bloomberg/comdb2

Respect maxquerytime during parallel sql execution

Opened this issue · 0 comments

Bug Description:

During parallel execution, the sqlengines executing in parallel don't respect maxquerytime set by client

Steps to reproduce
SET maxquerytime 3
select sleep(6) union all select sleep(6)
Client returns rc -107 (query timeout) but the individual shards continue to execute beyond 3 secs , verified by long request trace

Expected behavior
Individual shards should timeout as well

Additional context:
requires dohast_disable 0, dohsql_disable 0