Re: Flink Runtime Exception

Hi Andra, I would try increasing the memory per task manager, i.e. on a machine with 8 CPUs and 16GBs of memory, instead of spawning 8 TMs with 2GB each, I would try to spawn 2 TMs of 8GBs each. This might help with the spilling problem (in case that the CPU is not your bottleneck, this might even

Re: Flink Runtime Exception

On 19 Jun 2015, at 14:53, Andra Lungu wrote: > Another problem that I encountered during the same set of experiments(sorry > if I am asking too many questions, I am eager to get things fixed): > - for the same configuration, a piece of code runs perfectly on 10GB of > input, then for 38GB it runs

Re: Flink Runtime Exception

woops, sorry! Whenever I read the word "deadlock" I getting a bit nervous and distracted ;-) 2015-06-19 15:21 GMT+02:00 Till Rohrmann : > I think Andra wrote that there is *no deadlock*. > > On Fri, Jun 19, 2015 at 3:18 PM Fabian Hueske fhue...@gmail.com > wrote:

Re: Flink Runtime Exception

I think Andra wrote that there is *no deadlock*. On Fri, Jun 19, 2015 at 3:18 PM Fabian Hueske fhue...@gmail.com wrote: Hi Andra, > > The system should never deadlock. > There is a bug somewhere if that happens. > > Can you check if the program is really stuck? >

Re: Flink Runtime Exception

Hi Andra, The system should never deadlock. There is a bug somewhere if that happens. Can you check if the program is really stuck? Cheers, Fabian 2015-06-19 15:08 GMT+02:00 Till Rohrmann : > What does forever mean? Usually it's the case that you see a steep decline > in performance once the s

Re: Flink Runtime Exception

What does forever mean? Usually it's the case that you see a steep decline in performance once the system starts spilling data to disk because of the disk I/O bottleneck. The system always starts spilling to disk when it has no more memory left for its operations. So for example if you want to sor

Re: Flink Runtime Exception

Another problem that I encountered during the same set of experiments(sorry if I am asking too many questions, I am eager to get things fixed): - for the same configuration, a piece of code runs perfectly on 10GB of input, then for 38GB it runs forever (no deadlock). I believe that may occur becau

Re: Flink Runtime Exception

Yes, it was an issue for the milestone release. On Fri, Jun 19, 2015 at 2:18 PM Andra Lungu wrote: > Yes, so I am using flink-0.9.0-milestone-1. Was it a problem for this > version? > I'll just fetch the latest master if this is the case. > > On Fri, Jun 19, 2015 at 2:12 PM, Till Rohrmann > wro

Re: Flink Runtime Exception

Yes, so I am using flink-0.9.0-milestone-1. Was it a problem for this version? I'll just fetch the latest master if this is the case. On Fri, Jun 19, 2015 at 2:12 PM, Till Rohrmann wrote: > Hi Andra, > > the problem seems to be that the deployment of some tasks takes longer than > 100s. From the

Re: Flink Runtime Exception

Hi Andra, the problem seems to be that the deployment of some tasks takes longer than 100s. From the stack trace it looks as if you're not using the latest master. We had problems with previous version where the deployment call waited for the TM to completely download the user code jars. For larg

Flink Runtime Exception

Hi everyone, I ran a job this morning on 30 wally nodes. DOP 224. Worked like a charm. Then, I ran a similar job, on the exact same configuration, on the same input data set. The only difference between them is that the second job computes the degrees per vertex and, for vertices with degree high