redis-lock looks interesting. One line stood out though: "Since it's asynchronous, different functions could be holding different locks simultaneously. This is awesome!"
This is both good and bad. The good & bad part is that it's really easy to deadlock without knowing it since multiple functions may be circularly locked on a set of locks and other functions might be proceeding normally. Really hard to debug v/s a system that *actually* does get deadlocked. On Monday, June 18, 2012 8:39:53 AM UTC-7, Rakesh Pai wrote: > > > My favoriate configuration is to use clustered workers with a redis > backend > > (redis and hiredis are also awesome modules) to share objects between > > workers (like sessions or other saved variables). > > I can smell the concurrency problems with that. I had a similar issue > recently (not with cluster), and built this: > https://github.com/errorception/redis-lock > > Feedback welcome, if you have any thoughts. Will be a great fit for > someone who already has redis in their stack. > > Regards, > Rakesh Pai > > On Mon, Jun 18, 2012 at 8:20 PM, Evan <[email protected]> wrote: > > Agree! I've been using it with actionHero, and it makes scaling out > quite > > simple. > > My favoriate configuration is to use clustered workers with a redis > backend > > (redis and hiredis are also awesome modules) to share objects between > > workers (like sessions or other saved variables). > > > > On Saturday, June 16, 2012 11:06:26 AM UTC-7, carter-thaxton wrote: > >> > >> Just a friendly note to say the new cluster API is truly amazing. It > just > >> works! > >> It gave me ~6x speedup on one of my projects, with about 5 minutes of > >> development time. > >> > >> I've written similar integrations for TCP servers using fork() in C++ > >> before, and it's indeed a lot of work and quite subtle to get > everything > >> just right: > >> - Efficient piping of TCP streams to/from the child processes > >> - Graceful and automatic shutdown of the child processes > >> - Shared STDIO for straightforward console logging > >> - Signal handlers > >> - And lots more... > >> > >> Way to go!! > >> > > -- > > Job Board: http://jobs.nodejs.org/ > > Posting guidelines: > > https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines > > You received this message because you are subscribed to the Google > > Groups "nodejs" group. > > To post to this group, send email to [email protected] > > To unsubscribe from this group, send email to > > [email protected] > > For more options, visit this group at > > http://groups.google.com/group/nodejs?hl=en?hl=en > > > > -- > Rakesh Pai > Mumbai, India. > http://rakeshpai.me/ > http://twitter.com/rakesh314 > http://www.google.com/profiles/rakeshpai > -- Job Board: http://jobs.nodejs.org/ Posting guidelines: https://github.com/joyent/node/wiki/Mailing-List-Posting-Guidelines You received this message because you are subscribed to the Google Groups "nodejs" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/nodejs?hl=en?hl=en
