I just re-wrote much of the code to run on the main thread. And so now I
can see the error message (which was about a nil value in a place where I
was certain there would be no nil values). Which is great. But since the
app does a lot of I/O, I would like to for this app to be multi-threaded.
But I need to figure out a way that I can see errors that happen in
background threads.
On Sunday, October 8, 2017 at 11:46:26 PM UTC-4, tbc++ wrote:
>
> I don't think you can catch an Object on the JVM can you? What happens if
> you catch Throwable?
>
> On Sun, Oct 8, 2017 at 9:43 PM, <lawrence...@gmail.com <javascript:>>
> wrote:
>
>> So, for instance, this says that 10 documents were retried:
>>
>> {"message" {:num-slabs 1, :num-active-slabs 1, :enqueued 389, :retried
>> 10, :completed 378, :in-progress 1}}
>>
>> The only place that I call retry! is in this catch block (this function
>> runs in a background thread):
>>
>> (defn advance
>> [message]
>> {:pre [
>> (= (type message) durable_queue.Task)
>> ]}
>> (slingshot/try+
>> (doseq [[k v] @message]
>> (when-not (or
>> (= :how-many-rows-and-fields-from-database k)
>> (= :database-table k)
>> (= :denormalized-id k)
>> (= :topic k))
>> (transform
>> (get @message :denormalized-id :missing-denormalized-id)
>> k
>> v
>> (get @message :topic :missing-topic)
>> (get @message :how-many-rows-and-fields-from-database
>> :missing-how-many-rows-and-fields-from-database))))
>>
>> (durable/complete! message)
>> (catch Object o
>> (durable/retry! message)
>> (slingshot/throw+ {
>> :type ::advance
>> :error o
>> }
>> ))))
>>
>> So clearly, the code has a problem with some of the documents, and so an
>> error is thrown, and it is caught by the catch block. I would love to see
>> the error. I can not. If I put println in the catch block, the message
>> never shows up in the terminal. Likewise if I use Timbre.
>>
>> I would love to see the error object. My inability to see the error
>> objects has crippled my productivity.
>>
>> Can anyone tell me what I need to do to see these errors?
>>
>>
>>
>>
>>
>> On Sunday, October 8, 2017 at 11:13:10 PM UTC-4, lawrence...@gmail.com
>> wrote:
>>
>>> I'm trying to use a Zach Tellman's durable-queue to pass information
>>> from one part of my app to another. I have 3 queues. I also have a function
>>> that runs every 30 seconds, that is suppose to report some statistics for
>>> me:
>>>
>>> (fn []
>>> (future
>>> (slingshot/try+
>>> (errors/log "Resource usage: "
>>>
>>> (monitoring/show-stats-regarding-resources-used-by-this-app))
>>> (errors/log "Stats about from-mysql-to-tables-queue:
>>> " (durable/stats from-mysql-to-tables-queue))
>>> (errors/log "Stats about from-tables-to-topics-queue:
>>> " (durable/stats from-tables-to-topics-queue))
>>> (errors/log "Stats about the
>>> from-topics-to-persistence-queue: " (durable/stats
>>> from-topics-to-persistence-queue))
>>> (catch Object o
>>> (slingshot/throw+ {
>>> :type
>>> ::cycle-resource-usage-report
>>> :error o
>>> }
>>> )))))
>>>
>>> I think there is an error here, because I see the first (errors/log) but
>>> typically not anything after that. Can I rely on the fact that this should
>>> catch anything:
>>>
>>> (catch Object o
>>>
>>> Because it doesn't seem to work. Instead, the app runs for 8 minutes,
>>> then prints data about the queues just once, then never again. Assuming an
>>> error is occuring, why can't I see the error?
>>>
>>>
>>> {:timestamp "17-10-09 01:27:45", :level :trace, :hostname
>>> "ip-172-31-4-54", :message ("STARTING!!!!")}
>>>
>>> Loading class `com.mysql.jdbc.Driver'. This is deprecated. The new
>>> driver class is `com.mysql.cj.jdbc.Driver'. The driver is automatically
>>> registered via the SPI and manual loading of the driver class is generally
>>> unnecessary.
>>>
>>>
>>> {:timestamp "17-10-09 01:28:16", :level :trace, :hostname
>>> "ip-172-31-4-54", :message ("\n\n\n\n\nResource usage: " "Memory in use
>>> (percentage/used/max-heap): (\"67%\" \"2414M\" \"3568M\")\n\nCPU usage
>>> (how-many-cpu's/load-average): [4 1.56]\n\nFree memory in jvm:
>>> [1201758848]")}
>>>
>>>
>>> {:timestamp "17-10-09 01:28:49", :level :trace, :hostname
>>> "ip-172-31-4-54", :message ("\n\n\n\n\nResource usage: " "Memory in use
>>> (percentage/used/max-heap): (\"81%\" \"2900M\" \"3568M\")\n\nCPU usage
>>> (how-many-cpu's/load-average): [4 2.64]\n\nFree memory in jvm:
>>> [699798216]")}
>>>
>>> ;; removing redundant entries
>>>
>>> {:timestamp "17-10-09 01:34:48", :level :trace, :hostname "UnknownHost",
>>> :message ("\n\n\n\n\nResource usage: " "Memory in use
>>> (percentage/used/max-heap): (\"87%\" \"3118M\" \"3568M\")\n\nCPU usage
>>> (how-many-cpu's/load-average): [4 3.77]\n\nFree memory in jvm:
>>> [471249616]")}
>>>
>>>
>>> {:timestamp "17-10-09 01:35:17", :level :trace, :hostname
>>> "ip-172-31-4-54", :message ("\n\n\n\n\nResource usage: " "Memory in use
>>> (percentage/used/max-heap): (\"87%\" \"3120M\" \"3568M\")\n\nCPU usage
>>> (how-many-cpu's/load-average): [4 3.79]\n\nFree memory in jvm:
>>> [468602192]")}
>>>
>>>
>>> {:timestamp "17-10-09 01:35:35", :level :trace, :hostname
>>> "ip-172-31-4-54", :message ("\nStats about from-mysql-to-tables-queue: "
>>> {"message" {:num-slabs 1, :num-active-slabs 1, :enqueued 306671, :retried
>>> 0, :completed 306660, :in-progress 8}})}
>>>
>>>
>>> {:timestamp "17-10-09 01:35:50", :level :trace, :hostname
>>> "ip-172-31-4-54", :message ("\nStats about from-tables-to-topics-queue: "
>>> {})}
>>>
>>>
>>> {:timestamp "17-10-09 01:35:50", :level :trace, :hostname
>>> "ip-172-31-4-54", :message ("\nStats about the
>>> from-topics-to-persistence-queue: " {})}
>>>
>>>
>>> {:timestamp "17-10-09 01:36:15", :level :trace, :hostname
>>> "ip-172-31-4-54", :message ("\n\n\n\n\nResource usage: " "Memory in use
>>> (percentage/used/max-heap): (\"28%\" \"1007M\" \"3568M\")\n\nCPU usage
>>> (how-many-cpu's/load-average): [4 4.14]\n\nFree memory in jvm:
>>> [1868654000]")}
>>>
>>>
>>>
>>>
>>>
>>> So 8 minutes after the app starts, it print the lines with stats about
>>> the queues, just this one time, and then never again. I think an error must
>>> occur inside of the function that is suppose to report the queues. What
>>> would I have to do to see that error?
>>>
>>> My core/main function looks like this:
>>>
>>> (defn -main [& args]
>>>
>>> (slingshot/try+
>>>
>>> ;; 2017-10-04 -- see this:
>>> ;;
>>> https://stackoverflow.com/questions/28908835/ssl-peer-shut-down-incorrectly-in-java
>>> (System/setProperty "https.protocols" "TLSv1.1")
>>>
>>> (.addShutdownHook (Runtime/getRuntime)
>>> (Thread.
>>> #(do (println "addShutdownHook triggered. This app
>>> is shutting down.")
>>> (stop))))
>>>
>>> ;; 2017-07-12 -- stealing code from here:
>>> ;; https://stuartsierra.com/2015/05/27/clojure-uncaught-exceptions
>>> ;; Assuming require [clojure.tools.logging :as log]
>>> (Thread/setDefaultUncaughtExceptionHandler
>>> (reify Thread$UncaughtExceptionHandler
>>> (uncaughtException [_ thread ex]
>>> (println ex "Uncaught background exception on thread " (.getName
>>> thread) " Caught by Thread/setDefaultUncaughtExceptionHandler in
>>> core/main.")
>>> (throw (ex-info "/setDefaultUncaughtExceptionHandler. The app
>>> has died!")))))
>>>
>>> (start)
>>> (catch Exception e
>>> (println e)
>>> (stack/parse-exception e)
>>> (throw (ex-info "The app has died!")))
>>> (catch Object o
>>> (println o)
>>> (throw (ex-info "The app has died!")))))
>>>
>>> What else could I do to either kill the app at the moment of error or
>>> catch it at the top and print the error?
>>>
>> --
>> You received this message because you are subscribed to the Google
>> Groups "Clojure" group.
>> To post to this group, send email to clo...@googlegroups.com
>> <javascript:>
>> Note that posts from new members are moderated - please be patient with
>> your first post.
>> To unsubscribe from this group, send email to
>> clojure+u...@googlegroups.com <javascript:>
>> For more options, visit this group at
>> http://groups.google.com/group/clojure?hl=en
>> ---
>> You received this message because you are subscribed to the Google Groups
>> "Clojure" group.
>> To unsubscribe from this group and stop receiving emails from it, send an
>> email to clojure+u...@googlegroups.com <javascript:>.
>> For more options, visit https://groups.google.com/d/optout.
>>
>
>
>
> --
> “One of the main causes of the fall of the Roman Empire was that–lacking
> zero–they had no way to indicate successful termination of their C
> programs.”
> (Robert Firth)
>
--
You received this message because you are subscribed to the Google
Groups "Clojure" group.
To post to this group, send email to clojure@googlegroups.com
Note that posts from new members are moderated - please be patient with your
first post.
To unsubscribe from this group, send email to
clojure+unsubscr...@googlegroups.com
For more options, visit this group at
http://groups.google.com/group/clojure?hl=en
---
You received this message because you are subscribed to the Google Groups
"Clojure" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to clojure+unsubscr...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
