Clojurians Log v2

For search indexability: "Update failed"

Is there an associated stacktrace you can share?

hi, there is no stacktrace around these messages

for context, we switched from AWS + Aurora postgres to Azure + postgres this weekend past

we are seeing transaction timeouts which correlate exactly with these transactor logs. our peer tx timeout is 20 seconds

we've assessed several of these transactions and they are very limited in scope in terms of applicable datoms, in the 10s to 100s at most

Hm. Any transaction functions involved?

I don’t believe there are

oh, there are cases with transaction functions and without

the peers throw a generic timeout exception:

clojure.lang.ExceptionInfo: :db.error/transaction-timeout Transaction timed out.
    at datomic.error$raise.invokeStatic(error.clj:70)
    at datomic.error$raise.invoke(error.clj:58)
    at datomic.error$raise.invokeStatic(error.clj:68)
    at datomic.error$raise.invoke(error.clj:58)
    at datomic.peer$await_tx_result.invokeStatic(peer.clj:79)
    at datomic.peer$await_tx_result.invoke(peer.clj:70)
    at datomic.peer.Connection.transact(peer.clj:315)
    at datomic.api$transact.invokeStatic(api.clj:107)
    at datomic.api$transact.doInvoke(api.clj:105)
    at clojure.lang.RestFn.invoke(RestFn.java:467)

Unfortunately I have to run right now. It's worth taking a close look at those transaction functions. Happy to hop on a short call at 15:00 (UTC+1)

Thanks I will be around for the call :)

2025-11-27T12:52:14+00:00  INFO datomic.update {:message "Update failed", :pid 1636, :tid 137}

same :pid and :tid I wonder what :tid stands for, perhaps thread?

Please send logs to datomic support

Along with transactor properties file and any details about how you are configuring your pg driver and the azure pg instance

thank you, sent an email