Szabolcs, thanks for mentioning, this was fixed in the TLA a few commits back. thanks, avati 2007/11/21, Székelyi Szabolcs <cc@xxxxxxxxx>: > > Hi, > > This error may be just cosmetic, but with posix-locks loaded on the > server side, I get ENOTCONN instead of EAGAIN when trying to establish a > conflicting lock. > > According to the trace translator, posix-locks works fine on the server > side, it sets errno to EAGAIN. However, on the client side something > transforms it into ENOTCONN. Here's how I tried it: > > <server> > storage/posix > debug/trace <--- errno is correct (EAGAIN) > transport/server (tcp/server) > <network> > transport/client > debug/trace <--- errno is ENOTCONN > <client> > > Are you aware of this? Is this error significant? (I guess it is, since > EAGAIN is a temporary error instructing the caller to retry the > operation, while ENOTCONN looks like a fatal one.) > > Thanks, > -- > cc > > > > _______________________________________________ > Gluster-devel mailing list > Gluster-devel@xxxxxxxxxx > http://lists.nongnu.org/mailman/listinfo/gluster-devel > -- It always takes longer than you expect, even when you take into account Hofstadter's Law. -- Hofstadter's Law
