Re: [db-wg] Anything wrong with RIPE DB?

[Shane Kerr <]

Marcus Ruchti,

Marcus.Ruchti@localhost wrote:

> Do you know if anything happened with the RIPE DB last evening
> or even this morning?
>
> Our RIPE robot, reading out AS-sets and route objects, deleted a lot
> of prefixes. Started just a few minutes
> ago again without any changes to the Robot we got many prefixes back.
> This was of course service affecting for single-homed PI customers :-(
>
> Now I wonder if RIPE is doing something with the DB. I've seen the
> output has been changed (reduced mainly) but this does not affect my
> script. So any hints from your side are appreciated.
>
This is probably related to the update server being very busy right now.

We have paused the processing of large updates (more than 500 objects)
right now, to allow other updates to go through. We will resume them
when all of the other updates have gone through. Delays were about 90
minutes.

We have also gotten our query servers working properly again, so updates
are now immediately visible.

Apologies again for any problems that this has caused you.

--
Shane Kerr
Software Engineering Department Manager
RIPE NCC