> > > On irc somebody kindly guided me through getfattr (that won't show you > anything if you don't ask for it with -d -m "*.gluster.*") and eventually > managed to play more with the setups. > > Probably could be helpful putting those on the faq. >
Sure, I will add these entries in FAQ. > > Meanwhile I eventually solved my issue with writebehind translator by > updating to git sources (the 2.0.0rc1 is apparently buggy in that regard) We just made RC2 tarball, which has this fix and few other minor fixes after RC1. > and I'm trying to get writes pushing more data (so far the best I got is > 75Mb/s from 5 concurrent dd on a single node and 45Mb/s on a single > transfer), right now I'm using a dht+afr mix clientside and enabling afr > apparently hits severely the performance (almost halving them). > Is it 75MBps ? or Mbps? and what is the interconnect?? Regards, Amar
_______________________________________________ Gluster-users mailing list [email protected] http://zresearch.com/cgi-bin/mailman/listinfo/gluster-users
