0

While going through the AWR report found that the row lock waits on indexes and buffer busy waits are more (rest of the stats look normal).

Could someone tell me if we can interpret by the below if it (row lock waits & buffer busy waits) is the cause of slowness? Or the stats look ok based on the DB time.

Thank you for your time.

Can we reduce the row lock waits on indexes and reduce the buffer busy waits by further tuning of IX1_PCWAUDITLOG, IX1_PCWREQRESP & LOBSEG_NEW (image).

enter image description here

enter image description here

enter image description here

enter image description here

enter image description here

3
  • looks like your interconnect is heavily used - I have seen this on applications that do a lot of inserts on the same table and/or use select for update as feature for serialization. For testing you could try to set one node to be only only available and look if the application behaves better. – jmk Jan 16 '19 at 19:21
  • Thanks a lot for your reply, but can you help spot by what/which stats/sections from the AWR we can draw this. I could not find any spike related to interconnect. Please excuse if my interpretation is wrong and help me with your findings in detail if possible. – Dothertechie Jan 17 '19 at 13:18
  • A detailed analysis of the AWR report is well out of scope of a SE answer. The hint is are the global cache waits. I don't know your patch level there where some issues that where fixed during the lifetime of 11.2. Another hint are the SQLs ordered by elapsed time. Almost all the statements are insert and update statements. – jmk Jan 18 '19 at 20:12

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.