MySQL Bugs: #36843: Unoptimal handling of partially connected API in SUB_START

Bug #36843	Unoptimal handling of partially connected API in SUB_START_REQ
Submitted:	21 May 2008 6:46	Modified:	30 May 2008 14:05
Reporter:	Jonas Oreland	Email Updates:
Status:	Closed	Impact on me:	None
Category:	MySQL Cluster: Cluster (NDB) storage engine	Severity:	S3 (Non-critical)
Version:	*	OS:	Any
Assigned to:	Jonas Oreland	CPU Architecture:	Any

Description:
When receiving execSUB_START_REQ (which corresponds to NdbEventOperation.execute())

SUMA *at end* checks that node is properly connected.

Given that a execSUB_START_REQ can consume lots of CPU (if it's the first subscriber on the subscription) clever change would be to check this already at start.

How to repeat:
.

Suggested fix:
.

A patch for this bug has been committed. After review, it may
be pushed to the relevant source trees for release in the next
version. You can access the patch from:

  http://lists.mysql.com/commits/46906

ChangeSet@1.2207, 2008-05-21 09:02:12+02:00, jonas@perch.ndb.mysql.com +2 -0
  ndb - bug#36843
    Check connectivity also in beginning of SUB_START_REQ
    to avoid wasting resources with setup/teardown of subscription

A patch for this bug has been committed. After review, it may
be pushed to the relevant source trees for release in the next
version. You can access the patch from:

  http://lists.mysql.com/commits/46907

ChangeSet@1.2598, 2008-05-21 09:18:45+02:00, jonas@perch.ndb.mysql.com +2 -0
  ndb - bug#36843 (telco-6.*)
    check connectivity also at start of SUB_START_REQ

Pushed into 5.1.23-ndb-6.4.0

Pushed into 5.1.24-ndb-6.3.15

Pushed into 5.1.24-ndb-6.2.16

Documented in the 5.1.24-ndb-6.2.16 and 5.1.24-ndb-6.3.15 changelogs as follows:

        Checking of API node connections was not efficiently handled.

Pushed into 6.0.6-alpha  (revid:sp1r-jonas@perch.ndb.mysql.com-20080521071845-22292) (version source revid:jonas@mysql.com-20080808094047-4e1yiarqa2t3opg3) (pib:5)