I now have some time to do a few tests with Hindsight, so I tried to compile it on our targeted arm platform (raspberry pi), but I get the following error:
root@hl-mc-9999-dev:~/hindsight/release# cmake -DCMAKE_BUILD_TYPE=release .. -- The C compiler identification is GNU 4.7.2 -- The CXX compiler identification is GNU 4.7.2 -- Check for working C compiler: /usr/bin/gcc -- Check for working C compiler: /usr/bin/gcc -- works -- Detecting C compiler ABI info -- Detecting C compiler ABI info - done -- Detecting C compile features -- Detecting C compile features - done -- Check for working CXX compiler: /usr/bin/g++ -- Check for working CXX compiler: /usr/bin/g++ -- works -- Detecting CXX compiler ABI info -- Detecting CXX compiler ABI info - done -- Detecting CXX compile features -- Detecting CXX compile features - done -- Found LUASANDBOX: /usr/local/lib/libluasandbox.so -- Configuring done -- Generating done -- Build files have been written to: /root/hindsight/release root@hl-mc-9999-dev:~/hindsight/release# make Scanning dependencies of target hindsight [ 2%] Building C object src/CMakeFiles/hindsight.dir/hindsight.c.o [ 4%] Building C object src/CMakeFiles/hindsight.dir/ hs_analysis_plugins.c.o [ 6%] Building C object src/CMakeFiles/hindsight.dir/ hs_checkpoint_reader.c.o /root/hindsight/src/hs_checkpoint_reader.c: In function 'find_first_id': /root/hindsight/src/hs_checkpoint_reader.c:46:3: error: large integer implicitly truncated to unsigned type [-Werror=overflow] /root/hindsight/src/hs_checkpoint_reader.c:55:3: error: comparison is always false due to limited range of data type [-Werror=type-limits] cc1: all warnings being treated as errors src/CMakeFiles/hindsight.dir/build.make:100: recipe for target 'src/CMakeFiles/hindsight.dir/hs_checkpoint_reader.c.o' failed make[2]: *** [src/CMakeFiles/hindsight.dir/hs_checkpoint_reader.c.o] Error 1 CMakeFiles/Makefile2:947: recipe for target 'src/CMakeFiles/hindsight.dir/all' failed make[1]: *** [src/CMakeFiles/hindsight.dir/all] Error 2 Makefile:146: recipe for target 'all' failed make: *** [all] Error 2 Do you know what is going on here? I guess this is an issue with the arm platform only? Cheers, Bruno On 10 June 2015 at 18:41, bruno binet <[email protected]> wrote: > Thanks a lot for your answers. > > And yes, I'm very interested in bootstrapping a first prototype of my own > data pipeline based on Hindsight so that I can compare the performance on a > raspberry pi. > (here is the current state of our Heka-based data pipeline: > https://bitbucket.org/helioslite/heka-hl-sandboxes) > So it would be great if you can give me the first instructions on how to > build and setup Hindsight. > > Thanks. > Bruno > > On 10 June 2015 at 18:18, Michael Trinkala <[email protected]> wrote: > >> - It is usable and being actively developed with the intent to move it >> into production later this year. >> - We are currently running production data through it for testing but it >> is not deployed in an official capacity. It has been very stable but until >> a more robust set of tests have been build out I will not consider it >> production ready. >> - Yes, it can decode/encode Heka protobuf format >> - Yes, the router/message matcher is complete. The only difference is >> that it supports Lua string pattern matching instead of re2 regexp (Heka >> 'Hostname =~ /^foo/' vs Hindsight 'Hostname =~ "^foo"') >> - Yes, but you would need a lua-socket input and output sandbox (see >> benchmarks/hsr_run for related examples) >> - No documentation yet, only examples in the benchmarks directory. I >> could have you bootstrapped in about 30 minutes (and hopefully turn that >> into a getting started guide) if you are interested. >> >> Implementation wise the only missing piece is support for dynamically >> loading plugins. The actual code to accomplish it is very small (just >> detecting files in the load directory and moving them to the run directory) >> but ideally it would be fronted by a web server and a GUI with access >> control and validation (a much larger effort and actually a separate >> project). >> >> Trink >> >> On Wed, Jun 10, 2015 at 8:15 AM, bruno binet <[email protected]> >> wrote: >> >>> Hi, >>> >>> I recently discovered the work pushed into the Hindsight repository ( >>> https://github.com/trink/hindsight) which seems to be a lightweight >>> alternative to Heka, based on the lua sandbox. >>> The Hindsight vs Heka benchmarks are quite impressive. >>> >>> I'm currently running Heka on the raspberry pi (not so powerful) device >>> and the load average quickly increases and exceeds 1 when Heka is ingesting >>> data, so Hindsight could be a good fit for us if it can perform better than >>> Heka in terms of CPU cycles. >>> >>> What is the current status of Hindsight? Is it just an temporary >>> experiment or will it be maintained and actually used in production? >>> Is it currently usable and stable? >>> Is Hindsight able to decode and encode Heka protobuf format? >>> Does Hindsight have a complete router implementation to dispatch >>> messages to sandboxes like in Heka? >>> My use case is basically to read raw text data from UDP socket, parse >>> text data with lua patterns or lpeg, process data through a few lua sandbox >>> filters, then write output messages both to a file (protobuf heka format) >>> and a HTTP server (json format): can this be easily accomplished with >>> Hindsight? >>> Is there any documentation somewhere to get started with Hindsight? >>> >>> Thanks, >>> Bruno >>> >>> _______________________________________________ >>> Heka mailing list >>> [email protected] >>> https://mail.mozilla.org/listinfo/heka >>> >>> >> >
_______________________________________________ Heka mailing list [email protected] https://mail.mozilla.org/listinfo/heka
