Fri Jun 1 14:54:24 PDT 2007
- Previous message: [Slony1-general] Does slonik EXECUTE SCRIPT call waits for comands termination?
- Next message: [Slony1-general] if the slon daemon stop at all (on master and slaves) SL_NODELOCK must be empty ???
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
On 6/1/2007 1:21 PM, Andrew Hammond wrote: > On 6/1/07, Jan Wieck <JanWieck at yahoo.com> wrote: >> On 6/1/2007 10:52 AM, Dmitry Koterov wrote: >> > I also tried to use >> > >> > try { >> > ... >> > } on success { >> > WAIT FOR EVENT ( >> > ORIGIN = ALL, >> > CONFIRMED = ALL, >> > WAIT ON = 1 >> > ) >> > } >> >> Looking at the code, it seems that EXECUTE SCRIPT doesn't actually >> record the events seqno in the admin conninfo. Which means that WAIT FOR >> EVENT will wait for an earlier event, or even nothing. Unfortunately the >> same is true for SYNC in 1.2.9. > > Hmmm, I'm looking at slonik.c under slonik_ddl_script() and I see > (starting at line 3706) > > slon_mkquery(&query, > "select \"_%s\".ddlScript_prepare(%d, %d); ", > stmt->hdr.script->clustername, > stmt->ddl_setid, /* dstring_data(&script), */ > stmt->only_on_node); > > if (db_exec_evcommand((SlonikStmt *) stmt, adminfo1, &query) < 0) > > So, it looks like it's recording the event's seqno in the admin > conninfo. What am I missing? You're missing that it isn't ddlScript_prepare() that generates the event we want to wait for, but ddlScript_complete(). And that one isn't called via db_exec_evcommand(). Jan -- #======================================================================# # It's easier to get forgiveness for being wrong than for being right. # # Let's break this rule - forgive me. # #================================================== JanWieck at Yahoo.com #
- Previous message: [Slony1-general] Does slonik EXECUTE SCRIPT call waits for comands termination?
- Next message: [Slony1-general] if the slon daemon stop at all (on master and slaves) SL_NODELOCK must be empty ???
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
More information about the Slony1-general mailing list