API performance this morning

gr8 topic same 4 me

I was supposed to be showing a mate the wonders of Betfair today!

I guess I'd better leave it.

Once again a spectacular own goal by a company who should know better...

we are aware of the issue you are describing which some API users are experiencing. We hope to have this issue resolved shortly

Yeah I noticed that this morning and last night as well.

The response time looks normal but when you place a bet it seems slow and sticky.

All bet matching seems slow to me, not just API

I can't make any bets now.

Get an unknown error.

just went on to the main site to try and place a few - it says 'sorry site is being upgraded - try again in a few minutes'

this was posted earlier. This was a planned pause

We are rolling out the final part of the solution to yesterday's problems that means we need to suspend bet taking for a few minutes - so it will be paused from 11.15-11.20 GMT. After this, trading will resume as usual.

Thank you for your patience.

where was that posted?

http://community.betfair.com/service/go/thread/view/94166/27236093/bet-placement-1115---1120-gmt

Posted in Service section 1 minute before the outage. Do you call that adequate communication?

this was posted in service. We also updated banners on the sports page / homepage. I appreciate it was late notice, which isnt ideal. But this pause will hopefully mean is site is stable, and bet placement times are back down to normal response times.

Looks like it's worked. The response times are much better than they were earlier.

Dont hold your breath they have been saying for last month that the problems have been resolved and fixed

Bet placing times seems OK now.

I can confirm, since this mornings fix, bet placement times are back to normal.

Please can you confirm that that type of outage won't happen again?

They confirm that most days marky.

Here is todays confirmation that everything is fine

We've fixed this problem now.  We've disabled the original automated job and rebuilt it to update content safely.  We've tripled the capacity of our web server farm to spread our load even more thinly. We've fixed our process for disabling features so that we won't make things worse. We've updated our operational processes and introduced a whole new raft of monitoring to spot this type of issue.  We've also isolated the underlying web server issue so that we can change our content at will without triggering the switch to single-threading.

is there a voluteer to keep a log of all the issues and conclusions going forward so we can refer back to them if needed?