Fork me on GitHub
#datomic
<
2017-07-28
>
erichmond12:07:37

@robert-stuttaford how do you deal with needing to up the write capacity on dynamo before doing the restore? did you automate that?

robert-stuttaford12:07:23

@erichmond yes, we have a totally scripted process that ups DDB, restores, downs DDB, cycles the transactor, and restarts all the connected peer service apps on their instances

erichmond15:07:23

bash script?

erichmond15:07:33

are you ever in NYC?

erichmond15:07:40

Dinner on me, and we can talk shop, if so

erichmond15:07:44

we have the same stack basically

uwo19:07:34

does the time it takes to connect a peer increase as the size of the db increases? In dev, I’m noticing that some datomic instances are taking longer to connect to than others. (They’re all setup the same way with sql storage).

robert-stuttaford19:07:31

@uwo peers need to grab all the idents and the live index. so if your live index is large (ie just before indexing job) it’ll take longer to start up

robert-stuttaford19:07:50

we saw a substantial peer startup time when we doubled our transactor instance size 🙂

robert-stuttaford19:07:28

@erichmond i live in Cape Town, South Africa, so dinner might be a little tough to swing 😁

erichmond19:07:13

No problem! Just checked google maps. It appears the midway point between NYC and Cape Town is roughly the Cape Verdean Islands. We just each take a quick flight over there.

robert-stuttaford19:07:06

we’ll be the best two Clojure developers for miles!

hmaurer19:07:29

or you could both buy an oculus rift and have a virtual dinner together

hmaurer19:07:34

cheaper and more 2017

robert-stuttaford19:07:46

that won’t be awkward at all

hmaurer19:07:59

just throwing ideas 😄

uwo19:07:32

@robert-stuttaford do you schedule your own indexing jobs outside of imports?