Is it possible to save the queries created in the Datomic Console into the browser's Local Storage (on a per-db/uri basis)? I searched the docs and this ML but found no mentions of it.

Not at present.

@marshall late to respond but thanks for your solution to sample-query last week. worked perfectly

I got this error when I try to connect to remote transactor:

HornetQNotConnectedException HQ119007: Cannot connect to server(s). Tried with all available servers.  org.hornetq.core.client.impl.ServerLocatorImpl.createSessionFactory (ServerLocatorImpl.java:906)

I can use telnet connect those 3 ports — I opened(20177,20178,20179)

I use Dotomic in dev mode

isaac: using datomic free?

@bkamphaus: Are you in CO?

We are doing sweet machine-learning in Boulder on top of Datomic (❤️) and would love to chat.

http://thinktopic.com

@harold: yep, I’m in Longmont. Went to one of the Clojure meetups that Jeff and Jon organized. Happy to chat with you guys about how you’re using Datomic.

Radical! We'll be doing more meetups for sure.

In my peer, I have a ObjectCache hit ratio of 1 and no reported StorageGetBytes. But I see access to my Riak storage. Is that normal?

I would expect some access — peers read active transactor heartbeat from storage, also the transactor notifies peers of newly written data that needs to be read into memory index on peer.

But the access correlates with my load test. I have nearly no storage access during no load.

@bkamphaus: Sorry it was another process accessing the Riak cluster which is triggered by the same load test. I have no storage access from Datomic itself.

ah, ok, makes sense.

yes my oAuth tokens are also stored there...

@harold are you using any of the open source ML tools, or your own stack?

I got a ProvisionedThroughputExceededException while restoring a database to an underprovisioned DynamoDB store. Will the transactor and peers back off and retry instead of dying if DynamoDB's read or write capacity is exceeded?

to some extent, Datomic will respond to throttling (and metrics like StoragePutBackoffMsec or StorageGetBackoffMsec will be reported), but it is possible that it throttling is too severe that a transactor will fail to heartbeat and fall over, for example.

that makes sense.

Are the transactor and peers resilient to memcached dying as long as the hostname of the memcached server resolves when the transactor or peer starts?