update README.md about dynomite

[timiblossom]

Thanks - we will rework on the documents for sure.

[vpopescu]

As another brand-spanking new user, I'd also request you bring the command line options list up to date. For example, there is a "-g" option to enable gossip which I don't understand since I thought gossip was fundamental to dyanamo.

And if I turn it, on, bad things happen after about a few minutes of running. But I'm not sure I'm understanding what -g should do and how it's affected by my config file.

[Thu May 21 15:37:52 2015] dyn_util.c:299 free(0x7f57b8115790) @ dyn_dict.c:614
[Thu May 21 15:37:52 2015] dyn_util.c:253 malloc(48) at 0x7f57b8115790 @ dyn_dict.c:552
[Thu May 21 15:37:52 2015] dyn_util.c:320 [0] /lib/x86_64-linux-gnu/libpthread.so.0(+0x10d10) [0x7f57be083d10]
[Thu May 21 15:37:52 2015] dyn_util.c:320 [1] dynomite(dictFingerprint+0x4) [0x42a364]
[Thu May 21 15:37:52 2015] dyn_util.c:320 [2] dynomite(dictNext+0xd8) [0x42a588]
[Thu May 21 15:37:52 2015] dyn_util.c:320 [3] dynomite() [0x428cd4]
[Thu May 21 15:37:52 2015] dyn_util.c:320 [4] /lib/x86_64-linux-gnu/libpthread.so.0(+0x76aa) [0x7f57be07a6aa]
[Thu May 21 15:37:52 2015] dyn_util.c:320 [5] /lib/x86_64-linux-gnu/libc.so.6(clone+0x6d) [0x7f57bdaa7eed]
[Thu May 21 15:37:52 2015] dyn_signal.c:127 signal 11 (SIGSEGV) received, core dumping

...

Program terminated with signal SIGSEGV, Segmentation fault.
#0 dictFingerprint (d=d@entry=0x1000074) at dyn_dict.c:522
522 integers[0] = (long) d->ht[0].table;
(gdb) bt
#0 dictFingerprint (d=d@entry=0x1000074) at dyn_dict.c:522
#1 0x000000000042a588 in dictNext (iter=0x7f96d4115790) at dyn_dict.c:579
#2 0x0000000000428cd4 in gossip_forward_state (sp=) at dyn_gossip.c:209
#3 gossip_loop (arg=0x1000074) at dyn_gossip.c:760
#4 0x00007f96dbadd6aa in start_thread (arg=0x7f96da9fe700) at pthread_create.c:333
#5 0x00007f96db50aeed in clone () at ../sysdeps/unix/sysv/linux/x86_64/clone.S:109

[timiblossom]

@vpopescu we will fix this asap. Please run it with disable gossip and use the local service for now. Those 2 are kind of interchangeable.

[ipapapa]

@akbarahmed @shailesh33 since we are now doing the work to update YAML #368 this might be related. Some folks seem to be confused about the terminologies. Any recommendation on how to better organize the documentation? There is some good advice from @redboy1972.

[akbarahmed]

@ipapapa @shailesh33 I agree. Are you thinking about changing just the yaml variables, just the documentation, or both?

[ipapapa]

@akbarahmed there two parallel efforts. One by you #368 and one by me #370 where we are changing the YAML and the documentation. I think as long as the basic functionalities are backward compatible we should update both... I was curious though on what would be the best plan?

[akbarahmed]

@ipapapa I've posted some thoughts on the yaml options at #368.

I'm okay with deferring to you on the yaml updates since you're already working on it.

[ipapapa]

It seems we have a good discussion in #368 about renaming the YAM entries. The gossip enabler argument has also moved to the YAML #386 as well as other entries #387. Hence I think we have addressed this issue. If there are any further questions, we can reopen it.