot be posted to this locked post.
Post Details
Locked due to inactivity on Apr 28 2016
Added on Mar 26 2016
6 comments
1,369 views