tag:blogger.com,1999:blog-15622618.post8775830487953822132..comments2011-09-30T14:41:47.782-07:00Devananda vdvhttp://www.blogger.com/profile/03378341125039609409noreply@blogger.comBlogger2125tag:blogger.com,1999:blog-15622618.post-50294221810477568402011-09-30T14:41:47.782-07:002011-09-30T14:41:47.782-07:00AT-HE,<br /><br />Blindly skipping replication errors is a sure way to get the replication slave out of sync with the master. While this may get replication running again, the slave&#39;s data will probably become different from the master&#39;s data because the skipped statement did not run on the slave. Furthermore, depending on the particular event being skipped, you might not solve anything at all -- the very next statement may fail because it depended on the statement you skipped!<br /><br />Moral of the story: don&#39;t use SQL_SLAVE_SKIP_COUNTER unless you really, really know what you&#39;re doing!Devananda vdvhttp://www.blogger.com/profile/03378341125039609409noreply@blogger.comtag:blogger.com,1999:blog-15622618.post-46274575008166383042011-09-30T14:22:48.579-07:002011-09-30T14:22:48.579-07:00i readed somewhere that by skipping the error causing slave subprocess stop, can resume replication activity<br /><br />show slave status\G<br />-- if any &quot;slave running:no&quot; slave thread has stopped<br />-- below shows query that has stopped slave<br /><br />-- tell mysql to skip 1 (last, bugged) query<br />stop slave;<br />SET GLOBAL SQL_SLAVE_SKIP_COUNTER=1;<br />start slave;<br /><br />-- check slave status againAT-HEhttp://netgame.clnoreply@blogger.com